The expressive role of performance measurement systems

Description
The management control systems (MCS) literature has long recognized the importance of
values and beliefs (e.g., Ouchi, 1979; Simons, 1995). However, in this literature, values and
beliefs are typically presented in the context of mission statements or company slogans
that can play little substantive role in shaping actions and behaviors. In this paper we focus
on how MCS can play a more active role in values expression, and examine the potential for
performance measurement systems (PMS) to be used within organizations to express the
values and beliefs of organizational members. This use of PMS, which we term its expressive
role, is important as pluralistic and expressive forms of organizing are becoming more
prevalent. Furthermore, prior research indicates that enabling the expression of values and
beliefs by organizational members can generate energy and commitment that are
important to the achievement of organizational objectives

The expressive role of performance measurement systems:
A ?eld study of a mental health development project
Robert H. Chenhall
a
, Matthew Hall
b,?
, David Smith
a
a
Department of Accounting, Monash Business School, Monash University, Australia
b
Department of Accounting, London School of Economics and Political Science, United Kingdom
a b s t r a c t
The management control systems (MCS) literature has long recognized the importance of
values and beliefs (e.g., Ouchi, 1979; Simons, 1995). However, in this literature, values and
beliefs are typically presented in the context of mission statements or company slogans
that can play little substantive role in shaping actions and behaviors. In this paper we focus
on how MCS can play a more active role in values expression, and examine the potential for
performance measurement systems (PMS) to be used within organizations to express the
values and beliefs of organizational members. This use of PMS, which we term its expres-
sive role, is important as pluralistic and expressive forms of organizing are becoming more
prevalent. Furthermore, prior research indicates that enabling the expression of values and
beliefs by organizational members can generate energy and commitment that are
important to the achievement of organizational objectives. In a ?eld study of a mental
health development project in a non-government organization, we examine the design
and operational characteristics that are important for the expressive role of PMS. We also
examine the interplay between the expressive role and the instrumental role of PMS and
identify circumstances in which these roles can clash and/or be complementary.
Ó 2014 Elsevier Ltd. All rights reserved.
Introduction
The management control systems (MCS) literature has
long recognized that values and beliefs are important in
the functioning of organizations (Ouchi, 1979; Simons,
1995). Scholars have questioned, however, whether those
MCS often associated with values and beliefs, such as mis-
sion statements and company slogans, play any substantial
role in shaping actions and behaviors (e.g., Argyres &
McGahan, 2002: 48). Furthermore, even in contexts where
MCS are envisioned to play a more prominent role, it is
typically to promote conformance by organizational
members with corporate values and beliefs espoused by
senior management (Ouchi, 1979; Simons, 1995). In this
way, there has been little attention directed toward the
possibility for MCS to play a more active role in values
expression, particularly in the context of enabling a wider
variety of organizational members (and not only senior
managers) to express their beliefs and values as part of
their work in organizations.
Attention to a more active role for MCS in values
expression is important because recent research indicates
that organizations are increasingly operating in more
pluralistic contexts characterized by multiple objectives
(Denis, Langley, & Rouleau, 2007). In this context, in
addition to a purely instrumental rationale focused on
the pursuit of speci?c objectives, organizations can
have an expressive purpose (Berry, 2005; Etzioni, 1961;
Frumkin, 2005; Schultz, Hatch, & Larsen, 2000). This
expressive purpose re?ects a focus on helping individuals
to express their values, commitment and faith throughhttp://dx.doi.org/10.1016/j.aos.2014.11.002
0361-3682/Ó 2014 Elsevier Ltd. All rights reserved.
?
Corresponding author.
E-mail address: [email protected] (M. Hall).
Accounting, Organizations and Society xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Accounting, Organizations and Society
j our nal homepage: www. el sevi er. com/ l ocat e/ aos
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
their work in organizations (Frumkin, 2005; Gordon &
Babchuk, 1959; Mason, 1996; Schultz et al., 2000).
