Description
The purpose of this paper is to explore the impact on communication of changes in an
accounting standard arising from the transition to International Financial Reporting Standards.
It investigates inter and intragroup differences in measured connotative meaning of the old and new
definitions of “cash”, as held by three key groups of parties to the accounting communication process
(preparers, auditors and users); and determines the effect of changes in connotative meaning on
decision behaviour (outcomes).
Accounting Research Journal
The meaning of cash in the context of alternative accounting standards: IFRS
convergence and classification decisions
Tony Mortensen Richard Fisher
Article information:
To cite this document:
Tony Mortensen Richard Fisher, (2011),"The meaning of cash in the context of alternative accounting
standards", Accounting Research J ournal, Vol. 24 Iss 1 pp. 23 - 49
Permanent link to this document:
http://dx.doi.org/10.1108/10309611111148760
Downloaded on: 24 January 2016, At: 21:12 (PT)
References: this document contains references to 71 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 1363 times since 2011*
Users who downloaded this article also downloaded:
Salma Ibrahim, Li Xu, Genese Rogers, (2011),"Real and accrual-based earnings management and its legal
consequences: Evidence from seasoned equity offerings", Accounting Research J ournal, Vol. 24 Iss 1 pp.
50-78 http://dx.doi.org/10.1108/10309611111148779
Afshad J . Irani, Le (Emily) Xu, (2011),"Do stealth restatements convey material information?", Accounting
Research J ournal, Vol. 24 Iss 1 pp. 5-22 http://dx.doi.org/10.1108/10309611111148751
Nicholas Apergis, Christina Christou, Christis Hassapis, (2014),"Accounting standards convergence
dynamics: International evidence from club convergence and clustering", Accounting Research J ournal,
Vol. 27 Iss 3 pp. 226-248 http://dx.doi.org/10.1108/ARJ -06-2013-0031
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The meaning of cashinthe context
of alternative accounting
standards
IFRS convergence and classi?cation decisions
Tony Mortensen and Richard Fisher
University of Canterbury, Christchurch, New Zealand
Abstract
Purpose – The purpose of this paper is to explore the impact on communication of changes in an
accounting standard arising from the transition to International Financial Reporting Standards.
It investigates inter and intragroup differences in measured connotative meaning of the old and new
de?nitions of “cash”, as held by three key groups of parties to the accounting communication process
(preparers, auditors and users); and determines the effect of changes in connotative meaning on
decision behaviour (outcomes).
Design/methodology/approach – The study adopted a between-participants 2 £ 3 factorial design
whereby the ?rst factor re?ected the de?nition type: old vs new de?nition of the concept “cash”; while
the second re?ected three ?nancial reporting groups: preparers, auditors and users. The semantic
differential technique developed by Osgood, Suci and Tannenbaum was used to measure connotative
meaning.
Findings – The study ?nds that the three ?nancial reporting groups do not share the same meaning
of the concept “cash” and that the introduction of the new de?nition has changed the interpreted
connotative meaning for these three groups. A link between measured meaning and the decisions
made by the participants was also established.
Research limitations/implications – The explanatory power of the typical three (evaluative,
potency and activity) factor structure should be acknowledged; these factors typically explain
50 per cent of the total phenomena known as “meaning”. The study’s ?ndings make an important
contribution to the earnings management and creative accounting literature.
Practical implications – The ?ndings are particularly relevant to standard-setters and regulators
as a lack of shared meaning may lead to unnecessary misunderstandings and tensions among the
many parties to the reporting process.
Originality/value – The study extends prior measurement of meaning studies in accounting
through ?rst, the inclusion of all three major groups of parties to the accounting communication
process; second, examination of an accounting concept which is de?ned differently by two accounting
standards in the same jurisdiction; and last, investigation of the impact on decision behaviour
(outcomes) resulting from changes in meaning brought about through the introduction of a new
standard across the three groups.
Keywords International accounting standards, Cash ?ow statement, Connotative meaning,
Measurement of meaning, Semantic differential, Cash, Financial reporting
Paper type Research paper
1. Introduction
Effective communication between the various parties to the reporting process is the
cornerstone of ?nancial accounting (Shah, 1996). The importance of shared meaning,
and, in particular, shared connotative meaning[1], between preparers and the many
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
The meaning
of cash
23
Accounting Research Journal
Vol. 24 No. 1, 2011
pp. 23-49
qEmerald Group Publishing Limited
1030-9616
DOI 10.1108/10309611111148760
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
users of ?nancial reports in facilitating effective communication has been widely
acknowledged in the literature (Haried, 1972; Oliver, 1974; Flamholtz and Cook, 1978;
Adelberg and Farrelly, 1989; Bagranoff et al., 1994; Hronsky and Houghton, 2001; Wines,
2006). The provision of clear de?nitions by standard-setters contributes to the
communications process by providing a uniform basis from which all parties can
operate when dealing with the meaning associated with fundamental accounting terms
and concepts. This suggests that careful wording may signi?cantly reduce the potential
for miscommunication and “[. . .] reduce the justi?ability of aggressive reporting
decisions” (Hronsky and Houghton, 2001, p. 124), thus enhancing the reliability of
?nancial statements.
The global convergence to International Financial Reporting Standards (IFRS) has
seen extensive and unprecedented changes in many jurisdictions’ Generally Accepted
Accounting Practice (GAAP)[2], and poses a substantial challenge to the ?nancial
communications process as preparers, auditors and users confront unfamiliar
terminology and alternative de?nitions of existing concepts. This study explores the
complexity of this communication problemin the context of NewZealand’s transition to
IFRS. The study examines the connotative meaning of the concept “cash”, as it relates to
the cash ?ow statement, as interpreted by three key groups of parties to the accounting
communication process: preparers, auditors and users. Using an experimental design,
the study focuses on inter and intragroup changes in measured connotative meaning
resulting from the introduction of a new de?nition of cash, contained in New Zealand
Equivalent to International Accounting Standard (NZ IAS) 7, “cash ?ow statements”.
New Zealand’s old de?nition is found in FRS 10, “statement of cash ?ows” and has
remained in effect after the mandatory adoption date for NZ IFRS for many entities such
as small to medium-sized non-listed companies and not-for-pro?t entities. Such entities
are currently not subject to the full ?nancial reporting requirements of New Zealand’s
Financial Reporting Act 1993 (pending the outcome of a reviewof the ?nancial reporting
regime in New Zealand currently being undertaken by the Ministry of Economic
Development). The study also considers the effect that possible changes in meaning may
have on the decisions made by the different parties[3].
Many would regard cash as being among the most clear and least contentious
concepts in the domain of accounting (Mulford and Comiskey, 2005; Tergesen, 2002).
Despite this, the cash ?ow statement has recently been criticised for its susceptibility to
aggressive reporting practices (Mulford and Comiskey, 2002; Solomon, 2002; Tergesen,
2002; Broome, 2004). Clearly worded accounting standards are one means of reducing
the justi?ability of aggressive reporting decisions (Hronsky and Houghton, 2001).
However, in the New Zealand context, the respective de?nitions of cash contained in
FRS 10 and NZ IAS 7 use ambiguous and subjective terms, such as “highly liquid”,
“short term”, “at call”, “day-to-day cash management” and “insigni?cant risk”.
Somewhat ironically, the cash ?owstatement replaced the funds statement in the late
1980s, due in part to concerns about the lack of de?nitional clarity surrounding the term
“funds” and its consequential susceptibility to misunderstanding and potential
manipulation (Goldberg, 1951; Buzby and Falk, 1974; Henry, 1975; Heath, 1978; Clift,
1979; Chesley and Scheiner, 1982; Graci, 1982; Vicknair, 1983; Lawson, 1981; Lee, 1984;
Vent et al., 1995). If it is found in this study that FRS have failed to achieve a shared
understanding of a relatively straightforward concept, such as cash, then substantial
doubt must exist over standard-setters’ abilities to achieve shared meanings on more
ARJ
24,1
24
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
complex ?nancial terms and concepts among the key players in the accounting
communication process.
The study builds on the measurement of meaning in accounting literature that has
traditionally focused on the extent to which the meaning of various accounting concepts
are shared by different parties to the accounting communication process (e.g. among
preparers, users, or auditors). The study extends research in this area through ?rst, the
inclusion of all three major groups of parties to the accounting communication process;
second, examination of an accounting concept which is de?ned differently by two
accounting standards in the same jurisdiction; and last, investigation of the impact of
resulting changes in meaning brought about through the introduction of a newstandard
on decision behaviour (outcomes) across the three groups.
With New Zealand being an early adopter of IFRS[4], the results of this study have
far-reaching implications for standard-setters and regulators of countries adopting
IFRS. Further, the results provide an important contribution to the creative accounting
and earnings management literature, as the de?nitional clarity of accounting concepts is
believed to be an important component in reducing creative or aggressive reporting
practices.
With IFRS convergence, a subtle change in the de?nition of even a seemingly
straightforward item such as cash (for purposes of the cash ?ow statement) can result
in different interpretations (connotative meaning) for different parties to the ?nancial
reporting communication process, and, in turn, impact on accounting decisions. Future
research may build on the ?ndings of this study by investigating the issues raised in
the study in relation to other IFRS standards.
The next section provides a review of the literature. This is followed by the research
hypotheses and an outline of the research method employed in the study. The results
are then considered, followed by a discussion of the research ?ndings and conclusion.
2. Review of the literature
2.1 Cash and the cash ?ow statement
Since the introduction of the cash ?ow statement, its usefulness to decision makers has
received signi?cant attention (Epstein, 1992; Jones et al., 1995, 1998; Yap, 1996; Jones
and Ratnatunga, 1997; Sharma and Iselin, 2003). The literature highlights several
advantages of cash ?ow information over traditional information found in the balance
sheet and income statement (Neubert, 1959; Mason, 1961; Hicks and Hunt, 1980; Lee,
1984, 1992; Sharma and Iselin, 2003), with one of the most frequently cited being
increased reliability over traditional accrual-based accounts. Reasons for this
advantage include the establishment of objective criteria and de?nitional clarity.
The precursor to the cash ?ow statement was the funds statement, which was ?rst
seen in the USA in the early 1900s (Wilson, 1989). While originally welcomed by many
countries, it became apparent by the late 1970s that the funds statement was not going to
be a permanent addition to the annual report (or at least not in its existing form). One of
the key concerns echoed by professionals and academics alike surrounded the de?nition
of the term “fund” and its inconsistent application in practice (Goldberg, 1951; Buzby
and Falk, 1974; Henry, 1975; Heath, 1978; Clift, 1979; Lawson, 1981; Chesley and
Scheiner, 1982; Graci, 1982; Vicknair, 1983; Lee, 1984; Vent et al., 1995). Vent et al. (1995)
believed that the meaning of the word “fund” had never been clearly de?ned, and that
this lack of guidance had resulted in a wide diversity of practice in the preparation
The meaning
of cash
25
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of the statement. After many years of debate, national standard-setters began to
systematically replace the funds statement with the cash ?owstatement in the hope that
many of the issues raised regarding the problematic nature of the funds statement would
be resolved[5]. The concept of “fund” was effectively replaced by that of cash, which was
believed to be clearer and less ambiguous (Buzby and Falk, 1974).
The cash ?ow statement purports to present the inward and outward ?ows of cash.
Accordingly, a clear understanding of what constitutes cash is important within and
across all parties to the accounting communication process. New Zealand’s adoption of
IFRS saw it introduce NZ IAS 7. This standard (along with all other NZ IFRS) applies
to reporting entities, which include large issuers, subsidiaries of overseas companies
complying with IFRSs, and the public sector. Certain smaller entities are currently still
permitted to apply the old standard, FRS 10, which de?nes cash differently than
NZ IAS 7, raising the possibility that different parties to the communications process
may hold different meanings of the term.
2.2 De?nition of cash
Table I presents the de?nition of cash (as it relates to cash ?ow) under FRS 10 and the
more recent NZ IAS 7.
While the concept “cash” under FRS 10 includes currencies, as would be described in a
traditional sense (i.e. coins, notes and demand deposits), it also introduces the concept of
“highlyliquidinvestments” (Para. 4.1). However, this termis not separatelyde?nedinFRS
10. FRS 10’s de?nition also speci?callyencompasses certainforms of “at call borrowings”,
which effectively represent “negative” cash. In contrast, NZ IAS 7’s de?nition includes
both “cash on hand” and “demand deposits”, together with all other “cash equivalents”.
The latter is de?ned in terms of ease of conversion to known amounts of cash and
presenting an insigni?cant risk of changes in value. Whilst there is signi?cant overlap
between the two de?nitions, there are also clearly notable differences.
A further notable observation concerning the two de?nitions in Table I is the degree
to which many of the terms employed are inherently ambiguous, leading to potential
subjectivity in the interpretation and application of the de?nitions. For instance, phrases
such as “short term”, “highly liquid”, “at call”, “readily convertible”, “day-to-day cash
management” and “insigni?cant” are all open to alternative interpretation and little
guidance is provided in either standard.
To many users of ?nancial statements, cash ?ow is regarded as “real and not
subject to the vagaries of GAAP or the whims of the accountants” (xiii), while cash,
which underpins cash ?ow, is considered fact (Mulford and Comiskey, 2005).
FRS 10 – de?nition of “cash”
NZ IAS 7 – de?nition of “cash” and “cash
equivalents”
“[. . .] coins, notes, demand deposits and other
highly liquid investments in which an entity
invests as part of its day-to-day cash
management.” And “[. . .] borrowings from
?nancial institutions such as bank overdrafts,
where such borrowings are at call and are used as
part of day-to-day cash management” (Para. 4.1)
“[. . .] cash on hand and demand deposits” (Para. 6)
“cash equivalents” is de?ned as:
“[. . .] short term, highly liquid investments that
are readily convertible to known amounts of cash
and which are subject to an insigni?cant risk of
change in value” (Para. 6)
Table I.
De?nitions of cash
ARJ
24,1
26
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
However, as noted above, with the introduction of NZ IAS 7, there now exist two
de?nitions of cash in New Zealand, neither of which exactly overlap and both requiring
judgment in their application. The introduction of NZ IAS 7 has created a need to
explore the meanings assigned to the concept of cash within and between the differing
groups of parties to the ?nancial reporting communications process. Prior research
suggests that differences in meanings attributed to concepts de?ned in accounting
standards may increase the possibility for miscommunication, and have ?ow on
implications for decision making, including increasing the risk of aggressive reporting
practices (Hronsky and Houghton, 2001). Exploration of this issue is also consistent
with Mason and Gibbins’ (1991) call for researchers to examine new and existing
standards for possible ambiguities that may detract from the overall thrust of
particular accounting standards.
2.3 Measurement of meaning
The connotative meaning of cash in this study is measured using a well-established
method of measuring meaning, the semantic differential technique. The technique was
developed by Osgood and Suci (1955), and later re?ned by Osgood et al. (1957).
Osgood et al. narrowly de?ned meaning as that which explains the way in which
differences occur between the sender and receiver of information. While the
quanti?cation of meaning will always be controversial, Osgood et al. argued that
connotative meaning is responsible for in?uencing “individual behaviour” (p. 321).
The semantic differential technique is essentially a combination of controlled
association and scaling procedures which are used to identify and measure
a participant’s connotative meaning of a concept. Coupled with the use of factor
analysis, the process allows for the identi?cation of a participant’s connotative meaning
of a speci?c term or concept by locating their perceived meaning as a point in a
corresponding “semantic space” (Osgood et al., 1957, p. 25)[6].
Osgood et al. (1957) found that, for most everyday concepts in the general domain of
meaning, connotative meaning could be explained within a three factor “cognitive”
structure. These factors were identi?ed in their study and labelled as: evaluative,
represented by scale items such as good-bad, important-unimportant; potency,
represented by scale items such as hard-soft, strong-weak; and activity, made up of scale
items such as active-passive, fast-slow. The evaluative, potency and activity (EPA)
structure has been validated in a large number of subsequent studies across a wide
range of disciplines[7]. As these factors represent the axes within the semantic space
where meaning is believed to exist, a participant’s meaning of a speci?c termor concept
can be identi?ed within that semantic space.
2.4 Measurement of meaning in accounting
Haried (1972) conducted the earliest research in the area of measured meaning in
accounting. He undertook a study into what he called the “[. . .] semantic problems in
accounting communication [. . .]” (p. 376), looking at the proximity (or differences)
in meaning between that intended by the sender (e.g. the meaning of the term
“depreciation” as determined by the accountant) and that interpreted by the receivers
(e.g. the meaning of the term “depreciation” as determined by a banker or investor).
Attempting to validate Osgood et al.’s (1957) technique in the accounting domain,
Haried (1973) used a sample of concepts commonly found in ?nancial statements
The meaning
of cash
27
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
(e.g. goodwill, net income, accounts receivable) across both sophisticated
(e.g. accountants) and unsophisticated (e.g. investment club members) research
participants. This study revealed a seven, factor structure. However, subsequent
reanalysis of Haried’s (1973) data by Houghton (1988, p. 279) con?rmed that the “[. . .]
dimensions of space are similar to those ‘standard’ EPA factors that were originally
identi?ed by Osgood et al. (1957)”. Houghton’s (1988) reanalysis laid the foundations for
further analysis in the domain of accounting meaning, leading to a number of studies
examining a variety of terms, concepts and phrases used in accounting (Oliver, 1974;
Flamholtz and Cook, 1978; Houghton, 1987a, 1988; Houghton and Messier, 1990;
Hronsky and Houghton, 2001; Wines, 2006), using different groups of participants
(Houghton, 1987b; Houghton and Messier, 1990; Houghton and Hronsky, 1993), and
across different cultures (Bagranoff et al., 1994).
It has also been established that the level of sophistication of participants tends to
affect the degree of shared cognitive structure in which meaning is believed to be held
(Oliver, 1974; Houghton, 1987a, 1988; Houghton and Messier, 1990; Houghton and
Hronsky, 1993; Hronsky and Houghton, 2001; Wines, 2006). This is important as the
absence of a shared cognitive structure prevents further analysis of measured meaning.
That is, a shared structure is a necessary but not suf?cient condition to con?rm the
existence of shared connotative meaning.
Recent studies by Hronsky and Houghton (2001) and Wines (2006) have extended the
boundaries of measurement of meaning research by investigating the association
between measured meaning and decision outcomes. Hronsky and Houghton (2001)
found that practicing auditors held the same cognitive structure within which the
meaning of “extraordinary items” was held, and that their classi?cation decisions were
systematically associated with differences in the connotative meanings of de?nitions of
“extraordinary items” under new and old FRS. Wines (2006), on the other hand,
examined the concept of “auditor’s independence” using a sample of 89 undergraduate
students. He found that participants considered the concept within a two factor structure
comprising “emphasis” and “variability” dimensions, and that their perceptions of
independence in three case scenarios were systematically associated with the identi?ed
connotative meaning dimensions.
The current study extends the measurement of meaning literature in several ways.
First, in contrast to prior studies, it investigates the connotative meaning of a concept
where meaning is widely regarded as being well understood and relatively
uncontroversial (Mulford and Comiskey, 2005; Tergesen, 2002). Second, it considers
the degree of shared meaning between three important groups of parties to the ?nancial
communications process: preparers, users and auditors. Last, it investigates the
relationships between the decision behaviour (outcomes) of all three groups and their
respective connotations of the concept of cash.
3. Research question and hypotheses
The overall objectives of this study are to determine, ?rst, whether the three main
groups of parties to the accounting communication process share a similar meaning of
the concept of cash, as it relates to the cash ?ow statement; second, whether a change in
the regulated de?nition of cash can in?uence measured meaning and decision
outcomes; and third, whether there is empirical evidence to link the measured meaning
ARJ
24,1
28
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of this concept with the decisions made by each party. Five main hypotheses provide
the basis for addressing these research objectives.