A focus on the expressive dimension of organizations is
consistent with emerging research attempting to reposi-
tion the role of organizational members’ values, beliefs
and emotions at work – not as a nuisance or obstacle to
organizational progress, but rather as a potentially produc-
tive feature of the work environment (Boedker & Chua,
2013; Bolton & Houlihan, 2009; Dutton, Worline, Frost, &
Lilius, 2006; Huy, 1999). In particular, research indicates
that enabling the expression of values and beliefs by orga-
nizational members can generate energy and commitment,
enhance learning and exploration of alternatives, and
reduce value con?icts in ways that can aid the collective
achievement of organizational objectives (c.f., Frumkin,
2005; Huy, 1999).
In this study we examine the potential for PMS to be
used within organizations to express the values and beliefs
of organizational members. In particular, we consider
whether and how organizational members’ involvement
with PMS can potentially provide a context within which
their values and beliefs can be expressed. This focus on
organizational members’ substantive engagement with
PMS seeks to highlight a more active role for MCS in values
expression, moving beyond the typically passive role
ascribed to elements of MCS such as mission statements
and slogans. We also seek to pay particular attention to
the expression of values and beliefs held by a variety of
organizational members, and not only on how PMS can
be used instrumentally by senior managers to encourage
conformance with corporate values and beliefs. Speci?-
cally, we address three interrelated research questions:
Can PMS help organizational members to express their
values and beliefs? If so, what design and operational
characteristics of the PMS could help to facilitate this
process? How does the expressive role of PMS relate to
and/or con?ict with an instrumental use of PMS in
organizations?
To analyze the potential for an expressive role of PMS
we conducted a ?eld study of the development of a PMS
in the Sri Lankan of?ce of Voluntary Service Overseas
(VSO), a non-government organization (NGO) working in
the ?eld of international development. NGOs, like VSO,
are characterized typically by the instrumental pursuit of
speci?c objectives, such as poverty reduction and meeting
other developmental goals related to health, education and
empowerment. In addition, they are characterized by a sal-
ience of beliefs and values related to missions to help the
disadvantaged, where organizational members participate
in these organizations in order to help express their com-
mitment to these values and causes (Frumkin, 2005;
Mason, 1996). Our empirical setting provides us with a
powerful lens through which to examine the potential for
PMS to play an expressive role, as well as the way in which
this expressive role relates to the instrumental use of PMS
in organizations.
Our study contributes to the literature by developing
and advancing understanding of the expressive role of
PMS. We conceptualize this role as the capacity of a PMS
to facilitate the display of a variety of values and beliefs
held by organizational members. The ?ndings from our
?eld study indicate that the ability of a PMS to enable
organizational members to express their values and beliefs
is related to speci?c design and operational characteristics.
The creation of a ‘playful’ environment (c.f., Huy, 1999) in
which the PMS operates, and a PMS that is easily accessible
to organizational members, are both important in order for
the PMS to operate in an expressive role. Prompting the
expression of values and beliefs by organizational mem-
bers is then facilitated by the indicators included in the
PMS aligning with the important values and beliefs of
organizational members, where the development of a close
af?nity between the speci?c values and beliefs of organiza-
tional members and the precise content and wording of
those indicators is paramount.
Our identi?cation and conceptualization of the expres-
sive role of PMS contributes to the MCS literature (e.g.,
Simons, 1995) by exploring how values and beliefs can
be discussed and expressed as part of MCS and how this
actually shapes behavior. This is in contrast to, for exam-
ple, MCS in the form of a mission statement that is neither
referred to, nor thought about, by organizational members
(e.g., Argyres & McGahan, 2002: 48). This focus also con-
tributes to emerging research positioning the expression
of organizational members’ values and emotions as a
potentially productive feature of the work environment
(Boedker & Chua, 2013; Bolton & Houlihan, 2009; Dutton
et al., 2006; Huy, 1999). In particular, our study provides
insight into the way in which the expression of values
through PMS can help the achievement of organizational
objectives by mobilizing the energy, motivation and com-
mitment that values expression can create (c.f., Dutton
et al., 2006; Frumkin, 2005; Huy, 1999). Examination of
the expressive role of PMS is likely to be important for
those organizations that have an evident expressive pur-
pose, such as NGOs, and for organizations more generally
as pluralistic and expressive forms of organizing become
more prevalent (Denis et al., 2007; Schultz et al., 2000;
Huy, 1999; Dutton et al., 2006).