3.1 Hypotheses
Studies in the area of measurement of (connotative) meaning in accounting have
con?rmed the importance of establishing the degree of shared cognitive structure
between participants in order to subsequently determine and measure the meaning
associated witha given concept (Houghton, 1987a, 1988). Anumber of these studies have
provided evidence to suggest that preparers, auditors and speci?c users (e.g. bankers
and shareholders) do not share the same cognitive structure in which the meaning of
certain accounting concepts are held, due in part to differences in sophistication,
incentives and degree of homogeneity (Houghton, 1987b; Hronsky and Houghton, 2001).
Therefore, the following null hypothesis is proposed for the current study:
H
01
. There is no signi?cant difference in the cognitive (factor) structures within
which the meaning of “cash” is held between each of the three ?nancial
reporting groups.
As previously discussed, shared cognitive structure is a necessary but not suf?cient
condition for establishing shared connotative meaning (Houghton, 1988). As clari?ed by
Houghton and Messier (1990), there are two component parts to shared meaning in a
communications process. First, parties must share the same “cognitive structure”
(e.g. perhaps all sharing a three-dimensional factor structure conforming to the classic
EPA structure), and second, all must have similar factor scores on each dimension of
their respective cognitive structures (e.g. all parties view cash as being equally positive
on the evaluative dimension, all view cash as being equally dynamic on the activity
dimension, etc.). For this reason, a second hypothesis is required to provide a basis for
determining possible differences in the measured meaning between the three participant
groups of the old and new de?nitions of cash, respectively:
H
02
. There is no signi?cant difference in the measured meaning of the concept
“cash”, as de?ned in either FRS 10 or NZ IAS 7, between each of the three
?nancial reporting groups.
If pairs of reporting groups fail to exhibit a shared cognitive structure (as established
under H
01
), H
02
will not be able to be tested for those pairs. In this case, the connotative
meaning of cash will be assumed not to be shared and the analysis will proceed directly
to H
03
for those reporting group pairs.
Hronsky and Houghton (2001) found a signi?cant difference among auditors in the
measured meanings of old and new de?nitions of “extraordinary items”, respectively,
under Australian Accounting Standards. Earlier, Houghton and Messier (1990)
con?rmed the existence of signi?cant differences among auditors and bankers
(and between these two groups) in their interpretation of the meaning of differently
worded audit reports and different types of audit report opinions. These results suggest
that individual reporting groups may exhibit signi?cant intragroup differences in the
measured meanings of alternative de?nitions of speci?c accounting concepts.
The recent change in the de?nition of the concept of cash, brought about through
New Zealand’s transition to IFRS, might be expected to result in such intragroup
differences. Therefore, the third hypothesis is as follows:
The meaning
of cash
29
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
H
03
. There is no signi?cant difference in the measured meaning of the concept
“cash”, as de?ned in either FRS 10 or NZ IAS 7, within each of the three
?nancial reporting groups.
Hronsky and Houghton (2001) found evidence of a relationship between the decision
outcomes of each participant and the meaning they attributed to relevant accounting
concepts. In their study, a change in the de?nition of the term“extraordinary items” was
seen to be associated with changes in auditors’ perceived meaning of this term. In turn,
this change in meaning was found to lead to different classi?cation decisions across a
number of cases.
In the context of the current study, it is expected that participants will make
different decisions regarding the classi?cation of potential cash items (the degree to
which an item is or is not an item of cash for the purpose of the cash ?ow statement),
depending on which of the two de?nitions (FRS 10 and NZ IAS 7) they are exposed to.
Further, it is expected that any change in connotative meaning brought about by the
change in the de?nition of cash will be associated with a change in the decision
outcomes for each of the ?nancial reporting groups. This gives rise to the fourth and
?fth hypotheses, respectively:
H
04
. There is no signi?cant difference in the decision outcomes made by each of
the three ?nancial reporting groups on the basis of the old FRS 10 de?nition of
“cash” and the new NZ IAS 7 de?nition.
H
05
. The variability in the decision outcomes within each of the three ?nancial
reporting groups is not explained by the variability in the meaning of the
concept upon which the decisions are based.
4. Research method
This study consists of a between-participants 2 £ 3 factorial design in which the ?rst
factor re?ects the old and new de?nition of the concept of cash while the second factor
re?ects the three ?nancial reporting groups: preparers, auditors and users.
4.1 Participants
The three ?nancial reporting groups utilised in the current study are seen to play
a speci?c and distinct role in the accounting communication process (Bedford and
Baladouni, 1962). While prior measure of meaning studies have used each of these
?nancial reporting groups either individually or in pairs (Oliver, 1974; Flamholtz and
Cook, 1978; Houghton, 1987a; Houghton and Hronsky, 1993; Hronsky and Houghton,
2001), none have combined all three in conjunction with two de?nitions of the same
concept.
To help ensure consistency across the range of preparer participants, the “preparer”
group has been limited to those who hold a current chartered accounting (CA)
quali?cation, or international equivalent, and are currently working in an accounting
decision-making role[8]. The members’ list from The New Zealand Institute of
Chartered Accountants (NZICA) provided the source listing for this participant group.
The second group of research participants are “experienced auditors”, de?ned as
those having had no less than three years’ audit experience. Hronsky and Houghton
(2001) argued that third year auditor have suf?cient experience to make an initial
judgment on issues of technical meaning and even though they do not prepare the
ARJ
24,1
30
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
?nancial accounts, they have suf?cient in?uence over the ?nal presentation to
contribute to the accounting communication process. The auditor participant group
was randomly drawn from three of the Big 4 CA ?rms’ audit departments, operating in
the three major centres in New Zealand.
While it is agreed that there are many different classes of ?nancial statement users,
it is commonly accepted that investors represent a company’s principal consumer of
published ?nancial reports (Barth et al., 2001). The current study sampled all current
members of the Certi?ed Financial Analysts’ Society of New Zealand (CFA) as
representatives of the user participant group. Owing to the quali?cation requirements
of the CFA the results of this study should indicate a more consistent or stable
cognitive structure within the user group than if a more heterogeneous subpopulation
of investors had been sampled[9]. Given their quali?cation requirements, this group
may be considered “experienced” users.
As in prior measurement of meaning studies, the data analysis tool used in this study
is factor analysis. In order to establish the stability of the emerging factor pattern,
a researcher must ensure that the sample size is adequate to achieve statistical
signi?cance (DeVellis, 2003; Hair et al., 1998). For the current study, a sample size of
seven participants per semantic scale item was deemed adequate, resulting in a target
sample size of 154 participants (seven participants £ 22-scale items). Based on an
estimated 35 per cent response rate, a total of 446 participants were contacted by e-mail
requesting participation in the study. The response rate was acceptable, at 46 per cent
for preparers, 44 per cent for auditors and 30 per cent for users, providing an overall
response rate of 39 per cent. The response rate for each subgroup is represented in
Table II.
4.2 Research instrument and administration
E-mail and the internet were used to request, administer and record the information
pertaining to the research instrument[10]. Participants were randomly assigned to the old
and new de?nition of cash based on participant lists (sample frames) obtained from
NZICA, the three CA?rms andthe CFA. The “between-participants” design helpedensure
that each participant’s responses were independent of the other responses, and that their
responses were not confounded by such factors as practice or experience gained in other
treatments or fatigue or boredom from participating in a series of treatments.
The web-based research instrument was made up of four key tasks. Task (1)
required each participant to consider the concept of cash, as de?ned by their allocated
de?nition (either FRS 10 or NZ IAS 7), then assess its meaning according to the
22 seven-point semantic differential scales provided (Appendix 1)[11]. To address the
decision-making implications of the allocated de?nition, two further tasks (tasks (2)
and (3)) were presented to each participant. Task (2) presented a series of ten speci?c
Group
Total e-mails
sent
FRS 10
de?nition
NZ IAS 7
de?nition
Total
responses
Total response rate
(%)
Preparers 124 27 30 57 46
Auditors 156 35 33 68 44
Users 166 23 27 50 30
Total 446 85 90 175 120
Table II.
Response rates
The meaning
of cash
31
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
assets (referred to as cases for the purpose of this study). For each case, participants
were instructed to indicate whether or not they believed the case to represent “clearly
an item of cash” when applying their respective de?nition of the concept of cash.
As recommended by Hronsky and Houghton (2001), task (3) required each participant
to indicate the degree to which they believed each case noted in task (2) represented an
item of cash (as it relates to the cash ?ow statement), which helped determine the
degree of con?dence each participant had in each decision outcome. For this purpose,
participants responded on a six-point Likert scale (with “1” representing “clearly an
item of cash” and “6” representing “clearly not an item of cash”) (Appendix 2 for
examples of tasks (1) and (3)).
5. Results
The data analysis stage is divided into four parts. The ?rst part involves factor analysis
of the measurement of connotative meaning data obtained fromthe semantic differential
scale responses. The aim of this analysis is to establish and label the underlying factor
structure (cognitive structure) for each ?nancial reporting group. This helps establish
whether intra and intergroup comparisons can be made. Inso doing, the analysis enables
testing of H
01
and helps determine the feasibility of testing H
02
. The second part of the
analysis focuses on changes in factor placements along the dimensions making up each
reporting group’s factor structure in order to ascertain whether changes in measured
meaning may be attributed to the change in de?nition of cash (i.e. testing of H
03
). The
third part focuses on the analysis of the decision outcomes (the indication of whether or
not the asset presented in each case represented an itemof cash and the degree to which
the participant believes the assets listed in the instrument represent items of cash) and
whether these have been impacted by the change in regulated de?nition (i.e. testing
of H
04
). The ?nal part deals with the connection between the measured meanings and
decision outcome(s) (i.e. testing of H
05
). It is this analysis that provides the most vital
evidence of the relationship between the de?nition of an accounting concept and the
decisions made by participants resulting from their meaning.
5.1 Factor analysis
The determination of the number and nature of factors held by each ?nancial reporting
group is fundamental in determining the degree of shared cognitive structure (Houghton,
1988). Inorder to ?rst determine the number of factors withinwhich the meaning of cash is
held by each of the three ?nancial reporting groups, each group’s data are independently
factor analysed (with varimax rotation). This procedure reduces the responses relating to
the 22 semantic differential scale items to a smaller number of factors, and is consistent
with the approach followed in the measure of meaning literature (Osgood et al., 1957;
Houghton, 1987a, b, 1988; Houghton and Hronsky, 1993; Houghton and Messier, 1990;
Bagranoff et al., 1994; Hronsky and Houghton, 2001; Wines, 2006).
Initially selecting factors with eigenvalues greater than 1.0 resulted in a ?ve-factor
structure (solution) for both the preparer and user groups, and a six-factor structure for
the auditor group. However, after employing a standard scree test (Cattell, 1966) and
then the more rigorous factor comparability test recommended by both Nunnally
(1978) and Everett and Entrekin (1980) (and subsequently introduced to the
measurement of meaning literature by Houghton (1987a, b, 1988))[12], a stable two
factor solution was obtained for both preparers and auditors, while a (less complex)
ARJ
24,1
32
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
single-factor structure resulted for users. Table III presents a summary of the results of
the factor comparability test for each of the three ?nancial reporting groups.
The results suggest that the structure within which the meaning of the concept of
cash is held for users is relatively simple (as re?ected by a single factor cognitive
structure), and differs to that held by both preparers and auditors, who hold a more
complex two factor structure. While it is clear, then, that the user group is not
comparable with the other ?nancial reporting groups, further investigation into the
preparers’ and auditors’ factors is required to determine whether their respective two
factor structures are indeed comparable. That is, although two factors appear to
underlie both groups’ cognitive structures with respect to the concept of cash, it is not
clear that the two factors are comparable between the groups. Non-comparability could
arise due to each of the factors for each group comprising a different set of highly
loading semantic scales. To explore this issue, further insight into the relationship
between the respective factors is required. Consequently, a factor comparison of each
reporting group’s factors is reported in Table IV.
The results suggest that the ?rst factor for all three ?nancial reporting groups is of
a similar nature, with the simple average comparability coef?cient for these factors
being 0.92, and Cronbach’s a being 0.97. While the correlation for factor 1 between
auditors and preparers does not speci?cally reach the cutoff suggested by Everett and
Entrekin (1980) of 0.894, the correlation of 0.876 is very close and indicates a high level
of shared variance of 76.7 per cent. In contrast, preparers’ and auditors’ second factors
appear unrelated, with a comparability coef?cient not signi?cantly different from zero.
Correlations between factor scores of random split
halves
a
Group Factor 1 Factor 2
Preparers
No. of factors ¼ 2 0.918 0.923
Auditors
No. of factors ¼ 2 0.941 0.976
Users
No. of factors ¼ 1 0.996
Note:
a
Correlation coef?cients of at least 0.894 indicate a comparable and robust factor structure
(Everett and Entrekin, 1980)
Table III.
Within group factor
comparability
Factor Groups Correlation
Panel A
Factor 1 Preparers-auditors 0.876
*
Preparers-users 0.986
*
Auditors-users 0.903
*
Panel B
Factor 2 Preparers-auditors 0.049
Note: Signi?cance at:
*
0.01 level (two-tailed)
Table IV.
Between group factor
comparability
The meaning
of cash
33
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
In summary, the assumption of a shared cognitive structure between any pair of
?nancial reporting group is not justi?ed.
As discussed previously, determining whether different groups share connotative
meaning is a two-step process. First one determines whether comparable factor
(cognitive) structures exist between groups (H
01
), and then, and only then, can one test
for shared measured meaning between groups (H
02
). As the results suggest a lack of
shared cognitive structure between any pair of ?nancial reporting group, H
01
is
rejected and the study proceeds on the basis that all three ?nancial reporting groups
have different cognitive structures and therefore can be viewed as three separate factor
groups. Consequently, the testing of H
02
is neither possible nor necessary. However,
the testing of within group differences in measured meaning of the concept of cash, as
it relates to the cash ?ow statement, may proceed as planned (i.e. testing of H
03
).
5.2 Factor labelling
Having established the number of factors underpinning each group’s cognitive
structure, attention may turn to understanding the nature of these factors.
This is achieved through the process of factor labelling and requires examination of
the factor loadings of each semantic differential scale item. Consistent with the
measurement of meaning literature in accounting (Houghton, 1988; Hronsky and
Houghton, 2001; Wines, 2006), those scale items which load heavily are deemed to be
those with factor loadings greater than 0.5. The rotated factor matrix (using varimax
rotation) for the three ?nancial reporting groups is presented in Table V.
Given that all three ?nancial reporting groups in this study have fewer than three
factors underlying their respective cognitive structures, it is apparent that their
structures may depart somewhat from the classic EPA structure. Osgood et al. (1957)
acknowledged that variations to this structure may occur depending on the nature of
the respondents and concepts under examination. Similarly, Osgood (1976, p. 37)
suggested that the greater the emotional or attitudinal loading of the concept being
judged, the greater the tendency of the semantic framework to collapse to fewer
dimensions. Houghton (1987a, 1988) also ?nds the applicability of the EPA structure in
accounting only under certain circumstances (Houghton and Messier, 1990).
Factor 1 for all three ?nancial reporting groups (shown earlier to be comparable
between groups) is consistently associated with scale items associated with potency
and evaluative dimensions. Some of the highest loading items are of a potency nature
(e.g. weak-strong, intangible-tangible and estimated-exact). Evaluative scales loading
highly include items such as subjective-objective, unexpected-expected and
discretionary-required. With respect to the concept of cash, it is perhaps to be
expected that potency and evaluative scales load onto a single factor. For cash, the
measurable, exact, strong and tangible potency scales are similar in effect to the safe,
objective and expected evaluative scales. A recent study by Wines (2006) also ?nds a
dominant factor consisting of both potency and evaluative dimensions, which he labels
emphasis. Consistent with this nomenclature, this study adopts the emphasis label for
factor 1 for all ?nancial reporting groups. Table V reveals that the emphasis factor
dominates factor 2 for both preparers and auditors, as re?ected by its comparatively
high eigenvalues of 10.145 and 8.429 for preparers and auditors, respectively.
Relative to factor 1, the labelling of factor 2 for the preparer group is more
straightforward. This factor is dominated by scales associated with the activity
ARJ
24,1
34
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
dimension of the EPA structure, such as constant-variable and dynamic-static.
Accordingly, this factor is labelled activity.
As shown previously in Table IV, the auditors’ factor 2 is not comparable with the
preparers’ factor 2. Factor 2 for auditors consists of scales relating to all three
dimensions of the EPA structure. In particular, it has several activity and evaluative
scales loading onto it which are of a temporal nature (e.g. temporary-permanent, long
term-short term and expected-unexpected). Further, it contains the potency scale items
of direct-indirect and imaginary-real, both of which have previously been included in a
sub-set of the potency dimension referred to as substantiveness in several prior studies
(Houghton and Hronsky, 1993; Houghton and Messier, 1990; Bagranoff et al., 1994). If
cash is considered to be permanent and long term, then one might also expect it to be
considered real and direct. Consequently, the temporal and substantiveness aspects of
factor 2 appear to be logically related. Wines (2006) encountered a similar factor
in his study, which he labelled variability. Since six out of the eight heavily loading
factor 2 scales correspond to scales heavily loading on to Wines’ variability factor, it is
considered appropriate to label auditors’ factor 2 as variability.
Preparer Auditor User
factors
a
factors
a
factor
a
1 2 1 2 1
Scale
b
EPA Emphasis Activity Emphasis Variability Emphasis
Good-bad E 20.456 0.422
Unnecessary-necessary E 0.662 0.569
Subjective-objective E 0.754 0.798 0.887
Risky-safe E 0.736 0.696 0.783
Required-discretionary E 20.788 20.489 0.560 20.763
Adverse-bene?cial E 0.640 0.431 0.522
Uncontrollable-controllable E 0.705 0.488 0.462
Expected-unexpected E 20.780 0.681 20.761
Unmeasurable-measurable P 0.810 0.864 0.850
Estimated-exact P 0.846 0.866 0.788
Intangible-tangible P 0.870 0.720 0.861
Weak-strong P 0.932 0.846 0.894
Direct-indirect P 20.847 20.493 0.575 20.794
Incomplete-complete P 0.809 0.674 0.808
Imaginary-real P 0.833 0.508 20.504 0.850
Flexible-in?exible P 0.835 0.664 0.754
Unplanned-planned A 0.682 0.626
Constant-variable A 20.815 0.558 20.691
Permanent-temporary A 20.451 20.431 0.703 20.582
Active-passive A 20.486 0.619 20.572
Dynamic-static A 0.713 0.426
Long term-short term A 0.541
Initial eigenvalue 10.145 2.439 8.429 2.194 10.413
Percentage of variance 46.112 11.086 38.313 9.971 47.330
Cumulative percentage 46.112 57.198 38.313 48.284 47.330
Notes: Based on a two-factor solution for preparer and auditor groups, and a single factor solution for
users;
a
factor loadings ,0.40 are suppressed;
b
each bipolar adjectival scale ranged from “1” (left-hand
scale endpoint) to “7” (right-hand scale endpoint)
Table V.
Rotated factor matrix by
?nancial reporting group
The meaning
of cash
35
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
In summary, the identi?cation of a single factor structure (corresponding to
emphasis) for the user group indicates a simpler (unidimensional) structure than the
factor structure identi?ed by Osgood et al. (1957), while both the preparer and auditor
?nancial reporting groups conform to a more complex (dimensional) structure
(emphasis/activity for preparers and emphasis/variability for auditors).