The remainder of the paper is structured as follows.
The next section provides the theoretical framework for
the study. The third section details the research method.
The fourth section provides information on the case con-
text, with the ?fth and sixth sections presenting our empir-
ical analysis from our ?eld study. In the ?nal section we
discuss our ?ndings and provide concluding comments.
Theory development
Our conceptualization of the expressive role of PMS has
its roots in literature seeking to distinguish between
expressive and instrumental organizations, a typology ?rst
adopted by Gordon and Babchuk (1959). An expressive
organization is one whose activities can help individuals
to express their values, commitment and faith through
their work, while an instrumental organization is one
designed to maintain or create some normative aim
(Frumkin, 2005; Gordon & Babchuk, 1959; Mason, 1996).
Recent research has indicated that rather than being
either purely expressive or purely instrumental, organiza-
tions can operate in more pluralistic contexts characterized
2 R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
by multiple objectives (Chenhall, Hall, & Smith, 2010; Denis
et al., 2007). On this basis, the idea that all organizations
have an expressive purpose (to varying degrees) has gained
increasing attention. This is founded on the idea that orga-
nizations compete based on their values, or what they are
seen to stand for (Schultz et al., 2000). This can be the case
for both for-pro?t and non-pro?t organizations. In consid-
ering for-pro?t organizations, Shaw (2000) notes the
importance of values relating to innovation at 3Mand Hew-
lett–Packard and the manner in which these organizations
have used stories of organizational members past and pres-
ent to highlight these values to newmembers. On this basis,
rather than organizations being purely expressive or instru-
mental, it makes more sense to consider the expressive and
instrumental purposes that may be at play within any given
organization. As such, our study focuses on the role PMS
can play in supporting this expressive purpose, as well as
instrumental roles for PMS in organizations.
PMS, roles of accounting information, and the expression of
values and beliefs
In examining the expressive role of PMS, we focus on
the way in which MCS can be used to express values and
beliefs. Prior MCS research has tended to focus on the
values of top managers where, for example, they can use
interactive control systems to reveal their values and
preferences to other managers and employees in the
organization (Simons, 1995). However, this is very much
a top-down, hierarchical process whereby the values of
top managers are communicated to others in the organiza-
tion, with little capacity for the expression of values and
beliefs by employees at lower levels of the organization.
We seek to understand how PMS can help to express or
‘signal’ the values and beliefs of a variety of organizational
members, not only top managers. Prior MCS research also
tends to locate values under the domain of ‘belief systems,’
where practices such as mission statements are used to
valorize particular values (Simons, 1995). However, such
practices seemingly have little active or on-going role in
helping organizational members to discuss and express
their values as part of their work (e.g., Argyres &
McGahan, 2002: 48). In contrast, our focus is on how the
PMS can become involved in the communication, discus-
sion and expression of values by members of the organiza-
tion. As such, we consider how PMS can be used to discuss
and express the beliefs and values of organization mem-
bers, and how these systems can be used to shape
behavior.
In line with this discussion, we de?ne the expressive
role of PMS as the capacity of a PMS to facilitate the display
of a variety of values and beliefs held by organizational
members. In crafting and elaborating this de?nition, we
?rst draw on research into organizational symbolism.
Scholars have focused on a variety of different types of
organizational symbols related to the core and distinctive
values and beliefs of an organization, and, in particular,
have identi?ed the use of material objects and artefacts
as being a potentially powerful symbol in this process of
values expression (c.f. Pratt & Rafaeli, 1997; Trice &
Beyer, 1993).