5.3 Factor placement
Having established that the cognitive structure for each ?nancial reporting group is
both robust and stable, we now investigate the possible differences in meaning within
each structure for each reporting group. This will address H
03
. Given that the cognitive
structure of each group is different, the analysis will focus on each ?nancial reporting
group separately.
Mean placement (factor) scores indicate the position of the concept cash for each
group on the factor axes corresponding to their cognitive structures. Movements in
placements (along the factor axes) between the old and new de?nition provide evidence
of both the extent (degree) and direction of a change in meaning[13]. Table VI reports
the factor placements by ?nancial reporting group.
To determine whether there are any signi?cant differences in perceived connotative
meaning of the concept of cash between the two de?nition groups (FRS 10 vs NZ IAS 7)
for each reporting group, analysis of variance (ANOVA) was performed on the factor
scores by group (Table VII).
The ANOVA results reveal a signi?cant change in factor scores on the dominant
factor, emphasis, for all three ?nancial reporting groups in response to NZ IAS 7
(preparers: F ¼ 4.996, p , 0.005; auditors: F ¼ 23.363, p , 0.01; and users:
F ¼ 36.881, p , 0.01). For instance, for preparers, the factor placement for the
emphasis factor is 230 under the old de?nition but 27 under the new de?nition. The
signi?cance of F for this factor for preparers pursuant to the ANOVA indicates that
these two values are signi?cantly different. For all three groups, there has been a
positive shift along the emphasis factor axis, suggesting that the new de?nition is
perceived as being more potent and is evaluated more favourably. In other words,
“cash”, as it relates to cash ?ows under NZ IAS 7, is seen as stronger and more
tangible, direct, exact, real, complete and measurable, as well as being more required,
expected, objective, safe, controllable, necessary, bene?cial and planned,
Group Factor(s)
Preparers Emphasis Activity
FRS 10 (old) de?nition 230 222
NZ IAS 7 (new) de?nition 27 20
Auditors Emphasis Variability
FRS 10 (old) de?nition 249 5
NZ IAS 7 (new) de?nition 52 25
Users Emphasis
FRS 10 (old) de?nition 270
NZ IAS 7 (new) de?nition 60
Note: Consistent with prior measurement of meaning studies in accounting, factor placement scores
have been multiplied by 100 to aid interpretation
Table VI.
Factor placements by
?nancial reporting group
ARJ
24,1
36
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
relative to the de?nition found in FRS 10. The apparent changes, however, along the
activity factor axis for preparers (F ¼ 2.519, p ¼ 0.118), and the variability factor axis
for auditors (F ¼ 0.187, p ¼ 0.667), were not statistically signi?cant, perhaps re?ecting
the subordinate nature of these two factors relative to the emphasis factor.
Overall, the results of the ANOVA provide empirical evidence that a change in
connotative meaning has occurred between the old and new de?nition of the concept
cash for all three ?nancial reporting groups, leading to a rejection of the null H
03
.
5.4 Decision outcomes
In order to address H
04
, the analysis determines whether a change in the regulated
de?nition of cash is associated with changes in decision outcomes (relating to ten asset
classi?cation cases)[14]. Given the previously established differences in cognitive
structure between the three ?nancial reporting groups, each group’s results were
assessed independently. Table VIII presents a comparison of classi?cations prompted
by NZ IAS 7 and FRS 10 for each reporting group, together with the results of ANOVA.
Table VIII reveals that the effects of the treatment were consistently signi?cant for
cases 3, 9 and 10 across all ?nancial reporting groups. Both the “four-month treasury
bill” and “on-call bank overdraft” were rated as closer to not being items of cash, while
“preference shares redeemable in three months for a ?xed amount of cash” were seen
as being closer to an item of cash under NZ IAS 7 relative to FRS 10. Both NZ IAS 7 and
FRS 10 de?nitions allude to cash potentially including highly liquid investments,
however the former requires there to be minimal risk surrounding the realisable value
of the asset (i.e. investments included in cash must be convertible for known amounts
of cash and be subject to an insigni?cant risk of changes in value). Consequently, the
risk created by the relatively long maturity (four months) on the treasury bills most
likely contributed to it being viewed as closer to not being an item of cash for
participants prompted by the NZ IAS 7 de?nition relative to participants in the FRS
10 group. Conversely, the ?xed redeemable value nature of the preference shares may
have led some NZ IAS 7 participants to view them as having a limited risk, which
would likely contribute to this asset being rated closer to an item of cash than for the
same asset under FRS 10. The fact that FRS 10 makes direct mention of bank
overdrafts potentially being included in cash would have undoubtedly led these
participants to view such items as being closer to cash than participants viewing the
NZ IAS 7 de?nition. It is interesting to note that the extent of the treatment effects was
suf?cient enough to result in change in overall classi?cation of an item (either from
Group F
Preparers
Factor 1 – emphasis 4.996
*
Factor 2 – activity 2.519
Auditors
Factor 1 – emphasis 23.363
* *
Factor 2 – variability 0.187
Users
Factor 1 – emphasis 36.881
* *
Note: Signi?cance at:
*
0.05 and
* *
0.01 levels, respectively
Table VII.
ANOVA of factor
placements for old
and new de?nitions
The meaning
of cash
37
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
cash to not an item of cash or vice versa) for cases 3, 9 and 10 for the user group, and
case 10 for the auditor group.
Further, Table VIII reveals that the user and auditor groups made different
decisions under NZ IAS 7 relative to FRS 10 for case 6 (“readily tradable securities”).
These assets were more likely to be viewed as not being an itemof cash under NZ IAS 7,
presumably due to their inherent risk of changes in value. The last signi?cant
difference noted in Table VIII relates to auditors only. Case 5 (“gold bullion”) is seen as
being closer to not being an item of cash under the NZ IAS 7 treatment, again,
presumably due to its potential risk of change in value.
On the basis of the discussion above, it appears that by incorporating criteria
relating to the riskiness of certain assets (highly liquid investments), NZ IAS 7’s
de?nition is instrumental in clarifying classi?cation decisions of the three ?nancial
reporting groups, particularly for the auditor group. For highly liquid investments,
the new standard is seen as being “tougher” (that is, has more emphasis – is more
objective, less discretionary, more exact) by all groups. Given the evidence of
differences in classi?cation decisions between the two treatment groups (old vs new
standard) the null H
04
is rejected.
Mean decision score
a
Preparers Auditors Users
Case Item
Old
de?nition
New
de?nition
Old
de?nition
New
de?nition
Old
de?nition
b
New
de?nition
1 Coins and notes on
deposit 1.00 1.00 1.00 1.00 1.00 1.00
2 Accounts receivable 4.96 4.90 4.94 5.27 4.86 4.70
3 A four-month treasury
bill 4.26
c
5.07
* *
,d
3.83 5.79
* * *
,d,e
2.95
c
5.07
* * *
,e
4 Reserve bank bill 3.33 2.80 2.71 2.82 2.23 2.15
5 Gold bullion 4.52 4.33 3.89 4.76
*
4.05 4.41
6 Readily tradable equity
securities 4.48 4.37
d
4.63
f
5.24
*
,d
3.55
f
4.52
*
7 A three-month futures
contract
5.44 5.40 5.03 5.21 5.23 4.85
8 Non-cash payment for
goods and services
(barter) 5.30
g
5.63 5.80
g
5.73 5.73 5.81
9 On-call bank overdraft 1.44 3.27
* * *
1.31 2.45
* *
,e
1.45 4.33
* * *
,f
10 Preference shares
redeemable in three
months for a ?xed
amount of cash 4.89 4.07
*
,d,h
4.20 2.39
* * *
,d
4.00 2.63
* *
,h
Notes: Signi?cance of difference between old and new de?nitions at:
*
0.05,
* *
0.01 and
* * *
0.001
levels, respectively;
a
possible scores ranged from “1” (clearly an item of cash) to “6” (clearly not an item
of cash);
b
one user participant responded with “6” for all ten cases and, consequently, has been
excluded as an outlier from this and subsequent analyses;
c,f,g
signi?cant pairwise differences between
preparers and auditors, preparers and users, and auditors and users, respectively, who received the old
de?nition ( post hoc comparison using the Bonferroni test, p , 0.05);
d,e,h
signi?cant pairwise
differences between preparers and auditors, preparers and users, and auditors and users, respectively,
who received the new de?nition ( post hoc comparison using the Bonferroni test, p , 0.05)
Table VIII.
ANOVA of decision
scores by ?nancial
reporting group
ARJ
24,1
38
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
While the focus of this section has been on the effects of a change in de?nition on the
decision outcomes of each reporting group, it is interesting to note that Table VIII also
reveals differences between preparers, auditors and users for each de?nition. The old
de?nition was associated with differences for certain pairs of reporting groups on cases
3, 6 and 8, while the new de?nition revealed pairwise differences on cases 3, 6, 9 and 10.
The new de?nition has created less consistency among the key parties to the ?nancial
reporting process with respect to their reporting decisions.
5.5 Decision outcomes and measurement of meaning
In this part of the data analysis, multiple regression analysis is employed to
examine H
05
. This hypothesis examines whether the variability in the decision
outcomes within each of the three ?nancial reporting groups may be explained by the
variability in the meaning of the concept upon which the decisions are based.
For each ?nancial reporting group, classi?cation decisions are regressed on
factor scores (corresponding to emphasis and activity for preparers, emphasis and
variability for auditors, and emphasis for users). To control for the effects of differences
in the nature of case facts, nine dummy variables are included in each equation to
represent the ten cases[15]. Further, unlike Hronsky and Houghton (2001) who did
not control for the effects of old vs new de?nition, a dummy variable representing
de?nition type (old vs new) is also included in each regression. To determine whether
multicollinearity was an issue (particularly between de?nition type and the factors),
collinearity diagnostics were reviewed. These did not reveal any cause for concern with
collinearity (all variance in?ation factors fell below 1.90). Table IX presents the results
of the regression analysis.
Overall, the three models, as speci?ed, are signi?cant (F ¼ 48.796 ( p , 0.001),
F ¼ 80.340 ( p , 0.001) and F ¼ 46.292 ( p , 0.001)) and explain 50.2, 58.5 and
50.5 per cent of the variance in classi?cation decisions made by the preparer, auditor,
Estimate
Parameter Preparers Auditors Users
Constant 4.419
* * *
3.090
* * *
2.941
* * *
Factor 1 (emphasis) 20.036 20.147
*
20.074
Factor 2 (activity) 0.170
* *
Factor 2 (variability) 20.001
De?nition (old vs new) 0.070 0.482
* * *
0.512
* *
Case 1 23.456
* * *
22.324
* * *
22.224
* * *
Case 2 0.474 1.779
* * *
1.571
* * *
Case 3 0.228 1.456
* * *
0.898
* *
Case 4 21.404
* * *
20.559
*
20.980
* * *
Case 5 20.035 0.985
* * *
1.061
* * *
Case 6 20.035 1.603
* * *
0.878
* * *
Case 7 0.965
* * *
1.794
* * *
1.796
* * *
Case 8 1.018
* * *
2.441
* * *
2.551
* * *
Case 9 22.053
* * *
21.456
* * *
20.163
F 48.796
* * *
80.340
* * *
46.292
* * *
Adjusted R
2
0.502 0.584 0.505
Notes: Signi?cance at:
*
0.05,
* *
0.01 and
* * *
0.001 levels, respectively
Table IX.
Regression of
classi?cation decisions on
factors, de?nition and
cases by ?nancial
reporting group
The meaning
of cash
39
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
and user groups, respectively. It may be seen from a review of the coef?cients
corresponding to each factor that the relationship between measured meaning and
decision outcomes is driven by factor 2, “activity” (coef?cient ¼ 0.170, p , 0.01), for
the preparer group; and factor 1, “emphasis” (coef?cient ¼ -0.147, p , 0.05), for the
auditor group. The users’ emphasis factor was not signi?cantly related to decision
outcomes.
“Activity” appears to be the most in?uential dimension of meaning for preparers.
The positive association between this dimension of meaning and decision outcomes
suggests that low (high) placements on the “activity” axis are associated with low
(high) decision scores[16]. Accordingly, a ?nancial report preparer with a low “activity”
factor score (corresponding with a view that cash is dynamic, variable and ?exible),
would tend to rate assets they associate with such characteristics as closer to the
“clearly an item of cash” (i.e. lower) end of the decision scale, than a preparer with a
higher “activity” score.
The dimension of meaning most in?uential for auditors was “emphasis”. “Emphasis”
is negatively associated with auditors’ decision scores. Consequently, an auditor with
a high “emphasis” score (corresponding with a view that cash is measurable, exact,
strong, objective, tangible, etc.) would be expected to rate assets perceived to possess
such traits closer to the “clearly an item of cash” end of the scale than an auditor with a
lower “emphasis” score.
In summary, meaning was found to independently relate to classi?cation decisions
for two out of the three ?nancial reporting groups, providing support for the rejection
of H
05
. “Activity” was the dimension of measured meaning most likely to in?uence the
classi?cation decisions of preparers, while for auditors it was the “emphasis” factor.
In part, the nature of the de?nition (NZ IAS 7 vs FRS 10) and case facts were also seen
to directly in?uence decision outcomes.
6. Discussion and limitations
The results of the current study indicate that the three key groups of parties to the
?nancial reporting process do not share the same cognitive structure in which the
meaning of the concept of cash (as it relates to cash ?ow) is held, thus allowing H
01
to be
rejected. Users were found to have a simple (unidimensional) structure consisting of an
“emphasis” factor, whilst the structures within which meaning was held for preparers
and auditors was somewhat more complex. The preparers’ cognitive structure was
represented by a semantic space consisting of two factors: “emphasis” and “activity”.
The two axes making up auditors’ semantic space consisted of “emphasis” and
“variability”. The lack of shared cognitive structure between the three groups implies
that the meaning of the concept cash is also not shared between these groups, raising
concerns about the effectiveness of the communication between the key players in the
?nancial reporting process. The lack of shared understanding could, in extreme
circumstances, result in misallocations of capital, as decisions made by the users
are based on their own understanding of the information and not on any uni?ed basis.
The reliability of the cash ?ow statement could be hampered by this fact, raising
questions about its susceptibility to intentional and unintentional miscommunication.
The lack of shared meaning between auditors and users, in particular, highlights
a communication gap paralleling the well-documented “audit expectations gap” – the
gap between what the public expects of an audit, and what auditors themselves believe
ARJ
24,1
40
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
their role to encompass. Prior studies examining the latter gap have largely focussed
on determining the nature and extent of differences between the public’s and auditing
profession’s perceptions of the auditor’s role and duties (Libby, 1979; Humphrey et al.,
1993; Monroe and Woodliff, 1993; Porter, 1993; McEnroe and Martens, 2001). However,
few studies have considered the impact of differences in meanings ascribed to the audit
subject matter (or elements thereof). The ?ndings of this study suggest that this could
be an avenue worthy of further research.
While the results of the semantic differential technique in this study only partially
supported Osgood et al.’s (1957) three-factor structure (EPA), they were consistent with
the application of this technique by other accounting researchers in the measurement
of meaning literature. However, it is recommended that future research reconsider the
adequacy of the 22-item semantic scale in the contemporary accounting environment to
provide assurance that these scales remain suf?ciently sensitive to capture the subtle
nuances of current terminology in the domain of accounting. Indeed, both
Osgood et al. (1957) and Haried (1972) supported an ongoing review of the semantic
scales to ensure that they continued to meet the requirements of the terms and concepts
under examination.
Turning to the analysis of measured meaning for each of the three ?nancial
reporting groups (H
03
), the results suggest that each group’s cognitive structure was
sensitive enough to result in a measurable difference in connotative meaning brought
about by changes in the de?nition of cash (FRS 10 vs NZ IAS 7). The new de?nition
was associated with a consistently positive movement along the “emphasis” axis
(a dimension of meaning held by all three ?nancial reporting groups). However, no
signi?cant change occurred along the “activity” axis for preparers, whose cognitive
structure included this axis. Likewise, no signi?cant change occurred along auditors’
second axis, labelled “variability”.
The decision-outcome analysis relating to H
04
con?rmed that the newer standard’s
(NZ IAS 7) de?nition of cash was instrumental in clarifying classi?cation decisions
for all three ?nancial reporting groups, particularly for highly liquid investments.
The results highlight that even subtle changes in the wording of of?cial de?nitions of
accounting terms and concepts may potentially lead to different decisions being made
by preparers, auditors and users. This result should be considered by standard-setters
when looking at amending or introducing new de?nitions or accounting standards, to
ensure that any such changes in decision outcomes are consistent with those intended.
This issue is of particular signi?cance in jurisdictions where two or more ?nancial
reporting frameworks coexist (e.g. in New Zealand, Australia, China, the UK). The
study demonstrates that two standards containing similar but not identical de?nitions
could result in inconsistent ?nancial reporting decisions, thereby further reducing the
comparability of the resulting ?nancial reports.
While there are many in?uences on?nancial reportingjudgments, suchas incentives,
training, availability of relevant resources, this study ?nds that connotative meaning is
an important determinant. After controlling for the effects of case differences and
alternative de?nitions of cash, “activity” was found to be the dimension of measured
meaning most likely to in?uence the classi?cation decisions of preparers, while for
auditors it was the “emphasis” dimension. This result is consistent with Hronsky and
Houghton (2001) who found potency important for auditors in “extraordinary item”
classi?cation decisions.
The meaning
of cash
41
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The results of this study must be considered in light of several limitations. It is
important to acknowledge the explanatory power of the typical three (EPA) factor
structure. Osgood et al. (1957) explainthat these factors typically explain50 per cent of the
total phenomena knownas “meaning”, withthe remaining50per cent beingmade upof an
almost in?nite number of variables. This notwithstanding, the Osgoodmodel continues to
be used widely across a wide range of disciplines, including accounting, to study the
interpretation or process of constructing meaning (Hronsky and Houghton, 2001).
The use of an experimental design in this study may have implications for the
generalisability of the study’s ?ndings. While a particular strength of the study is the
relatively large sample size and the inclusion of representatives of the three main
groups of parties to the ?nancial process (preparers, auditors and users), it is
acknowledged that ?ndings of the study may not generalise to groups not represented
in the study’s samples, such as unsophisticated users and non-quali?ed accountants.
While every effort was made to ensure each of the ten cases were representative of
assets found in practice, they naturally do not include all possible assets, nor are they
included in proportion to the frequency they would likely appear in practice.
7. Conclusion
Focusing on three key groups of parties in the ?nancial reporting process (preparers,
auditors and users), this study examined the impact on connotative meaning and
decision behaviour of a change in the promulgated de?nition of “cash”, as it relates to
the cash ?ow statement, necessitated by convergence with IFRS in New Zealand.
Results suggest that the three ?nancial reporting groups do not hold the meaning of
cash within the same cognitive structure. This ?nding supports the proposition that
the receivers of accounting information may not be interpreting that information in the
manner intended by the sender. Given that accounting communication is reliant on
shared meaning, such miscommunication may be leading to unnecessary
misunderstandings and tensions among the many parties to the reporting process.
A further ?nding of this study was that the measured meaning of cash for each of the
three ?nancial reporting groups was sensitive to the alternative de?nition of cash adopted
by NZ IAS 7. Similarly, classi?cation decisions were found to be in?uenced by the
particular de?nition of cash assigned to the research participants. These results suggest
that subtle differences inregulatedde?nitions of accountingconcepts have the potential to
cause unanticipated changes in meaning and consequential changes in decision
outcomes – even for seemingly uncontroversial and uncontested concepts in accounting.
In ?nding a direct association between measured meaning and decision outcomes for
both auditor and preparer groups, this study extends the work of Hronsky and Houghton
(2001) and con?rms the importance of connotative meaning in the domain of accounting.
Notes
1. Osgood et al. (1957, p. 321) describe connotative meaning as the emotional or metaphorical
meaning attached to a speci?c symbol (word, phrase, etc.). In contrast, denotative meaning
refers to the literal meaning of a symbol and involves the communication of an objective
description about the object (Adelberg and Farrelly, 1989). Agreement on denotative meaning
will generally be more easily reached between individuals (subject to similarities in knowledge
of the given ?eld under investigation) than on connotative meaning (Haried, 1973). After
considering both forms of meaning, Adelberg and Farrelly (1989) conclude that differences in
connotative meaning cause the most concern in the accounting communication process.
ARJ
24,1
42
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2. The International Accounting Standards Board (IASB) Chair, Sir David Tweedie, predicts
that 150 countries will have adopted IFRS by 2012 (Pickard, 2007).
3. According to Osgood et al. (195, p. 1597), connotative meaning “[. . .] is an important
intervening variable in human behaviour” and its measurement provides a basis for making
predictions about overt behavioural responses.
4. All New Zealand reporting entities were required to apply IFRS in preparing external
?nancial reports for reporting periods commencing on or after 1 January 2007. However,
entities were permitted to adopt IFRS from 1 January 2005.
5. In 1985, Canada’s accounting regulatory body introduced the cash ?ow statement. By 1992,
authorities responsible for standard-setting in New Zealand, the USA, South Africa, the UK
and the International Accounting Standards Board’s had all followed Canada’s lead.
6. Osgood et al. (1957, p. 25) de?ne semantic space as: “[. . .] a region of some unknown
dimensionality and Euclidian in character. Each semantic scale [. . .] is assumed to represent
a straight line function that passes through the origin of this space, and a sample of such
scales then represents a multidimensional space.”
7. Recent applications of Osgood et al.’s (1957) technique may be found in ?elds such as
information science (Yoon, 2008), marketing (Hsu et al., 2000), medicine (Fitzgerald et al.,
2008; Nekolaichuk et al., 1999), psychology (Xiong et al., 2006) and sociology (Noland et al.,
2004; Pierce et al., 2003).
8. This includes senior accountants, ?nancial controllers, chief ?nancial of?cers or a role
similar in nature.
9. To become a member of the CFA, an investment professional must meet all of the criteria
listed below:
.
hold a bachelor’s degree from an accredited institution or have equivalent education or
work experience;
.
pass level I of the CFA exam or such other appropriate examination as approved by the
Board of Governors or pass the self-administered Standards of Practice Examination;
.
have 48 months involvement in the investment decision-making process; and
.
agree to adhere to and sign the member’s agreement, a professional conduct statement,
and any additional documentation requested by the CFA Institute.
10. The use of the internet as an appropriate survey method has been supported in the literature
(Couper et al., 2001; Dillman, 2007), andpartlyjusti?edbythe response rate inthe current survey.
11. The 22 seven-point semantic differential scales are recognised in the literature as the most
appropriate scale system to measure connotative meaning in the accounting domain
(Houghton, 1987a, 1988; Houghton and Hronsky, 1993; Houghton and Messier, 1990;
Bagranoff, 1990; Bagranoff et al., 1994; Hronsky and Houghton, 2001; Wines, 2006).
12. Factor compatibility is a form of split-halves test, in which each (randomly selected) half is
subjected to identical factor analyses. The two resulting factor score coef?cient matrices may
then be used (normally with the aid of statistical software) to compute two sets of factor scores
for each participant. The Pearson correlation between these two sets of duplicate factor scores
can “be used to judge the comparability of factors in the analysis” (Nunnally, 1978, p. 433) and
is referred to as the “comparability coef?cient” (Everett and Entrekin, 1980, p. 166).
Comparability is said to exist when all factor(s) are robust and stable, as indicated by
comparability coef?cient scores of .0.894 (corresponding to 80 per cent shared variance).
Factor comparability is determined by initially specifying the number of factors to be
extracted, starting with the number of factors determined from a scree test, then reducing by
one factor each time until comparability (comparability coef?cient scores . 0.894) is achieved
The meaning
of cash
43
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
for all identi?ed factors. Practically, this can be achieved by selecting the largest correlation in
the resulting factor comparability matrix, deleting its row and column, then repeating the
process until all the rows and columns are deleted from the matrix (Everett, 1983).
13. This change can be described as the change in semantic space in which meaning is held for
each of the ?nancial reporting groups (Houghton, 1988).
14. The results of the decision outcomes for task (2) (the binary “yes” or “no” classi?cation) were
correlated with the corresponding outcomes on task (3) (asset classi?cation made on a scale
ranging from “1” (clearly an item of cash) to “6” (clearly not an item of cash)). As may be
expected, these two sets of decisions were highly correlated (Spearman’s r ¼ 0.80, p , 0.01)
suggesting that the responses from tasks (2) and (3) may be used interchangeably. Task (3)’s
responses were used in subsequent data analysis because of their better distributional
qualities (Hronsky and Houghton, 2001).
15. A categorical variable with j categories only requires j 2 1 dummy variables “[. . .] in order
to capture all the distributional information contained in the original set of distinctions”
(Hardy, 1993, p. 7).
16. Decisions scores have a theoretical range from “1” (clearly an item of cash) to “6” (clearly not
an item of cash).
References
Adelberg, A.H. and Farrelly, G.E. (1989), “Measuring the meaning of ?nancial statements
terminology: a psycholinguistics approach”, Accounting and Finance, Vol. 29 No. 1, pp. 33-61.
Bagranoff, N.A. (1990), “The semantic differential: a prescription for use in accounting research”,
Journal of Accounting Literature, Vol. 9, pp. 65-80.
Bagranoff, N.A., Houghton, K.A. and Hronsky, J. (1994), “The structure of meaning in
accounting: a cross-cultural experiment”, Behavioural Research in Accounting, Vol. 6,
pp. 35-57 (supplement).
Barth, M.E., Beaver, W.H. and Landsman, W.R. (2001), “The relevance of the value relevance
literature of ?nancial accounting standard setting: another view”, Journal of Accounting
and Economics, Vol. 31, pp. 77-104.
Bedford, N.M. and Baladouni, V. (1962), “A communication theory in accounting”, Accounting
Review, Vol. 37 No. 4, pp. 650-9.
Broome, O.W. (2004), “Statement of cash ?ows: time for change!”, Financial Analysts Journal,
Vol. 60 No. 2, pp. 16-23.
Buzby, S.L. and Falk, H. (1974), “A new approach to the funds statement”, Journal of
Accountancy, Vol. 137 No. 1, pp. 55-62.
Cattell, R.B. (1966), “The scree test for the number of factors”, Multivariate Behavioural Research,
Vol. 1, pp. 245-76.
Chesley, G.R. and Scheiner, J.H. (1982), “The statement of changes in ?nancial position:
an empirical investigation of Canadian and US users in nonpublic companies”,
International Journal of Accounting, Vol. 17 No. 2, pp. 49-58.
Clift, R.C. (1979), The Funds Statement, Australian Accountants Foundation, Melbourne.
Couper, M.P., Traugott, M.W. and Lamias, M.J. (2001), “Web survey design and administration”,
The Public Opinion Quarterly, Vol. 65 No. 4, pp. 230-53.
DeVellis, R.F. (2003), Scale Development Theory and Applications, 2nd ed., Sage, London.
Dillman, D.A. (2007), Mail and Internet Surveys: The Tailored Design Method, 2nd ed., Wiley,
New York, NY.
ARJ
24,1
44
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Epstein, M. (1992), “Howuseful is the statement of cash ?ows?”, Management Accounting, Vol. 74
No. 1, pp. 52-6.
Everett, J.E. (1983), “Factor comparability as a means of determining the number of factors and
their rotation”, Multivariate Behavioural Research, Vol. 18, pp. 197-218.
Everett, J.E. and Entrekin, L.V. (1980), “Factor comparability and the advantages of multiple
group factor analysis”, Multivariate Behavioural Research, Vol. 15, pp. 165-80.
Fitzgerald, J.T., Stans?eld, R.B., Tang, T., Oh, M., Frohna, A., Armbruster, B., Gruppen, L. and
Anderson, R. (2008), “Patient and provider perceptions of diabetes: measuring and
evaluating differences”, Patient Education and Counseling, Vol. 70, pp. 118-25.
Flamholtz, E. and Cook, E. (1978), “Connotative meaning and its role in accounting change: a ?eld
study”, Accounting, Organisations and Society, Vol. 3 No. 2, pp. 115-39.
Goldberg, L. (1951), “The funds statement reconsidered”, The Accounting Review, Vol. 26 No. 4,
pp. 485-91.
Graci, S.P. (1982), “Effects of the statement of changes in ?nancial position on the short term loan
decisions”, PhD dissertation, University of Arkansas, Fayetteville, AR.
Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1998), Multivariate Data Analysis,
5th ed., Prentice-Hall, Upper Saddle River, NJ.
Hardy, M. (1993), Regression with Dummy Variables, University Paper Series on Quantitative
Applications in the Social Sciences, 07-093, Sage, Newbury Park, CA.
Haried, A.A. (1972), “The semantic dimensions of ?nancial statements”, Journal of Accounting
Research, Vol. 10 No. 2, pp. 376-91.
Haried, A.A. (1973), “Measurement of meaning in ?nancial reports”, Journal of Accounting
Research, Vol. 11 No. 1, pp. 117-45.
Heath, L.C. (1978), “Let’s scrap the funds statement”, Journal of Accountancy, Vol. 146 No. 4,
pp. 94-103.
Henry, E.J. (1975), “A new funds statement formal for greater disclosure”, Journal of
Accountancy, Vol. 139, pp. 56-62.
Hicks, B.E. and Hunt, P. (1980), Cash Flow Accounting, The International Group for Cash Flow
Accounting and the Research and Publication Division, School of Commerce and
Administration, Laurentian University, Sudbury.
Houghton, K.A. (1987a), “True and fair view: an empirical study of connotative meaning”,
Accounting, Organisations and Society, Vol. 12 No. 2, pp. 143-53.
Houghton, K.A. (1987b), “The development of meaning in accounting: an intertemporal study”,
Accounting and Finance, Vol. 27 No. 2, pp. 25-40.
Houghton, K.A. (1988), “The measurement of meaning in accounting: a critical analysis of the
principal evidence”, Accounting, Organisations and Society, Vol. 13 No. 3, pp. 263-80.
Houghton, K.A. and Hronsky, J.J.F. (1993), “The sharing of meaning between accounting
students and members of the accounting profession”, Accounting and Finance, Vol. 33
No. 2, pp. 85-101.
Houghton, K.A. and Messier, W.F. (1990), “The wording of audit reports: its impact on the
meaning of the message communicated”, in Moriarty, S. (Ed.), Accounting Communication
and Monitoring, Centre for Economic and Management Research, The University of
Oklahoma, Norman, OK, pp. 85-114.
Hronsky, J.J.F. and Houghton, K.A. (2001), “The meaning of a de?ned accounting concept:
regulatory changes and the effect on auditors’ decision making”, Accounting,
Organizations and Society, Vol. 26 No. 2, pp. 123-39.
The meaning
of cash
45
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Hsu, S.H., Chuang, M.C. andChang, C.C. (2000), “Asemantic differential studyof designers’ andusers’
product form perception”, International Journal of Industrial Ergonomics, Vol. 25, pp. 375-91.
Humphrey, C.G., Moizer, P. and Turley, W.S. (1993), “The audit expectation gap in Britain:
an empirical investigation”, Accounting & Business Research, Vol. 23, pp. 395-411.
Jones, S. and Ratnatunga, J. (1997), “The decision usefulness of cash-?ow statements by
Australian reporting entities: some further evidence”, The British Accounting Review,
Vol. 29 No. 1, pp. 67-85.
Jones, S., Romano, C.A. and Smynios, K.S. (1995), “An evaluation of the decision usefulness of
cash ?ow statements by Australian reporting entities”, Accounting & Business Research,
Vol. 25 No. 29, pp. 115-29.
Jones, S., Sharma, R. and Mock, K. (1998), “Managerial evaluations of the relevance of cash vs
accrual-based ?nancial reports in the Australian food industry”, Australian Accounting
Review, Vol. 8, pp. 51-9.
Lawson, G.H. (1981), “A new approach to cash ?ow analysis”, The Accountant, Vol. 10 No. 17,
pp. 13-20.
Lee, T.A. (1984), Cash Flow Accounting, Van Nostrand Reinhold, London.
Lee, T.A. (1992), “Making cash ?ow statements useful”, Accountancy, April, p. 35.
Libby, R. (1979), “Bankers’ and auditors’ perceptions of the message communicated by the audit
report”, Journal of Accounting Research, Vol. 17, pp. 99-122.
McEnroe, J. and Martens, S. (2001), “Auditors’ and investors’ perceptions of the ‘expectation
gap’”, Accounting Horizons, Vol. 15, pp. 345-58.
Mason, A.K. and Gibbins, M. (1991), “Judgement and US accounting standards”,
Accounting Horizons, Vol. 5, pp. 14-24.
Mason, P. (1961), Cash Flow Analysis and the Funds Statements, Accounting Research Study
No. 2, AICPA, New York, NY.
Monroe, G.S. and Woodliff, D. (1993), “The effect of education on the audit expectation gap”,
Accounting and Finance, Vol. 33, pp. 61-78.
Mulford, C.W. and Comiskey, E.E. (2002), The Financial Numbers Game, Detecting Creating
Accounting Practices, Wiley, New York, NY.
Mulford, C.W and Comiskey, E.E (2005), Creative Cash Flow Reporting: Uncovering Sustainable
Financial Performance, Wiley, Hoboken, NJ.
Nekolaichuk, C.L., Jevne, R.F. and Maguire, T.O. (1999), “Structuring the meaning of hope in
health and illness”, Social Science & Medicine, Vol. 48, pp. 591-605.
Neubert, W.H. (1959), “Money ?ow and the ?rm”, The Accounting Review, Vol. 34, pp. 84-90.
Noland, V.J., Daley, E.M., Drolet, J.C., Fetro, J.V., McCormack-Brown, K.R., Hassell, C.D. and
McDermott, R.J. (2004), “Connotative interpretations of sexuality-related terms”, Sex Roles,
Vol. 51, pp. 523-34.
Nunnally, J.C. (1978), Psychometric Theory, 2nd ed., McGraw-Hill, New York, NY.
Oliver, B. (1974), “The semantic differential: a device for measuring the interprofessional
communication of selected accounting concepts”, Journal of Accounting Research, Vol. 12
No. 2, pp. 299-319.
Osgood, C.E. (1976), Focus on Meaning, Mouton, The Hague.
Osgood, C.E. and Suci, G.J. (1955), “Factor analysis of meaning”, Journal of Experimental
Psychology, Vol. 50, pp. 325-38.
ARJ
24,1
46
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Osgood, C.E., Suci, G.J. and Tannenbaum, P.H. (1957), The Measurement of Meaning, University
of Illinois Press, Urbana, IL.
Pickard, G. (2007), “Simplifying global accounting”, Journal of Accountancy, July, pp. 36-9.
Pierce, W.D., Sydie, T.A. and Stratkotter, R. (2003), “Social concepts and judgments: a semantic
differential analysis of the concepts feminist, man, and woman”, Psychology of Women
Quarterly, Vol. 27, pp. 338-46.
Porter, B. (1993), “An empirical study of the audit expectation-performance gap”, Accounting &
Business Research, Vol. 23, pp. 49-68.
Shah, A.K. (1996), “Creative compliance in ?nancial reporting”, Accounting, Organizations and
Society, Vol. 21 No. 1, pp. 23-39.
Sharma, D.S. and Iselin, E.R. (2003), “The decisions usefulness of reported cash ?ow and accrual
information in a behavioural ?eld experiment”, Accounting & Business Research, Vol. 33
No. 2, pp. 123-35.
Solomon, D. (2002), “Adelphia overstated cash ?ow, revenue over past two years”,
Wall Street Journal, June, p. B5.
Tergesen, A. (2002), “Cash-?ow hocus-pocus; this seemingly simple number can be anything but
straightforward”, Business Week, No. 3791, p. 130.
Vent, G.A., Cowling, J.F. and Sevalstad, S. (1995), “Cash ?ow comparability: accounting for
long-term debt under SFAS 95”, Accounting Horizons, Vol. 9 No. 4, p. 88.
Vicknair, D.B. (1983), “An investigation of users’ perceptions concerning corporate funds ?ow
information”, DBA dissertation, University of Tennessee, Knoxville, TN.
Wilson, A. (1989), “If it isn’t useful, get rid of it”, Accountancy, Vol. 104 No. 1151, p. 29.
Wines, G. (2006), “The connotative meaning of independence in alternative audit contexts:
an exploratory study”, Paci?c Accounting Review, Vol. 18, pp. 90-122.
Xiong, M.J., Logan, G.D. and Franks, J.J. (2006), “Testing the semantic differential as a model of task
processes with the implicit association test”, Memory &Cognition, Vol. 34 No. 7, pp. 1452-63.
Yap, C. (1996), “Cash ?ow statements: how useful?”, The Australian Accountant, Vol. 66 No. 11,
pp. 36-7.
Yoon, J. (2008), “Searching for an image conveying connotative meanings: an exploratory
cross-cultural study”, Library & Information Science Research, Vol. 30, pp. 312-18.
Further reading
Osgood, C.E. (1964), “Semantic differential technique in the comparative study of cultures”,
American Anthropologist, New Series, Vol. 66, pp. 171-200.
(The Appendices follow overleaf.)
The meaning
of cash
47
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Appendix 1
Figure A1.
Accounting concept: “CASH”– as it relates to the cash flow statement
Definition provided in FRS 10
"… coins, notes, demand deposits and other highly liquid investments in which an
entity invests as part of its day to day cash management." And "…borrowings from
financial institutions such as bank overdrafts, where such borrowings are at call and are
used as part of day-to-day cash management." (Para 4.1)
Based on the definition of CASH provided above, you believe cash to be:
Exact Estimated
Bad Good
Measurable
Unmeasurable
Necessary
Unnecessary
Planned
Unplanned
Objective
Subjective
Tangible
Intangible
Strong
Weak
Indirect
Direct
Variable
Constant
Safe
Risky
Complete
Discretionary
Required
Real
Imaginary
Beneficial
Adverse
Temporary
Permanent
Controllable
Uncontrollable
Unexpected
Expected
Passive
Active
Static
Dynamic
Long-term Short-term
Inflexible Flexible
ARJ
24,1
48
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Appendix 2
Corresponding author
Richard Fisher can be contacted at: [email protected]
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
Figure A2.
Accounting concept: “CASH”– as it relates to the cash flow statement
Definition provided in FRS 10
“… coins, notes, demand deposits and other highly liquid investments in which an
entity invests as part of its day to day cash management.” And “…borrowings from
financial institutions such as bank overdrafts, where such borrowings are at call and
are used as part of day-to-day cash management.” (Para 4.1)
Based on the above definition, what degree do you believe the following represent
items of CASH.
Yes
1. Coins and notes on deposit
2. Accounts receivable
3. A four month treasury bill
4. Reserve bank bill
5. Gold bullion
6. Readily tradable equity securities
7. A three month futures contract
8. Non-cash payments for goods and service
barter
9. On-call bank overdraft
10. Preference shares redeemable in three months
for a fixed amount of cash
No
The meaning
of cash
49
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_207362497.pdf
The purpose of this paper is to explore the impact on communication of changes in an
accounting standard arising from the transition to International Financial Reporting Standards.
It investigates inter and intragroup differences in measured connotative meaning of the old and new
definitions of “cash”, as held by three key groups of parties to the accounting communication process
(preparers, auditors and users); and determines the effect of changes in connotative meaning on
decision behaviour (outcomes).
Accounting Research Journal
The meaning of cash in the context of alternative accounting standards: IFRS
convergence and classification decisions
Tony Mortensen Richard Fisher
Article information:
To cite this document:
Tony Mortensen Richard Fisher, (2011),"The meaning of cash in the context of alternative accounting
standards", Accounting Research J ournal, Vol. 24 Iss 1 pp. 23 - 49
Permanent link to this document:
http://dx.doi.org/10.1108/10309611111148760
Downloaded on: 24 January 2016, At: 21:12 (PT)
References: this document contains references to 71 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 1363 times since 2011*
Users who downloaded this article also downloaded:
Salma Ibrahim, Li Xu, Genese Rogers, (2011),"Real and accrual-based earnings management and its legal
consequences: Evidence from seasoned equity offerings", Accounting Research J ournal, Vol. 24 Iss 1 pp.
50-78 http://dx.doi.org/10.1108/10309611111148779
Afshad J . Irani, Le (Emily) Xu, (2011),"Do stealth restatements convey material information?", Accounting
Research J ournal, Vol. 24 Iss 1 pp. 5-22 http://dx.doi.org/10.1108/10309611111148751
Nicholas Apergis, Christina Christou, Christis Hassapis, (2014),"Accounting standards convergence
dynamics: International evidence from club convergence and clustering", Accounting Research J ournal,
Vol. 27 Iss 3 pp. 226-248 http://dx.doi.org/10.1108/ARJ -06-2013-0031
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The meaning of cashinthe context
of alternative accounting
standards
IFRS convergence and classi?cation decisions
Tony Mortensen and Richard Fisher
University of Canterbury, Christchurch, New Zealand
Abstract
Purpose – The purpose of this paper is to explore the impact on communication of changes in an
accounting standard arising from the transition to International Financial Reporting Standards.
It investigates inter and intragroup differences in measured connotative meaning of the old and new
de?nitions of “cash”, as held by three key groups of parties to the accounting communication process
(preparers, auditors and users); and determines the effect of changes in connotative meaning on
decision behaviour (outcomes).
Design/methodology/approach – The study adopted a between-participants 2 £ 3 factorial design
whereby the ?rst factor re?ected the de?nition type: old vs new de?nition of the concept “cash”; while
the second re?ected three ?nancial reporting groups: preparers, auditors and users. The semantic
differential technique developed by Osgood, Suci and Tannenbaum was used to measure connotative
meaning.
Findings – The study ?nds that the three ?nancial reporting groups do not share the same meaning
of the concept “cash” and that the introduction of the new de?nition has changed the interpreted
connotative meaning for these three groups. A link between measured meaning and the decisions
made by the participants was also established.
Research limitations/implications – The explanatory power of the typical three (evaluative,
potency and activity) factor structure should be acknowledged; these factors typically explain
50 per cent of the total phenomena known as “meaning”. The study’s ?ndings make an important
contribution to the earnings management and creative accounting literature.
Practical implications – The ?ndings are particularly relevant to standard-setters and regulators
as a lack of shared meaning may lead to unnecessary misunderstandings and tensions among the
many parties to the reporting process.
Originality/value – The study extends prior measurement of meaning studies in accounting
through ?rst, the inclusion of all three major groups of parties to the accounting communication
process; second, examination of an accounting concept which is de?ned differently by two accounting
standards in the same jurisdiction; and last, investigation of the impact on decision behaviour
(outcomes) resulting from changes in meaning brought about through the introduction of a new
standard across the three groups.
Keywords International accounting standards, Cash ?ow statement, Connotative meaning,
Measurement of meaning, Semantic differential, Cash, Financial reporting
Paper type Research paper
1. Introduction
Effective communication between the various parties to the reporting process is the
cornerstone of ?nancial accounting (Shah, 1996). The importance of shared meaning,
and, in particular, shared connotative meaning[1], between preparers and the many
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
The meaning
of cash
23
Accounting Research Journal
Vol. 24 No. 1, 2011
pp. 23-49
qEmerald Group Publishing Limited
1030-9616
DOI 10.1108/10309611111148760
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
users of ?nancial reports in facilitating effective communication has been widely
acknowledged in the literature (Haried, 1972; Oliver, 1974; Flamholtz and Cook, 1978;
Adelberg and Farrelly, 1989; Bagranoff et al., 1994; Hronsky and Houghton, 2001; Wines,
2006). The provision of clear de?nitions by standard-setters contributes to the
communications process by providing a uniform basis from which all parties can
operate when dealing with the meaning associated with fundamental accounting terms
and concepts. This suggests that careful wording may signi?cantly reduce the potential
for miscommunication and “[. . .] reduce the justi?ability of aggressive reporting
decisions” (Hronsky and Houghton, 2001, p. 124), thus enhancing the reliability of
?nancial statements.
The global convergence to International Financial Reporting Standards (IFRS) has
seen extensive and unprecedented changes in many jurisdictions’ Generally Accepted
Accounting Practice (GAAP)[2], and poses a substantial challenge to the ?nancial
communications process as preparers, auditors and users confront unfamiliar
terminology and alternative de?nitions of existing concepts. This study explores the
complexity of this communication problemin the context of NewZealand’s transition to
IFRS. The study examines the connotative meaning of the concept “cash”, as it relates to
the cash ?ow statement, as interpreted by three key groups of parties to the accounting
communication process: preparers, auditors and users. Using an experimental design,
the study focuses on inter and intragroup changes in measured connotative meaning
resulting from the introduction of a new de?nition of cash, contained in New Zealand
Equivalent to International Accounting Standard (NZ IAS) 7, “cash ?ow statements”.
New Zealand’s old de?nition is found in FRS 10, “statement of cash ?ows” and has
remained in effect after the mandatory adoption date for NZ IFRS for many entities such
as small to medium-sized non-listed companies and not-for-pro?t entities. Such entities
are currently not subject to the full ?nancial reporting requirements of New Zealand’s
Financial Reporting Act 1993 (pending the outcome of a reviewof the ?nancial reporting
regime in New Zealand currently being undertaken by the Ministry of Economic
Development). The study also considers the effect that possible changes in meaning may
have on the decisions made by the different parties[3].
Many would regard cash as being among the most clear and least contentious
concepts in the domain of accounting (Mulford and Comiskey, 2005; Tergesen, 2002).
Despite this, the cash ?ow statement has recently been criticised for its susceptibility to
aggressive reporting practices (Mulford and Comiskey, 2002; Solomon, 2002; Tergesen,
2002; Broome, 2004). Clearly worded accounting standards are one means of reducing
the justi?ability of aggressive reporting decisions (Hronsky and Houghton, 2001).
However, in the New Zealand context, the respective de?nitions of cash contained in
FRS 10 and NZ IAS 7 use ambiguous and subjective terms, such as “highly liquid”,
“short term”, “at call”, “day-to-day cash management” and “insigni?cant risk”.
Somewhat ironically, the cash ?owstatement replaced the funds statement in the late
1980s, due in part to concerns about the lack of de?nitional clarity surrounding the term
“funds” and its consequential susceptibility to misunderstanding and potential
manipulation (Goldberg, 1951; Buzby and Falk, 1974; Henry, 1975; Heath, 1978; Clift,
1979; Chesley and Scheiner, 1982; Graci, 1982; Vicknair, 1983; Lawson, 1981; Lee, 1984;
Vent et al., 1995). If it is found in this study that FRS have failed to achieve a shared
understanding of a relatively straightforward concept, such as cash, then substantial
doubt must exist over standard-setters’ abilities to achieve shared meanings on more
ARJ
24,1
24
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
complex ?nancial terms and concepts among the key players in the accounting
communication process.
The study builds on the measurement of meaning in accounting literature that has
traditionally focused on the extent to which the meaning of various accounting concepts
are shared by different parties to the accounting communication process (e.g. among
preparers, users, or auditors). The study extends research in this area through ?rst, the
inclusion of all three major groups of parties to the accounting communication process;
second, examination of an accounting concept which is de?ned differently by two
accounting standards in the same jurisdiction; and last, investigation of the impact of
resulting changes in meaning brought about through the introduction of a newstandard
on decision behaviour (outcomes) across the three groups.
With New Zealand being an early adopter of IFRS[4], the results of this study have
far-reaching implications for standard-setters and regulators of countries adopting
IFRS. Further, the results provide an important contribution to the creative accounting
and earnings management literature, as the de?nitional clarity of accounting concepts is
believed to be an important component in reducing creative or aggressive reporting
practices.
With IFRS convergence, a subtle change in the de?nition of even a seemingly
straightforward item such as cash (for purposes of the cash ?ow statement) can result
in different interpretations (connotative meaning) for different parties to the ?nancial
reporting communication process, and, in turn, impact on accounting decisions. Future
research may build on the ?ndings of this study by investigating the issues raised in
the study in relation to other IFRS standards.
The next section provides a review of the literature. This is followed by the research
hypotheses and an outline of the research method employed in the study. The results
are then considered, followed by a discussion of the research ?ndings and conclusion.
2. Review of the literature
2.1 Cash and the cash ?ow statement
Since the introduction of the cash ?ow statement, its usefulness to decision makers has
received signi?cant attention (Epstein, 1992; Jones et al., 1995, 1998; Yap, 1996; Jones
and Ratnatunga, 1997; Sharma and Iselin, 2003). The literature highlights several
advantages of cash ?ow information over traditional information found in the balance
sheet and income statement (Neubert, 1959; Mason, 1961; Hicks and Hunt, 1980; Lee,
1984, 1992; Sharma and Iselin, 2003), with one of the most frequently cited being
increased reliability over traditional accrual-based accounts. Reasons for this
advantage include the establishment of objective criteria and de?nitional clarity.
The precursor to the cash ?ow statement was the funds statement, which was ?rst
seen in the USA in the early 1900s (Wilson, 1989). While originally welcomed by many
countries, it became apparent by the late 1970s that the funds statement was not going to
be a permanent addition to the annual report (or at least not in its existing form). One of
the key concerns echoed by professionals and academics alike surrounded the de?nition
of the term “fund” and its inconsistent application in practice (Goldberg, 1951; Buzby
and Falk, 1974; Henry, 1975; Heath, 1978; Clift, 1979; Lawson, 1981; Chesley and
Scheiner, 1982; Graci, 1982; Vicknair, 1983; Lee, 1984; Vent et al., 1995). Vent et al. (1995)
believed that the meaning of the word “fund” had never been clearly de?ned, and that
this lack of guidance had resulted in a wide diversity of practice in the preparation
The meaning
of cash
25
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of the statement. After many years of debate, national standard-setters began to
systematically replace the funds statement with the cash ?owstatement in the hope that
many of the issues raised regarding the problematic nature of the funds statement would
be resolved[5]. The concept of “fund” was effectively replaced by that of cash, which was
believed to be clearer and less ambiguous (Buzby and Falk, 1974).
The cash ?ow statement purports to present the inward and outward ?ows of cash.
Accordingly, a clear understanding of what constitutes cash is important within and
across all parties to the accounting communication process. New Zealand’s adoption of
IFRS saw it introduce NZ IAS 7. This standard (along with all other NZ IFRS) applies
to reporting entities, which include large issuers, subsidiaries of overseas companies
complying with IFRSs, and the public sector. Certain smaller entities are currently still
permitted to apply the old standard, FRS 10, which de?nes cash differently than
NZ IAS 7, raising the possibility that different parties to the communications process
may hold different meanings of the term.
2.2 De?nition of cash
Table I presents the de?nition of cash (as it relates to cash ?ow) under FRS 10 and the
more recent NZ IAS 7.
While the concept “cash” under FRS 10 includes currencies, as would be described in a
traditional sense (i.e. coins, notes and demand deposits), it also introduces the concept of
“highlyliquidinvestments” (Para. 4.1). However, this termis not separatelyde?nedinFRS
10. FRS 10’s de?nition also speci?callyencompasses certainforms of “at call borrowings”,
which effectively represent “negative” cash. In contrast, NZ IAS 7’s de?nition includes
both “cash on hand” and “demand deposits”, together with all other “cash equivalents”.
The latter is de?ned in terms of ease of conversion to known amounts of cash and
presenting an insigni?cant risk of changes in value. Whilst there is signi?cant overlap
between the two de?nitions, there are also clearly notable differences.
A further notable observation concerning the two de?nitions in Table I is the degree
to which many of the terms employed are inherently ambiguous, leading to potential
subjectivity in the interpretation and application of the de?nitions. For instance, phrases
such as “short term”, “highly liquid”, “at call”, “readily convertible”, “day-to-day cash
management” and “insigni?cant” are all open to alternative interpretation and little
guidance is provided in either standard.
To many users of ?nancial statements, cash ?ow is regarded as “real and not
subject to the vagaries of GAAP or the whims of the accountants” (xiii), while cash,
which underpins cash ?ow, is considered fact (Mulford and Comiskey, 2005).
FRS 10 – de?nition of “cash”
NZ IAS 7 – de?nition of “cash” and “cash
equivalents”
“[. . .] coins, notes, demand deposits and other
highly liquid investments in which an entity
invests as part of its day-to-day cash
management.” And “[. . .] borrowings from
?nancial institutions such as bank overdrafts,
where such borrowings are at call and are used as
part of day-to-day cash management” (Para. 4.1)
“[. . .] cash on hand and demand deposits” (Para. 6)
“cash equivalents” is de?ned as:
“[. . .] short term, highly liquid investments that
are readily convertible to known amounts of cash
and which are subject to an insigni?cant risk of
change in value” (Para. 6)
Table I.
De?nitions of cash
ARJ
24,1
26
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
However, as noted above, with the introduction of NZ IAS 7, there now exist two
de?nitions of cash in New Zealand, neither of which exactly overlap and both requiring
judgment in their application. The introduction of NZ IAS 7 has created a need to
explore the meanings assigned to the concept of cash within and between the differing
groups of parties to the ?nancial reporting communications process. Prior research
suggests that differences in meanings attributed to concepts de?ned in accounting
standards may increase the possibility for miscommunication, and have ?ow on
implications for decision making, including increasing the risk of aggressive reporting
practices (Hronsky and Houghton, 2001). Exploration of this issue is also consistent
with Mason and Gibbins’ (1991) call for researchers to examine new and existing
standards for possible ambiguities that may detract from the overall thrust of
particular accounting standards.
2.3 Measurement of meaning
The connotative meaning of cash in this study is measured using a well-established
method of measuring meaning, the semantic differential technique. The technique was
developed by Osgood and Suci (1955), and later re?ned by Osgood et al. (1957).
Osgood et al. narrowly de?ned meaning as that which explains the way in which
differences occur between the sender and receiver of information. While the
quanti?cation of meaning will always be controversial, Osgood et al. argued that
connotative meaning is responsible for in?uencing “individual behaviour” (p. 321).
The semantic differential technique is essentially a combination of controlled
association and scaling procedures which are used to identify and measure
a participant’s connotative meaning of a concept. Coupled with the use of factor
analysis, the process allows for the identi?cation of a participant’s connotative meaning
of a speci?c term or concept by locating their perceived meaning as a point in a
corresponding “semantic space” (Osgood et al., 1957, p. 25)[6].
Osgood et al. (1957) found that, for most everyday concepts in the general domain of
meaning, connotative meaning could be explained within a three factor “cognitive”
structure. These factors were identi?ed in their study and labelled as: evaluative,
represented by scale items such as good-bad, important-unimportant; potency,
represented by scale items such as hard-soft, strong-weak; and activity, made up of scale
items such as active-passive, fast-slow. The evaluative, potency and activity (EPA)
structure has been validated in a large number of subsequent studies across a wide
range of disciplines[7]. As these factors represent the axes within the semantic space
where meaning is believed to exist, a participant’s meaning of a speci?c termor concept
can be identi?ed within that semantic space.
2.4 Measurement of meaning in accounting
Haried (1972) conducted the earliest research in the area of measured meaning in
accounting. He undertook a study into what he called the “[. . .] semantic problems in
accounting communication [. . .]” (p. 376), looking at the proximity (or differences)
in meaning between that intended by the sender (e.g. the meaning of the term
“depreciation” as determined by the accountant) and that interpreted by the receivers
(e.g. the meaning of the term “depreciation” as determined by a banker or investor).
Attempting to validate Osgood et al.’s (1957) technique in the accounting domain,
Haried (1973) used a sample of concepts commonly found in ?nancial statements
The meaning
of cash
27
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
(e.g. goodwill, net income, accounts receivable) across both sophisticated
(e.g. accountants) and unsophisticated (e.g. investment club members) research
participants. This study revealed a seven, factor structure. However, subsequent
reanalysis of Haried’s (1973) data by Houghton (1988, p. 279) con?rmed that the “[. . .]
dimensions of space are similar to those ‘standard’ EPA factors that were originally
identi?ed by Osgood et al. (1957)”. Houghton’s (1988) reanalysis laid the foundations for
further analysis in the domain of accounting meaning, leading to a number of studies
examining a variety of terms, concepts and phrases used in accounting (Oliver, 1974;
Flamholtz and Cook, 1978; Houghton, 1987a, 1988; Houghton and Messier, 1990;
Hronsky and Houghton, 2001; Wines, 2006), using different groups of participants
(Houghton, 1987b; Houghton and Messier, 1990; Houghton and Hronsky, 1993), and
across different cultures (Bagranoff et al., 1994).
It has also been established that the level of sophistication of participants tends to
affect the degree of shared cognitive structure in which meaning is believed to be held
(Oliver, 1974; Houghton, 1987a, 1988; Houghton and Messier, 1990; Houghton and
Hronsky, 1993; Hronsky and Houghton, 2001; Wines, 2006). This is important as the
absence of a shared cognitive structure prevents further analysis of measured meaning.
That is, a shared structure is a necessary but not suf?cient condition to con?rm the
existence of shared connotative meaning.
Recent studies by Hronsky and Houghton (2001) and Wines (2006) have extended the
boundaries of measurement of meaning research by investigating the association
between measured meaning and decision outcomes. Hronsky and Houghton (2001)
found that practicing auditors held the same cognitive structure within which the
meaning of “extraordinary items” was held, and that their classi?cation decisions were
systematically associated with differences in the connotative meanings of de?nitions of
“extraordinary items” under new and old FRS. Wines (2006), on the other hand,
examined the concept of “auditor’s independence” using a sample of 89 undergraduate
students. He found that participants considered the concept within a two factor structure
comprising “emphasis” and “variability” dimensions, and that their perceptions of
independence in three case scenarios were systematically associated with the identi?ed
connotative meaning dimensions.
The current study extends the measurement of meaning literature in several ways.
First, in contrast to prior studies, it investigates the connotative meaning of a concept
where meaning is widely regarded as being well understood and relatively
uncontroversial (Mulford and Comiskey, 2005; Tergesen, 2002). Second, it considers
the degree of shared meaning between three important groups of parties to the ?nancial
communications process: preparers, users and auditors. Last, it investigates the
relationships between the decision behaviour (outcomes) of all three groups and their
respective connotations of the concept of cash.
3. Research question and hypotheses
The overall objectives of this study are to determine, ?rst, whether the three main
groups of parties to the accounting communication process share a similar meaning of
the concept of cash, as it relates to the cash ?ow statement; second, whether a change in
the regulated de?nition of cash can in?uence measured meaning and decision
outcomes; and third, whether there is empirical evidence to link the measured meaning
ARJ
24,1
28
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of this concept with the decisions made by each party. Five main hypotheses provide
the basis for addressing these research objectives.
3.1 Hypotheses
Studies in the area of measurement of (connotative) meaning in accounting have
con?rmed the importance of establishing the degree of shared cognitive structure
between participants in order to subsequently determine and measure the meaning
associated witha given concept (Houghton, 1987a, 1988). Anumber of these studies have
provided evidence to suggest that preparers, auditors and speci?c users (e.g. bankers
and shareholders) do not share the same cognitive structure in which the meaning of
certain accounting concepts are held, due in part to differences in sophistication,
incentives and degree of homogeneity (Houghton, 1987b; Hronsky and Houghton, 2001).
Therefore, the following null hypothesis is proposed for the current study:
H
01
. There is no signi?cant difference in the cognitive (factor) structures within
which the meaning of “cash” is held between each of the three ?nancial
reporting groups.
As previously discussed, shared cognitive structure is a necessary but not suf?cient
condition for establishing shared connotative meaning (Houghton, 1988). As clari?ed by
Houghton and Messier (1990), there are two component parts to shared meaning in a
communications process. First, parties must share the same “cognitive structure”
(e.g. perhaps all sharing a three-dimensional factor structure conforming to the classic
EPA structure), and second, all must have similar factor scores on each dimension of
their respective cognitive structures (e.g. all parties view cash as being equally positive
on the evaluative dimension, all view cash as being equally dynamic on the activity
dimension, etc.). For this reason, a second hypothesis is required to provide a basis for
determining possible differences in the measured meaning between the three participant
groups of the old and new de?nitions of cash, respectively:
H
02
. There is no signi?cant difference in the measured meaning of the concept
“cash”, as de?ned in either FRS 10 or NZ IAS 7, between each of the three
?nancial reporting groups.
If pairs of reporting groups fail to exhibit a shared cognitive structure (as established
under H
01
), H
02
will not be able to be tested for those pairs. In this case, the connotative
meaning of cash will be assumed not to be shared and the analysis will proceed directly
to H
03
for those reporting group pairs.
Hronsky and Houghton (2001) found a signi?cant difference among auditors in the
measured meanings of old and new de?nitions of “extraordinary items”, respectively,
under Australian Accounting Standards. Earlier, Houghton and Messier (1990)
con?rmed the existence of signi?cant differences among auditors and bankers
(and between these two groups) in their interpretation of the meaning of differently
worded audit reports and different types of audit report opinions. These results suggest
that individual reporting groups may exhibit signi?cant intragroup differences in the
measured meanings of alternative de?nitions of speci?c accounting concepts.
The recent change in the de?nition of the concept of cash, brought about through
New Zealand’s transition to IFRS, might be expected to result in such intragroup
differences. Therefore, the third hypothesis is as follows:
The meaning
of cash
29
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
H
03
. There is no signi?cant difference in the measured meaning of the concept
“cash”, as de?ned in either FRS 10 or NZ IAS 7, within each of the three
?nancial reporting groups.
Hronsky and Houghton (2001) found evidence of a relationship between the decision
outcomes of each participant and the meaning they attributed to relevant accounting
concepts. In their study, a change in the de?nition of the term“extraordinary items” was
seen to be associated with changes in auditors’ perceived meaning of this term. In turn,
this change in meaning was found to lead to different classi?cation decisions across a
number of cases.
In the context of the current study, it is expected that participants will make
different decisions regarding the classi?cation of potential cash items (the degree to
which an item is or is not an item of cash for the purpose of the cash ?ow statement),
depending on which of the two de?nitions (FRS 10 and NZ IAS 7) they are exposed to.
Further, it is expected that any change in connotative meaning brought about by the
change in the de?nition of cash will be associated with a change in the decision
outcomes for each of the ?nancial reporting groups. This gives rise to the fourth and
?fth hypotheses, respectively:
H
04
. There is no signi?cant difference in the decision outcomes made by each of
the three ?nancial reporting groups on the basis of the old FRS 10 de?nition of
“cash” and the new NZ IAS 7 de?nition.
H
05
. The variability in the decision outcomes within each of the three ?nancial
reporting groups is not explained by the variability in the meaning of the
concept upon which the decisions are based.
4. Research method
This study consists of a between-participants 2 £ 3 factorial design in which the ?rst
factor re?ects the old and new de?nition of the concept of cash while the second factor
re?ects the three ?nancial reporting groups: preparers, auditors and users.
4.1 Participants
The three ?nancial reporting groups utilised in the current study are seen to play
a speci?c and distinct role in the accounting communication process (Bedford and
Baladouni, 1962). While prior measure of meaning studies have used each of these
?nancial reporting groups either individually or in pairs (Oliver, 1974; Flamholtz and
Cook, 1978; Houghton, 1987a; Houghton and Hronsky, 1993; Hronsky and Houghton,
2001), none have combined all three in conjunction with two de?nitions of the same
concept.
To help ensure consistency across the range of preparer participants, the “preparer”
group has been limited to those who hold a current chartered accounting (CA)
quali?cation, or international equivalent, and are currently working in an accounting
decision-making role[8]. The members’ list from The New Zealand Institute of
Chartered Accountants (NZICA) provided the source listing for this participant group.
The second group of research participants are “experienced auditors”, de?ned as
those having had no less than three years’ audit experience. Hronsky and Houghton
(2001) argued that third year auditor have suf?cient experience to make an initial
judgment on issues of technical meaning and even though they do not prepare the
ARJ
24,1
30
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
?nancial accounts, they have suf?cient in?uence over the ?nal presentation to
contribute to the accounting communication process. The auditor participant group
was randomly drawn from three of the Big 4 CA ?rms’ audit departments, operating in
the three major centres in New Zealand.
While it is agreed that there are many different classes of ?nancial statement users,
it is commonly accepted that investors represent a company’s principal consumer of
published ?nancial reports (Barth et al., 2001). The current study sampled all current
members of the Certi?ed Financial Analysts’ Society of New Zealand (CFA) as
representatives of the user participant group. Owing to the quali?cation requirements
of the CFA the results of this study should indicate a more consistent or stable
cognitive structure within the user group than if a more heterogeneous subpopulation
of investors had been sampled[9]. Given their quali?cation requirements, this group
may be considered “experienced” users.
As in prior measurement of meaning studies, the data analysis tool used in this study
is factor analysis. In order to establish the stability of the emerging factor pattern,
a researcher must ensure that the sample size is adequate to achieve statistical
signi?cance (DeVellis, 2003; Hair et al., 1998). For the current study, a sample size of
seven participants per semantic scale item was deemed adequate, resulting in a target
sample size of 154 participants (seven participants £ 22-scale items). Based on an
estimated 35 per cent response rate, a total of 446 participants were contacted by e-mail
requesting participation in the study. The response rate was acceptable, at 46 per cent
for preparers, 44 per cent for auditors and 30 per cent for users, providing an overall
response rate of 39 per cent. The response rate for each subgroup is represented in
Table II.
4.2 Research instrument and administration
E-mail and the internet were used to request, administer and record the information
pertaining to the research instrument[10]. Participants were randomly assigned to the old
and new de?nition of cash based on participant lists (sample frames) obtained from
NZICA, the three CA?rms andthe CFA. The “between-participants” design helpedensure
that each participant’s responses were independent of the other responses, and that their
responses were not confounded by such factors as practice or experience gained in other
treatments or fatigue or boredom from participating in a series of treatments.
The web-based research instrument was made up of four key tasks. Task (1)
required each participant to consider the concept of cash, as de?ned by their allocated
de?nition (either FRS 10 or NZ IAS 7), then assess its meaning according to the
22 seven-point semantic differential scales provided (Appendix 1)[11]. To address the
decision-making implications of the allocated de?nition, two further tasks (tasks (2)
and (3)) were presented to each participant. Task (2) presented a series of ten speci?c
Group
Total e-mails
sent
FRS 10
de?nition
NZ IAS 7
de?nition
Total
responses
Total response rate
(%)
Preparers 124 27 30 57 46
Auditors 156 35 33 68 44
Users 166 23 27 50 30
Total 446 85 90 175 120
Table II.
Response rates
The meaning
of cash
31
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
assets (referred to as cases for the purpose of this study). For each case, participants
were instructed to indicate whether or not they believed the case to represent “clearly
an item of cash” when applying their respective de?nition of the concept of cash.
As recommended by Hronsky and Houghton (2001), task (3) required each participant
to indicate the degree to which they believed each case noted in task (2) represented an
item of cash (as it relates to the cash ?ow statement), which helped determine the
degree of con?dence each participant had in each decision outcome. For this purpose,
participants responded on a six-point Likert scale (with “1” representing “clearly an
item of cash” and “6” representing “clearly not an item of cash”) (Appendix 2 for
examples of tasks (1) and (3)).
5. Results
The data analysis stage is divided into four parts. The ?rst part involves factor analysis
of the measurement of connotative meaning data obtained fromthe semantic differential
scale responses. The aim of this analysis is to establish and label the underlying factor
structure (cognitive structure) for each ?nancial reporting group. This helps establish
whether intra and intergroup comparisons can be made. Inso doing, the analysis enables
testing of H
01
and helps determine the feasibility of testing H
02
. The second part of the
analysis focuses on changes in factor placements along the dimensions making up each
reporting group’s factor structure in order to ascertain whether changes in measured
meaning may be attributed to the change in de?nition of cash (i.e. testing of H
03
). The
third part focuses on the analysis of the decision outcomes (the indication of whether or
not the asset presented in each case represented an itemof cash and the degree to which
the participant believes the assets listed in the instrument represent items of cash) and
whether these have been impacted by the change in regulated de?nition (i.e. testing
of H
04
). The ?nal part deals with the connection between the measured meanings and
decision outcome(s) (i.e. testing of H
05
). It is this analysis that provides the most vital
evidence of the relationship between the de?nition of an accounting concept and the
decisions made by participants resulting from their meaning.
5.1 Factor analysis
The determination of the number and nature of factors held by each ?nancial reporting
group is fundamental in determining the degree of shared cognitive structure (Houghton,
1988). Inorder to ?rst determine the number of factors withinwhich the meaning of cash is
held by each of the three ?nancial reporting groups, each group’s data are independently
factor analysed (with varimax rotation). This procedure reduces the responses relating to
the 22 semantic differential scale items to a smaller number of factors, and is consistent
with the approach followed in the measure of meaning literature (Osgood et al., 1957;
Houghton, 1987a, b, 1988; Houghton and Hronsky, 1993; Houghton and Messier, 1990;
Bagranoff et al., 1994; Hronsky and Houghton, 2001; Wines, 2006).
Initially selecting factors with eigenvalues greater than 1.0 resulted in a ?ve-factor
structure (solution) for both the preparer and user groups, and a six-factor structure for
the auditor group. However, after employing a standard scree test (Cattell, 1966) and
then the more rigorous factor comparability test recommended by both Nunnally
(1978) and Everett and Entrekin (1980) (and subsequently introduced to the
measurement of meaning literature by Houghton (1987a, b, 1988))[12], a stable two
factor solution was obtained for both preparers and auditors, while a (less complex)
ARJ
24,1
32
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
single-factor structure resulted for users. Table III presents a summary of the results of
the factor comparability test for each of the three ?nancial reporting groups.
The results suggest that the structure within which the meaning of the concept of
cash is held for users is relatively simple (as re?ected by a single factor cognitive
structure), and differs to that held by both preparers and auditors, who hold a more
complex two factor structure. While it is clear, then, that the user group is not
comparable with the other ?nancial reporting groups, further investigation into the
preparers’ and auditors’ factors is required to determine whether their respective two
factor structures are indeed comparable. That is, although two factors appear to
underlie both groups’ cognitive structures with respect to the concept of cash, it is not
clear that the two factors are comparable between the groups. Non-comparability could
arise due to each of the factors for each group comprising a different set of highly
loading semantic scales. To explore this issue, further insight into the relationship
between the respective factors is required. Consequently, a factor comparison of each
reporting group’s factors is reported in Table IV.
The results suggest that the ?rst factor for all three ?nancial reporting groups is of
a similar nature, with the simple average comparability coef?cient for these factors
being 0.92, and Cronbach’s a being 0.97. While the correlation for factor 1 between
auditors and preparers does not speci?cally reach the cutoff suggested by Everett and
Entrekin (1980) of 0.894, the correlation of 0.876 is very close and indicates a high level
of shared variance of 76.7 per cent. In contrast, preparers’ and auditors’ second factors
appear unrelated, with a comparability coef?cient not signi?cantly different from zero.
Correlations between factor scores of random split
halves
a
Group Factor 1 Factor 2
Preparers
No. of factors ¼ 2 0.918 0.923
Auditors
No. of factors ¼ 2 0.941 0.976
Users
No. of factors ¼ 1 0.996
Note:
a
Correlation coef?cients of at least 0.894 indicate a comparable and robust factor structure
(Everett and Entrekin, 1980)
Table III.
Within group factor
comparability
Factor Groups Correlation
Panel A
Factor 1 Preparers-auditors 0.876
*
Preparers-users 0.986
*
Auditors-users 0.903
*
Panel B
Factor 2 Preparers-auditors 0.049
Note: Signi?cance at:
*
0.01 level (two-tailed)
Table IV.
Between group factor
comparability
The meaning
of cash
33
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
In summary, the assumption of a shared cognitive structure between any pair of
?nancial reporting group is not justi?ed.
As discussed previously, determining whether different groups share connotative
meaning is a two-step process. First one determines whether comparable factor
(cognitive) structures exist between groups (H
01
), and then, and only then, can one test
for shared measured meaning between groups (H
02
). As the results suggest a lack of
shared cognitive structure between any pair of ?nancial reporting group, H
01
is
rejected and the study proceeds on the basis that all three ?nancial reporting groups
have different cognitive structures and therefore can be viewed as three separate factor
groups. Consequently, the testing of H
02
is neither possible nor necessary. However,
the testing of within group differences in measured meaning of the concept of cash, as
it relates to the cash ?ow statement, may proceed as planned (i.e. testing of H
03
).
5.2 Factor labelling
Having established the number of factors underpinning each group’s cognitive
structure, attention may turn to understanding the nature of these factors.
This is achieved through the process of factor labelling and requires examination of
the factor loadings of each semantic differential scale item. Consistent with the
measurement of meaning literature in accounting (Houghton, 1988; Hronsky and
Houghton, 2001; Wines, 2006), those scale items which load heavily are deemed to be
those with factor loadings greater than 0.5. The rotated factor matrix (using varimax
rotation) for the three ?nancial reporting groups is presented in Table V.
Given that all three ?nancial reporting groups in this study have fewer than three
factors underlying their respective cognitive structures, it is apparent that their
structures may depart somewhat from the classic EPA structure. Osgood et al. (1957)
acknowledged that variations to this structure may occur depending on the nature of
the respondents and concepts under examination. Similarly, Osgood (1976, p. 37)
suggested that the greater the emotional or attitudinal loading of the concept being
judged, the greater the tendency of the semantic framework to collapse to fewer
dimensions. Houghton (1987a, 1988) also ?nds the applicability of the EPA structure in
accounting only under certain circumstances (Houghton and Messier, 1990).
Factor 1 for all three ?nancial reporting groups (shown earlier to be comparable
between groups) is consistently associated with scale items associated with potency
and evaluative dimensions. Some of the highest loading items are of a potency nature
(e.g. weak-strong, intangible-tangible and estimated-exact). Evaluative scales loading
highly include items such as subjective-objective, unexpected-expected and
discretionary-required. With respect to the concept of cash, it is perhaps to be
expected that potency and evaluative scales load onto a single factor. For cash, the
measurable, exact, strong and tangible potency scales are similar in effect to the safe,
objective and expected evaluative scales. A recent study by Wines (2006) also ?nds a
dominant factor consisting of both potency and evaluative dimensions, which he labels
emphasis. Consistent with this nomenclature, this study adopts the emphasis label for
factor 1 for all ?nancial reporting groups. Table V reveals that the emphasis factor
dominates factor 2 for both preparers and auditors, as re?ected by its comparatively
high eigenvalues of 10.145 and 8.429 for preparers and auditors, respectively.
Relative to factor 1, the labelling of factor 2 for the preparer group is more
straightforward. This factor is dominated by scales associated with the activity
ARJ
24,1
34
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
dimension of the EPA structure, such as constant-variable and dynamic-static.
Accordingly, this factor is labelled activity.
As shown previously in Table IV, the auditors’ factor 2 is not comparable with the
preparers’ factor 2. Factor 2 for auditors consists of scales relating to all three
dimensions of the EPA structure. In particular, it has several activity and evaluative
scales loading onto it which are of a temporal nature (e.g. temporary-permanent, long
term-short term and expected-unexpected). Further, it contains the potency scale items
of direct-indirect and imaginary-real, both of which have previously been included in a
sub-set of the potency dimension referred to as substantiveness in several prior studies
(Houghton and Hronsky, 1993; Houghton and Messier, 1990; Bagranoff et al., 1994). If
cash is considered to be permanent and long term, then one might also expect it to be
considered real and direct. Consequently, the temporal and substantiveness aspects of
factor 2 appear to be logically related. Wines (2006) encountered a similar factor
in his study, which he labelled variability. Since six out of the eight heavily loading
factor 2 scales correspond to scales heavily loading on to Wines’ variability factor, it is
considered appropriate to label auditors’ factor 2 as variability.
Preparer Auditor User
factors
a
factors
a
factor
a
1 2 1 2 1
Scale
b
EPA Emphasis Activity Emphasis Variability Emphasis
Good-bad E 20.456 0.422
Unnecessary-necessary E 0.662 0.569
Subjective-objective E 0.754 0.798 0.887
Risky-safe E 0.736 0.696 0.783
Required-discretionary E 20.788 20.489 0.560 20.763
Adverse-bene?cial E 0.640 0.431 0.522
Uncontrollable-controllable E 0.705 0.488 0.462
Expected-unexpected E 20.780 0.681 20.761
Unmeasurable-measurable P 0.810 0.864 0.850
Estimated-exact P 0.846 0.866 0.788
Intangible-tangible P 0.870 0.720 0.861
Weak-strong P 0.932 0.846 0.894
Direct-indirect P 20.847 20.493 0.575 20.794
Incomplete-complete P 0.809 0.674 0.808
Imaginary-real P 0.833 0.508 20.504 0.850
Flexible-in?exible P 0.835 0.664 0.754
Unplanned-planned A 0.682 0.626
Constant-variable A 20.815 0.558 20.691
Permanent-temporary A 20.451 20.431 0.703 20.582
Active-passive A 20.486 0.619 20.572
Dynamic-static A 0.713 0.426
Long term-short term A 0.541
Initial eigenvalue 10.145 2.439 8.429 2.194 10.413
Percentage of variance 46.112 11.086 38.313 9.971 47.330
Cumulative percentage 46.112 57.198 38.313 48.284 47.330
Notes: Based on a two-factor solution for preparer and auditor groups, and a single factor solution for
users;
a
factor loadings ,0.40 are suppressed;
b
each bipolar adjectival scale ranged from “1” (left-hand
scale endpoint) to “7” (right-hand scale endpoint)
Table V.
Rotated factor matrix by
?nancial reporting group
The meaning
of cash
35
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
In summary, the identi?cation of a single factor structure (corresponding to
emphasis) for the user group indicates a simpler (unidimensional) structure than the
factor structure identi?ed by Osgood et al. (1957), while both the preparer and auditor
?nancial reporting groups conform to a more complex (dimensional) structure
(emphasis/activity for preparers and emphasis/variability for auditors).
5.3 Factor placement
Having established that the cognitive structure for each ?nancial reporting group is
both robust and stable, we now investigate the possible differences in meaning within
each structure for each reporting group. This will address H
03
. Given that the cognitive
structure of each group is different, the analysis will focus on each ?nancial reporting
group separately.
Mean placement (factor) scores indicate the position of the concept cash for each
group on the factor axes corresponding to their cognitive structures. Movements in
placements (along the factor axes) between the old and new de?nition provide evidence
of both the extent (degree) and direction of a change in meaning[13]. Table VI reports
the factor placements by ?nancial reporting group.
To determine whether there are any signi?cant differences in perceived connotative
meaning of the concept of cash between the two de?nition groups (FRS 10 vs NZ IAS 7)
for each reporting group, analysis of variance (ANOVA) was performed on the factor
scores by group (Table VII).
The ANOVA results reveal a signi?cant change in factor scores on the dominant
factor, emphasis, for all three ?nancial reporting groups in response to NZ IAS 7
(preparers: F ¼ 4.996, p , 0.005; auditors: F ¼ 23.363, p , 0.01; and users:
F ¼ 36.881, p , 0.01). For instance, for preparers, the factor placement for the
emphasis factor is 230 under the old de?nition but 27 under the new de?nition. The
signi?cance of F for this factor for preparers pursuant to the ANOVA indicates that
these two values are signi?cantly different. For all three groups, there has been a
positive shift along the emphasis factor axis, suggesting that the new de?nition is
perceived as being more potent and is evaluated more favourably. In other words,
“cash”, as it relates to cash ?ows under NZ IAS 7, is seen as stronger and more
tangible, direct, exact, real, complete and measurable, as well as being more required,
expected, objective, safe, controllable, necessary, bene?cial and planned,
Group Factor(s)
Preparers Emphasis Activity
FRS 10 (old) de?nition 230 222
NZ IAS 7 (new) de?nition 27 20
Auditors Emphasis Variability
FRS 10 (old) de?nition 249 5
NZ IAS 7 (new) de?nition 52 25
Users Emphasis
FRS 10 (old) de?nition 270
NZ IAS 7 (new) de?nition 60
Note: Consistent with prior measurement of meaning studies in accounting, factor placement scores
have been multiplied by 100 to aid interpretation
Table VI.
Factor placements by
?nancial reporting group
ARJ
24,1
36
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
relative to the de?nition found in FRS 10. The apparent changes, however, along the
activity factor axis for preparers (F ¼ 2.519, p ¼ 0.118), and the variability factor axis
for auditors (F ¼ 0.187, p ¼ 0.667), were not statistically signi?cant, perhaps re?ecting
the subordinate nature of these two factors relative to the emphasis factor.
Overall, the results of the ANOVA provide empirical evidence that a change in
connotative meaning has occurred between the old and new de?nition of the concept
cash for all three ?nancial reporting groups, leading to a rejection of the null H
03
.
5.4 Decision outcomes
In order to address H
04
, the analysis determines whether a change in the regulated
de?nition of cash is associated with changes in decision outcomes (relating to ten asset
classi?cation cases)[14]. Given the previously established differences in cognitive
structure between the three ?nancial reporting groups, each group’s results were
assessed independently. Table VIII presents a comparison of classi?cations prompted
by NZ IAS 7 and FRS 10 for each reporting group, together with the results of ANOVA.
Table VIII reveals that the effects of the treatment were consistently signi?cant for
cases 3, 9 and 10 across all ?nancial reporting groups. Both the “four-month treasury
bill” and “on-call bank overdraft” were rated as closer to not being items of cash, while
“preference shares redeemable in three months for a ?xed amount of cash” were seen
as being closer to an item of cash under NZ IAS 7 relative to FRS 10. Both NZ IAS 7 and
FRS 10 de?nitions allude to cash potentially including highly liquid investments,
however the former requires there to be minimal risk surrounding the realisable value
of the asset (i.e. investments included in cash must be convertible for known amounts
of cash and be subject to an insigni?cant risk of changes in value). Consequently, the
risk created by the relatively long maturity (four months) on the treasury bills most
likely contributed to it being viewed as closer to not being an item of cash for
participants prompted by the NZ IAS 7 de?nition relative to participants in the FRS
10 group. Conversely, the ?xed redeemable value nature of the preference shares may
have led some NZ IAS 7 participants to view them as having a limited risk, which
would likely contribute to this asset being rated closer to an item of cash than for the
same asset under FRS 10. The fact that FRS 10 makes direct mention of bank
overdrafts potentially being included in cash would have undoubtedly led these
participants to view such items as being closer to cash than participants viewing the
NZ IAS 7 de?nition. It is interesting to note that the extent of the treatment effects was
suf?cient enough to result in change in overall classi?cation of an item (either from
Group F
Preparers
Factor 1 – emphasis 4.996
*
Factor 2 – activity 2.519
Auditors
Factor 1 – emphasis 23.363
* *
Factor 2 – variability 0.187
Users
Factor 1 – emphasis 36.881
* *
Note: Signi?cance at:
*
0.05 and
* *
0.01 levels, respectively
Table VII.
ANOVA of factor
placements for old
and new de?nitions
The meaning
of cash
37
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
cash to not an item of cash or vice versa) for cases 3, 9 and 10 for the user group, and
case 10 for the auditor group.
Further, Table VIII reveals that the user and auditor groups made different
decisions under NZ IAS 7 relative to FRS 10 for case 6 (“readily tradable securities”).
These assets were more likely to be viewed as not being an itemof cash under NZ IAS 7,
presumably due to their inherent risk of changes in value. The last signi?cant
difference noted in Table VIII relates to auditors only. Case 5 (“gold bullion”) is seen as
being closer to not being an item of cash under the NZ IAS 7 treatment, again,
presumably due to its potential risk of change in value.
On the basis of the discussion above, it appears that by incorporating criteria
relating to the riskiness of certain assets (highly liquid investments), NZ IAS 7’s
de?nition is instrumental in clarifying classi?cation decisions of the three ?nancial
reporting groups, particularly for the auditor group. For highly liquid investments,
the new standard is seen as being “tougher” (that is, has more emphasis – is more
objective, less discretionary, more exact) by all groups. Given the evidence of
differences in classi?cation decisions between the two treatment groups (old vs new
standard) the null H
04
is rejected.
Mean decision score
a
Preparers Auditors Users
Case Item
Old
de?nition
New
de?nition
Old
de?nition
New
de?nition
Old
de?nition
b
New
de?nition
1 Coins and notes on
deposit 1.00 1.00 1.00 1.00 1.00 1.00
2 Accounts receivable 4.96 4.90 4.94 5.27 4.86 4.70
3 A four-month treasury
bill 4.26
c
5.07
* *
,d
3.83 5.79
* * *
,d,e
2.95
c
5.07
* * *
,e
4 Reserve bank bill 3.33 2.80 2.71 2.82 2.23 2.15
5 Gold bullion 4.52 4.33 3.89 4.76
*
4.05 4.41
6 Readily tradable equity
securities 4.48 4.37
d
4.63
f
5.24
*
,d
3.55
f
4.52
*
7 A three-month futures
contract
5.44 5.40 5.03 5.21 5.23 4.85
8 Non-cash payment for
goods and services
(barter) 5.30
g
5.63 5.80
g
5.73 5.73 5.81
9 On-call bank overdraft 1.44 3.27
* * *
1.31 2.45
* *
,e
1.45 4.33
* * *
,f
10 Preference shares
redeemable in three
months for a ?xed
amount of cash 4.89 4.07
*
,d,h
4.20 2.39
* * *
,d
4.00 2.63
* *
,h
Notes: Signi?cance of difference between old and new de?nitions at:
*
0.05,
* *
0.01 and
* * *
0.001
levels, respectively;
a
possible scores ranged from “1” (clearly an item of cash) to “6” (clearly not an item
of cash);
b
one user participant responded with “6” for all ten cases and, consequently, has been
excluded as an outlier from this and subsequent analyses;
c,f,g
signi?cant pairwise differences between
preparers and auditors, preparers and users, and auditors and users, respectively, who received the old
de?nition ( post hoc comparison using the Bonferroni test, p , 0.05);
d,e,h
signi?cant pairwise
differences between preparers and auditors, preparers and users, and auditors and users, respectively,
who received the new de?nition ( post hoc comparison using the Bonferroni test, p , 0.05)
Table VIII.
ANOVA of decision
scores by ?nancial
reporting group
ARJ
24,1
38
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
While the focus of this section has been on the effects of a change in de?nition on the
decision outcomes of each reporting group, it is interesting to note that Table VIII also
reveals differences between preparers, auditors and users for each de?nition. The old
de?nition was associated with differences for certain pairs of reporting groups on cases
3, 6 and 8, while the new de?nition revealed pairwise differences on cases 3, 6, 9 and 10.
The new de?nition has created less consistency among the key parties to the ?nancial
reporting process with respect to their reporting decisions.
5.5 Decision outcomes and measurement of meaning
In this part of the data analysis, multiple regression analysis is employed to
examine H
05
. This hypothesis examines whether the variability in the decision
outcomes within each of the three ?nancial reporting groups may be explained by the
variability in the meaning of the concept upon which the decisions are based.
For each ?nancial reporting group, classi?cation decisions are regressed on
factor scores (corresponding to emphasis and activity for preparers, emphasis and
variability for auditors, and emphasis for users). To control for the effects of differences
in the nature of case facts, nine dummy variables are included in each equation to
represent the ten cases[15]. Further, unlike Hronsky and Houghton (2001) who did
not control for the effects of old vs new de?nition, a dummy variable representing
de?nition type (old vs new) is also included in each regression. To determine whether
multicollinearity was an issue (particularly between de?nition type and the factors),
collinearity diagnostics were reviewed. These did not reveal any cause for concern with
collinearity (all variance in?ation factors fell below 1.90). Table IX presents the results
of the regression analysis.
Overall, the three models, as speci?ed, are signi?cant (F ¼ 48.796 ( p , 0.001),
F ¼ 80.340 ( p , 0.001) and F ¼ 46.292 ( p , 0.001)) and explain 50.2, 58.5 and
50.5 per cent of the variance in classi?cation decisions made by the preparer, auditor,
Estimate
Parameter Preparers Auditors Users
Constant 4.419
* * *
3.090
* * *
2.941
* * *
Factor 1 (emphasis) 20.036 20.147
*
20.074
Factor 2 (activity) 0.170
* *
Factor 2 (variability) 20.001
De?nition (old vs new) 0.070 0.482
* * *
0.512
* *
Case 1 23.456
* * *
22.324
* * *
22.224
* * *
Case 2 0.474 1.779
* * *
1.571
* * *
Case 3 0.228 1.456
* * *
0.898
* *
Case 4 21.404
* * *
20.559
*
20.980
* * *
Case 5 20.035 0.985
* * *
1.061
* * *
Case 6 20.035 1.603
* * *
0.878
* * *
Case 7 0.965
* * *
1.794
* * *
1.796
* * *
Case 8 1.018
* * *
2.441
* * *
2.551
* * *
Case 9 22.053
* * *
21.456
* * *
20.163
F 48.796
* * *
80.340
* * *
46.292
* * *
Adjusted R
2
0.502 0.584 0.505
Notes: Signi?cance at:
*
0.05,
* *
0.01 and
* * *
0.001 levels, respectively
Table IX.
Regression of
classi?cation decisions on
factors, de?nition and
cases by ?nancial
reporting group
The meaning
of cash
39
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
and user groups, respectively. It may be seen from a review of the coef?cients
corresponding to each factor that the relationship between measured meaning and
decision outcomes is driven by factor 2, “activity” (coef?cient ¼ 0.170, p , 0.01), for
the preparer group; and factor 1, “emphasis” (coef?cient ¼ -0.147, p , 0.05), for the
auditor group. The users’ emphasis factor was not signi?cantly related to decision
outcomes.
“Activity” appears to be the most in?uential dimension of meaning for preparers.
The positive association between this dimension of meaning and decision outcomes
suggests that low (high) placements on the “activity” axis are associated with low
(high) decision scores[16]. Accordingly, a ?nancial report preparer with a low “activity”
factor score (corresponding with a view that cash is dynamic, variable and ?exible),
would tend to rate assets they associate with such characteristics as closer to the
“clearly an item of cash” (i.e. lower) end of the decision scale, than a preparer with a
higher “activity” score.
The dimension of meaning most in?uential for auditors was “emphasis”. “Emphasis”
is negatively associated with auditors’ decision scores. Consequently, an auditor with
a high “emphasis” score (corresponding with a view that cash is measurable, exact,
strong, objective, tangible, etc.) would be expected to rate assets perceived to possess
such traits closer to the “clearly an item of cash” end of the scale than an auditor with a
lower “emphasis” score.
In summary, meaning was found to independently relate to classi?cation decisions
for two out of the three ?nancial reporting groups, providing support for the rejection
of H
05
. “Activity” was the dimension of measured meaning most likely to in?uence the
classi?cation decisions of preparers, while for auditors it was the “emphasis” factor.
In part, the nature of the de?nition (NZ IAS 7 vs FRS 10) and case facts were also seen
to directly in?uence decision outcomes.
6. Discussion and limitations
The results of the current study indicate that the three key groups of parties to the
?nancial reporting process do not share the same cognitive structure in which the
meaning of the concept of cash (as it relates to cash ?ow) is held, thus allowing H
01
to be
rejected. Users were found to have a simple (unidimensional) structure consisting of an
“emphasis” factor, whilst the structures within which meaning was held for preparers
and auditors was somewhat more complex. The preparers’ cognitive structure was
represented by a semantic space consisting of two factors: “emphasis” and “activity”.
The two axes making up auditors’ semantic space consisted of “emphasis” and
“variability”. The lack of shared cognitive structure between the three groups implies
that the meaning of the concept cash is also not shared between these groups, raising
concerns about the effectiveness of the communication between the key players in the
?nancial reporting process. The lack of shared understanding could, in extreme
circumstances, result in misallocations of capital, as decisions made by the users
are based on their own understanding of the information and not on any uni?ed basis.
The reliability of the cash ?ow statement could be hampered by this fact, raising
questions about its susceptibility to intentional and unintentional miscommunication.
The lack of shared meaning between auditors and users, in particular, highlights
a communication gap paralleling the well-documented “audit expectations gap” – the
gap between what the public expects of an audit, and what auditors themselves believe
ARJ
24,1
40
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
their role to encompass. Prior studies examining the latter gap have largely focussed
on determining the nature and extent of differences between the public’s and auditing
profession’s perceptions of the auditor’s role and duties (Libby, 1979; Humphrey et al.,
1993; Monroe and Woodliff, 1993; Porter, 1993; McEnroe and Martens, 2001). However,
few studies have considered the impact of differences in meanings ascribed to the audit
subject matter (or elements thereof). The ?ndings of this study suggest that this could
be an avenue worthy of further research.
While the results of the semantic differential technique in this study only partially
supported Osgood et al.’s (1957) three-factor structure (EPA), they were consistent with
the application of this technique by other accounting researchers in the measurement
of meaning literature. However, it is recommended that future research reconsider the
adequacy of the 22-item semantic scale in the contemporary accounting environment to
provide assurance that these scales remain suf?ciently sensitive to capture the subtle
nuances of current terminology in the domain of accounting. Indeed, both
Osgood et al. (1957) and Haried (1972) supported an ongoing review of the semantic
scales to ensure that they continued to meet the requirements of the terms and concepts
under examination.
Turning to the analysis of measured meaning for each of the three ?nancial
reporting groups (H
03
), the results suggest that each group’s cognitive structure was
sensitive enough to result in a measurable difference in connotative meaning brought
about by changes in the de?nition of cash (FRS 10 vs NZ IAS 7). The new de?nition
was associated with a consistently positive movement along the “emphasis” axis
(a dimension of meaning held by all three ?nancial reporting groups). However, no
signi?cant change occurred along the “activity” axis for preparers, whose cognitive
structure included this axis. Likewise, no signi?cant change occurred along auditors’
second axis, labelled “variability”.
The decision-outcome analysis relating to H
04
con?rmed that the newer standard’s
(NZ IAS 7) de?nition of cash was instrumental in clarifying classi?cation decisions
for all three ?nancial reporting groups, particularly for highly liquid investments.
The results highlight that even subtle changes in the wording of of?cial de?nitions of
accounting terms and concepts may potentially lead to different decisions being made
by preparers, auditors and users. This result should be considered by standard-setters
when looking at amending or introducing new de?nitions or accounting standards, to
ensure that any such changes in decision outcomes are consistent with those intended.
This issue is of particular signi?cance in jurisdictions where two or more ?nancial
reporting frameworks coexist (e.g. in New Zealand, Australia, China, the UK). The
study demonstrates that two standards containing similar but not identical de?nitions
could result in inconsistent ?nancial reporting decisions, thereby further reducing the
comparability of the resulting ?nancial reports.
While there are many in?uences on?nancial reportingjudgments, suchas incentives,
training, availability of relevant resources, this study ?nds that connotative meaning is
an important determinant. After controlling for the effects of case differences and
alternative de?nitions of cash, “activity” was found to be the dimension of measured
meaning most likely to in?uence the classi?cation decisions of preparers, while for
auditors it was the “emphasis” dimension. This result is consistent with Hronsky and
Houghton (2001) who found potency important for auditors in “extraordinary item”
classi?cation decisions.
The meaning
of cash
41
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The results of this study must be considered in light of several limitations. It is
important to acknowledge the explanatory power of the typical three (EPA) factor
structure. Osgood et al. (1957) explainthat these factors typically explain50 per cent of the
total phenomena knownas “meaning”, withthe remaining50per cent beingmade upof an
almost in?nite number of variables. This notwithstanding, the Osgoodmodel continues to
be used widely across a wide range of disciplines, including accounting, to study the
interpretation or process of constructing meaning (Hronsky and Houghton, 2001).
The use of an experimental design in this study may have implications for the
generalisability of the study’s ?ndings. While a particular strength of the study is the
relatively large sample size and the inclusion of representatives of the three main
groups of parties to the ?nancial process (preparers, auditors and users), it is
acknowledged that ?ndings of the study may not generalise to groups not represented
in the study’s samples, such as unsophisticated users and non-quali?ed accountants.
While every effort was made to ensure each of the ten cases were representative of
assets found in practice, they naturally do not include all possible assets, nor are they
included in proportion to the frequency they would likely appear in practice.
7. Conclusion
Focusing on three key groups of parties in the ?nancial reporting process (preparers,
auditors and users), this study examined the impact on connotative meaning and
decision behaviour of a change in the promulgated de?nition of “cash”, as it relates to
the cash ?ow statement, necessitated by convergence with IFRS in New Zealand.
Results suggest that the three ?nancial reporting groups do not hold the meaning of
cash within the same cognitive structure. This ?nding supports the proposition that
the receivers of accounting information may not be interpreting that information in the
manner intended by the sender. Given that accounting communication is reliant on
shared meaning, such miscommunication may be leading to unnecessary
misunderstandings and tensions among the many parties to the reporting process.
A further ?nding of this study was that the measured meaning of cash for each of the
three ?nancial reporting groups was sensitive to the alternative de?nition of cash adopted
by NZ IAS 7. Similarly, classi?cation decisions were found to be in?uenced by the
particular de?nition of cash assigned to the research participants. These results suggest
that subtle differences inregulatedde?nitions of accountingconcepts have the potential to
cause unanticipated changes in meaning and consequential changes in decision
outcomes – even for seemingly uncontroversial and uncontested concepts in accounting.
In ?nding a direct association between measured meaning and decision outcomes for
both auditor and preparer groups, this study extends the work of Hronsky and Houghton
(2001) and con?rms the importance of connotative meaning in the domain of accounting.
Notes
1. Osgood et al. (1957, p. 321) describe connotative meaning as the emotional or metaphorical
meaning attached to a speci?c symbol (word, phrase, etc.). In contrast, denotative meaning
refers to the literal meaning of a symbol and involves the communication of an objective
description about the object (Adelberg and Farrelly, 1989). Agreement on denotative meaning
will generally be more easily reached between individuals (subject to similarities in knowledge
of the given ?eld under investigation) than on connotative meaning (Haried, 1973). After
considering both forms of meaning, Adelberg and Farrelly (1989) conclude that differences in
connotative meaning cause the most concern in the accounting communication process.
ARJ
24,1
42
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2. The International Accounting Standards Board (IASB) Chair, Sir David Tweedie, predicts
that 150 countries will have adopted IFRS by 2012 (Pickard, 2007).
3. According to Osgood et al. (195, p. 1597), connotative meaning “[. . .] is an important
intervening variable in human behaviour” and its measurement provides a basis for making
predictions about overt behavioural responses.
4. All New Zealand reporting entities were required to apply IFRS in preparing external
?nancial reports for reporting periods commencing on or after 1 January 2007. However,
entities were permitted to adopt IFRS from 1 January 2005.
5. In 1985, Canada’s accounting regulatory body introduced the cash ?ow statement. By 1992,
authorities responsible for standard-setting in New Zealand, the USA, South Africa, the UK
and the International Accounting Standards Board’s had all followed Canada’s lead.
6. Osgood et al. (1957, p. 25) de?ne semantic space as: “[. . .] a region of some unknown
dimensionality and Euclidian in character. Each semantic scale [. . .] is assumed to represent
a straight line function that passes through the origin of this space, and a sample of such
scales then represents a multidimensional space.”
7. Recent applications of Osgood et al.’s (1957) technique may be found in ?elds such as
information science (Yoon, 2008), marketing (Hsu et al., 2000), medicine (Fitzgerald et al.,
2008; Nekolaichuk et al., 1999), psychology (Xiong et al., 2006) and sociology (Noland et al.,
2004; Pierce et al., 2003).
8. This includes senior accountants, ?nancial controllers, chief ?nancial of?cers or a role
similar in nature.
9. To become a member of the CFA, an investment professional must meet all of the criteria
listed below:
.
hold a bachelor’s degree from an accredited institution or have equivalent education or
work experience;
.
pass level I of the CFA exam or such other appropriate examination as approved by the
Board of Governors or pass the self-administered Standards of Practice Examination;
.
have 48 months involvement in the investment decision-making process; and
.
agree to adhere to and sign the member’s agreement, a professional conduct statement,
and any additional documentation requested by the CFA Institute.
10. The use of the internet as an appropriate survey method has been supported in the literature
(Couper et al., 2001; Dillman, 2007), andpartlyjusti?edbythe response rate inthe current survey.
11. The 22 seven-point semantic differential scales are recognised in the literature as the most
appropriate scale system to measure connotative meaning in the accounting domain
(Houghton, 1987a, 1988; Houghton and Hronsky, 1993; Houghton and Messier, 1990;
Bagranoff, 1990; Bagranoff et al., 1994; Hronsky and Houghton, 2001; Wines, 2006).
12. Factor compatibility is a form of split-halves test, in which each (randomly selected) half is
subjected to identical factor analyses. The two resulting factor score coef?cient matrices may
then be used (normally with the aid of statistical software) to compute two sets of factor scores
for each participant. The Pearson correlation between these two sets of duplicate factor scores
can “be used to judge the comparability of factors in the analysis” (Nunnally, 1978, p. 433) and
is referred to as the “comparability coef?cient” (Everett and Entrekin, 1980, p. 166).
Comparability is said to exist when all factor(s) are robust and stable, as indicated by
comparability coef?cient scores of .0.894 (corresponding to 80 per cent shared variance).
Factor comparability is determined by initially specifying the number of factors to be
extracted, starting with the number of factors determined from a scree test, then reducing by
one factor each time until comparability (comparability coef?cient scores . 0.894) is achieved
The meaning
of cash
43
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
for all identi?ed factors. Practically, this can be achieved by selecting the largest correlation in
the resulting factor comparability matrix, deleting its row and column, then repeating the
process until all the rows and columns are deleted from the matrix (Everett, 1983).
13. This change can be described as the change in semantic space in which meaning is held for
each of the ?nancial reporting groups (Houghton, 1988).
14. The results of the decision outcomes for task (2) (the binary “yes” or “no” classi?cation) were
correlated with the corresponding outcomes on task (3) (asset classi?cation made on a scale
ranging from “1” (clearly an item of cash) to “6” (clearly not an item of cash)). As may be
expected, these two sets of decisions were highly correlated (Spearman’s r ¼ 0.80, p , 0.01)
suggesting that the responses from tasks (2) and (3) may be used interchangeably. Task (3)’s
responses were used in subsequent data analysis because of their better distributional
qualities (Hronsky and Houghton, 2001).
15. A categorical variable with j categories only requires j 2 1 dummy variables “[. . .] in order
to capture all the distributional information contained in the original set of distinctions”
(Hardy, 1993, p. 7).
16. Decisions scores have a theoretical range from “1” (clearly an item of cash) to “6” (clearly not
an item of cash).
References
Adelberg, A.H. and Farrelly, G.E. (1989), “Measuring the meaning of ?nancial statements
terminology: a psycholinguistics approach”, Accounting and Finance, Vol. 29 No. 1, pp. 33-61.
Bagranoff, N.A. (1990), “The semantic differential: a prescription for use in accounting research”,
Journal of Accounting Literature, Vol. 9, pp. 65-80.
Bagranoff, N.A., Houghton, K.A. and Hronsky, J. (1994), “The structure of meaning in
accounting: a cross-cultural experiment”, Behavioural Research in Accounting, Vol. 6,
pp. 35-57 (supplement).
Barth, M.E., Beaver, W.H. and Landsman, W.R. (2001), “The relevance of the value relevance
literature of ?nancial accounting standard setting: another view”, Journal of Accounting
and Economics, Vol. 31, pp. 77-104.
Bedford, N.M. and Baladouni, V. (1962), “A communication theory in accounting”, Accounting
Review, Vol. 37 No. 4, pp. 650-9.
Broome, O.W. (2004), “Statement of cash ?ows: time for change!”, Financial Analysts Journal,
Vol. 60 No. 2, pp. 16-23.
Buzby, S.L. and Falk, H. (1974), “A new approach to the funds statement”, Journal of
Accountancy, Vol. 137 No. 1, pp. 55-62.
Cattell, R.B. (1966), “The scree test for the number of factors”, Multivariate Behavioural Research,
Vol. 1, pp. 245-76.
Chesley, G.R. and Scheiner, J.H. (1982), “The statement of changes in ?nancial position:
an empirical investigation of Canadian and US users in nonpublic companies”,
International Journal of Accounting, Vol. 17 No. 2, pp. 49-58.
Clift, R.C. (1979), The Funds Statement, Australian Accountants Foundation, Melbourne.
Couper, M.P., Traugott, M.W. and Lamias, M.J. (2001), “Web survey design and administration”,
The Public Opinion Quarterly, Vol. 65 No. 4, pp. 230-53.
DeVellis, R.F. (2003), Scale Development Theory and Applications, 2nd ed., Sage, London.
Dillman, D.A. (2007), Mail and Internet Surveys: The Tailored Design Method, 2nd ed., Wiley,
New York, NY.
ARJ
24,1
44
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Epstein, M. (1992), “Howuseful is the statement of cash ?ows?”, Management Accounting, Vol. 74
No. 1, pp. 52-6.
Everett, J.E. (1983), “Factor comparability as a means of determining the number of factors and
their rotation”, Multivariate Behavioural Research, Vol. 18, pp. 197-218.
Everett, J.E. and Entrekin, L.V. (1980), “Factor comparability and the advantages of multiple
group factor analysis”, Multivariate Behavioural Research, Vol. 15, pp. 165-80.
Fitzgerald, J.T., Stans?eld, R.B., Tang, T., Oh, M., Frohna, A., Armbruster, B., Gruppen, L. and
Anderson, R. (2008), “Patient and provider perceptions of diabetes: measuring and
evaluating differences”, Patient Education and Counseling, Vol. 70, pp. 118-25.
Flamholtz, E. and Cook, E. (1978), “Connotative meaning and its role in accounting change: a ?eld
study”, Accounting, Organisations and Society, Vol. 3 No. 2, pp. 115-39.
Goldberg, L. (1951), “The funds statement reconsidered”, The Accounting Review, Vol. 26 No. 4,
pp. 485-91.
Graci, S.P. (1982), “Effects of the statement of changes in ?nancial position on the short term loan
decisions”, PhD dissertation, University of Arkansas, Fayetteville, AR.
Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1998), Multivariate Data Analysis,
5th ed., Prentice-Hall, Upper Saddle River, NJ.
Hardy, M. (1993), Regression with Dummy Variables, University Paper Series on Quantitative
Applications in the Social Sciences, 07-093, Sage, Newbury Park, CA.
Haried, A.A. (1972), “The semantic dimensions of ?nancial statements”, Journal of Accounting
Research, Vol. 10 No. 2, pp. 376-91.
Haried, A.A. (1973), “Measurement of meaning in ?nancial reports”, Journal of Accounting
Research, Vol. 11 No. 1, pp. 117-45.
Heath, L.C. (1978), “Let’s scrap the funds statement”, Journal of Accountancy, Vol. 146 No. 4,
pp. 94-103.
Henry, E.J. (1975), “A new funds statement formal for greater disclosure”, Journal of
Accountancy, Vol. 139, pp. 56-62.
Hicks, B.E. and Hunt, P. (1980), Cash Flow Accounting, The International Group for Cash Flow
Accounting and the Research and Publication Division, School of Commerce and
Administration, Laurentian University, Sudbury.
Houghton, K.A. (1987a), “True and fair view: an empirical study of connotative meaning”,
Accounting, Organisations and Society, Vol. 12 No. 2, pp. 143-53.
Houghton, K.A. (1987b), “The development of meaning in accounting: an intertemporal study”,
Accounting and Finance, Vol. 27 No. 2, pp. 25-40.
Houghton, K.A. (1988), “The measurement of meaning in accounting: a critical analysis of the
principal evidence”, Accounting, Organisations and Society, Vol. 13 No. 3, pp. 263-80.
Houghton, K.A. and Hronsky, J.J.F. (1993), “The sharing of meaning between accounting
students and members of the accounting profession”, Accounting and Finance, Vol. 33
No. 2, pp. 85-101.
Houghton, K.A. and Messier, W.F. (1990), “The wording of audit reports: its impact on the
meaning of the message communicated”, in Moriarty, S. (Ed.), Accounting Communication
and Monitoring, Centre for Economic and Management Research, The University of
Oklahoma, Norman, OK, pp. 85-114.
Hronsky, J.J.F. and Houghton, K.A. (2001), “The meaning of a de?ned accounting concept:
regulatory changes and the effect on auditors’ decision making”, Accounting,
Organizations and Society, Vol. 26 No. 2, pp. 123-39.
The meaning
of cash
45
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Hsu, S.H., Chuang, M.C. andChang, C.C. (2000), “Asemantic differential studyof designers’ andusers’
product form perception”, International Journal of Industrial Ergonomics, Vol. 25, pp. 375-91.
Humphrey, C.G., Moizer, P. and Turley, W.S. (1993), “The audit expectation gap in Britain:
an empirical investigation”, Accounting & Business Research, Vol. 23, pp. 395-411.
Jones, S. and Ratnatunga, J. (1997), “The decision usefulness of cash-?ow statements by
Australian reporting entities: some further evidence”, The British Accounting Review,
Vol. 29 No. 1, pp. 67-85.
Jones, S., Romano, C.A. and Smynios, K.S. (1995), “An evaluation of the decision usefulness of
cash ?ow statements by Australian reporting entities”, Accounting & Business Research,
Vol. 25 No. 29, pp. 115-29.
Jones, S., Sharma, R. and Mock, K. (1998), “Managerial evaluations of the relevance of cash vs
accrual-based ?nancial reports in the Australian food industry”, Australian Accounting
Review, Vol. 8, pp. 51-9.
Lawson, G.H. (1981), “A new approach to cash ?ow analysis”, The Accountant, Vol. 10 No. 17,
pp. 13-20.
Lee, T.A. (1984), Cash Flow Accounting, Van Nostrand Reinhold, London.
Lee, T.A. (1992), “Making cash ?ow statements useful”, Accountancy, April, p. 35.
Libby, R. (1979), “Bankers’ and auditors’ perceptions of the message communicated by the audit
report”, Journal of Accounting Research, Vol. 17, pp. 99-122.
McEnroe, J. and Martens, S. (2001), “Auditors’ and investors’ perceptions of the ‘expectation
gap’”, Accounting Horizons, Vol. 15, pp. 345-58.
Mason, A.K. and Gibbins, M. (1991), “Judgement and US accounting standards”,
Accounting Horizons, Vol. 5, pp. 14-24.
Mason, P. (1961), Cash Flow Analysis and the Funds Statements, Accounting Research Study
No. 2, AICPA, New York, NY.
Monroe, G.S. and Woodliff, D. (1993), “The effect of education on the audit expectation gap”,
Accounting and Finance, Vol. 33, pp. 61-78.
Mulford, C.W. and Comiskey, E.E. (2002), The Financial Numbers Game, Detecting Creating
Accounting Practices, Wiley, New York, NY.
Mulford, C.W and Comiskey, E.E (2005), Creative Cash Flow Reporting: Uncovering Sustainable
Financial Performance, Wiley, Hoboken, NJ.
Nekolaichuk, C.L., Jevne, R.F. and Maguire, T.O. (1999), “Structuring the meaning of hope in
health and illness”, Social Science & Medicine, Vol. 48, pp. 591-605.
Neubert, W.H. (1959), “Money ?ow and the ?rm”, The Accounting Review, Vol. 34, pp. 84-90.
Noland, V.J., Daley, E.M., Drolet, J.C., Fetro, J.V., McCormack-Brown, K.R., Hassell, C.D. and
McDermott, R.J. (2004), “Connotative interpretations of sexuality-related terms”, Sex Roles,
Vol. 51, pp. 523-34.
Nunnally, J.C. (1978), Psychometric Theory, 2nd ed., McGraw-Hill, New York, NY.
Oliver, B. (1974), “The semantic differential: a device for measuring the interprofessional
communication of selected accounting concepts”, Journal of Accounting Research, Vol. 12
No. 2, pp. 299-319.
Osgood, C.E. (1976), Focus on Meaning, Mouton, The Hague.
Osgood, C.E. and Suci, G.J. (1955), “Factor analysis of meaning”, Journal of Experimental
Psychology, Vol. 50, pp. 325-38.
ARJ
24,1
46
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Osgood, C.E., Suci, G.J. and Tannenbaum, P.H. (1957), The Measurement of Meaning, University
of Illinois Press, Urbana, IL.
Pickard, G. (2007), “Simplifying global accounting”, Journal of Accountancy, July, pp. 36-9.
Pierce, W.D., Sydie, T.A. and Stratkotter, R. (2003), “Social concepts and judgments: a semantic
differential analysis of the concepts feminist, man, and woman”, Psychology of Women
Quarterly, Vol. 27, pp. 338-46.
Porter, B. (1993), “An empirical study of the audit expectation-performance gap”, Accounting &
Business Research, Vol. 23, pp. 49-68.
Shah, A.K. (1996), “Creative compliance in ?nancial reporting”, Accounting, Organizations and
Society, Vol. 21 No. 1, pp. 23-39.
Sharma, D.S. and Iselin, E.R. (2003), “The decisions usefulness of reported cash ?ow and accrual
information in a behavioural ?eld experiment”, Accounting & Business Research, Vol. 33
No. 2, pp. 123-35.
Solomon, D. (2002), “Adelphia overstated cash ?ow, revenue over past two years”,
Wall Street Journal, June, p. B5.
Tergesen, A. (2002), “Cash-?ow hocus-pocus; this seemingly simple number can be anything but
straightforward”, Business Week, No. 3791, p. 130.
Vent, G.A., Cowling, J.F. and Sevalstad, S. (1995), “Cash ?ow comparability: accounting for
long-term debt under SFAS 95”, Accounting Horizons, Vol. 9 No. 4, p. 88.
Vicknair, D.B. (1983), “An investigation of users’ perceptions concerning corporate funds ?ow
information”, DBA dissertation, University of Tennessee, Knoxville, TN.
Wilson, A. (1989), “If it isn’t useful, get rid of it”, Accountancy, Vol. 104 No. 1151, p. 29.
Wines, G. (2006), “The connotative meaning of independence in alternative audit contexts:
an exploratory study”, Paci?c Accounting Review, Vol. 18, pp. 90-122.
Xiong, M.J., Logan, G.D. and Franks, J.J. (2006), “Testing the semantic differential as a model of task
processes with the implicit association test”, Memory &Cognition, Vol. 34 No. 7, pp. 1452-63.
Yap, C. (1996), “Cash ?ow statements: how useful?”, The Australian Accountant, Vol. 66 No. 11,
pp. 36-7.
Yoon, J. (2008), “Searching for an image conveying connotative meanings: an exploratory
cross-cultural study”, Library & Information Science Research, Vol. 30, pp. 312-18.
Further reading
Osgood, C.E. (1964), “Semantic differential technique in the comparative study of cultures”,
American Anthropologist, New Series, Vol. 66, pp. 171-200.
(The Appendices follow overleaf.)
The meaning
of cash
47
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Appendix 1
Figure A1.
Accounting concept: “CASH”– as it relates to the cash flow statement
Definition provided in FRS 10
"… coins, notes, demand deposits and other highly liquid investments in which an
entity invests as part of its day to day cash management." And "…borrowings from
financial institutions such as bank overdrafts, where such borrowings are at call and are
used as part of day-to-day cash management." (Para 4.1)
Based on the definition of CASH provided above, you believe cash to be:
Exact Estimated
Bad Good
Measurable
Unmeasurable
Necessary
Unnecessary
Planned
Unplanned
Objective
Subjective
Tangible
Intangible
Strong
Weak
Indirect
Direct
Variable
Constant
Safe
Risky
Complete
Discretionary
Required
Real
Imaginary
Beneficial
Adverse
Temporary
Permanent
Controllable
Uncontrollable
Unexpected
Expected
Passive
Active
Static
Dynamic
Long-term Short-term
Inflexible Flexible
ARJ
24,1
48
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Appendix 2
Corresponding author
Richard Fisher can be contacted at: [email protected]
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
Figure A2.
Accounting concept: “CASH”– as it relates to the cash flow statement
Definition provided in FRS 10
“… coins, notes, demand deposits and other highly liquid investments in which an
entity invests as part of its day to day cash management.” And “…borrowings from
financial institutions such as bank overdrafts, where such borrowings are at call and
are used as part of day-to-day cash management.” (Para 4.1)
Based on the above definition, what degree do you believe the following represent
items of CASH.
Yes
1. Coins and notes on deposit
2. Accounts receivable
3. A four month treasury bill
4. Reserve bank bill
5. Gold bullion
6. Readily tradable equity securities
7. A three month futures contract
8. Non-cash payments for goods and service
barter
9. On-call bank overdraft
10. Preference shares redeemable in three months
for a fixed amount of cash
No
The meaning
of cash
49
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
2
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_207362497.pdf