PMS typically (if not always) involve the development
of material objects and artefacts, such as indicators,
spreadsheets, and performance reports. Smith and
Stewart (2011) argue that the process of transferring unex-
pressed beliefs and values into comprehensible symbols
and material artefacts is important for the expression of
values. Boedker and Chua (2013) argue that emotions like
hope, desire, fear and passion can come alive and circulate
through material artefacts such as accounting templates,
re?ecting their role in not only engaging intellectual and
reasoning skills, but also playing on people’s passions
and feelings. Other research indicates that measurement
and evaluation practices can provide an important mecha-
nism through which individuals in organizations can
re?ect upon and express their beliefs and values (Abma,
1997; Dart & Davies, 2003; Greene, 1999; Howes, 1992).
This research suggests the development and operation of
a PMS, particularly its manifestations in material artefacts,
could help staff to think about, re?ect on and articulate
their values and beliefs.
In elaborating this idea, we draw on Huy’s (1999) work
on emotional dynamics, speci?cally on the role of ‘display
freedom.’ Huy (1999) refers to display freedom as the
organization’s ability to facilitate the variety of emotions
that can legitimately be displayed and felt in the organiza-
tion. We adapt this conceptualization of display freedom to
refer to the capacity of a PMS to facilitate the display of a
variety of values and beliefs held by organizational
members. Huy’s (1999) work suggests that the creation
of an environment in which there is freedom from the fear
of reprisal for holding ‘inappropriate’ views, and in which
there is ‘time’ and ‘space’ to have the ability for re?ection,
experimentation and expression of new ideas, is likely to
support the development of display freedom. In contrast,
display freedom is low when organizational members are
faced with an organizational climate where they feel
restricted in their ability to share their values and
beliefs (for example, where there is fear of reprisal, or
embarrassment).
On ?rst consideration, it may appear that an instrumen-
tal use of accounting places little or no explicit emphasis on
the expression of values and beliefs held by organizational
members, given that its focus is on ef?ciency and the pur-
suit of instrumental ends. However, a large body of more
critical accounting scholarship illustrates how the (often
typical) focus in PMS on ?nancial goals can serve to pro-
mote a narrow range of values related to ef?ciency and
purely economic or pro?t-based ends (e.g., Cushen, 2013;
Dent, 1991; Ezzamel, Willmott, & Worthington, 2008). This
suggests that a strong focus on economic goals may restrict
the expressive role of PMS.
1
In contrast, an expressive role
of PMS need not be restricted to a focus on improvements
in operations and ef?ciency, even though the energy gener-
ated from the process of articulating beliefs and values can
complement efforts to improve organizational effectiveness
(c.f., Frumkin, 2005). Frumkin (2005) notes that the instru-
mental and expressive perspectives may not necessarily be
1
An exception to this could be in situations where organizational
members happen to value highly goals related to organizational ef?ciency
and pro?t-based ends.
R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx 3
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
in tension. Speci?cally, PMS playing an instrumental role
may support expressive purposes by serving to highlight
and focus attention on areas seen as important to the values
of organizational members. As above, however, this may
only be possible where there is scope for the inclusion of
organizational members’ values in the PMS that go beyond
the purely economic or instrumental.
The above discussion examined the role of organiza-
tional members’ values and beliefs in the development of
a PMS. However, when and how a PMS could provide
organizational members with opportunities to express
their values and beliefs is unclear, and is likely to depend
on the presence of certain design and operational
characteristics.
Characteristics of PMS enabling the expression of values and
beliefs
In considering the characteristics that may enable the
expression of values and beliefs of organizational mem-
bers, we ?rst draw on research suggesting that the engage-
ment of organizational members in the performance
measurement process is likely to be important for the
expressive role of PMS. This can involve opening up infor-
mation production activities to a wider variety of partici-
pants in order to make the PMS more accessible and
visible to organizational members (c.f., Deetz, 1995;
Greene, 1999). This could involve providing organizational
members with the opportunity to ask questions, offer
views and provide suggestions during the development
and operation of the PMS (c.f., Hirschheim & Klein, 1994).
Organizational members could be given power to negotiate
collectively the development of the PMS, where engage-
ment is more than a mere ‘feel good’ exercise akin to a
‘sham ritual’ (Ebrahim, 2003; Najam, 1996). Engagement
of organizational members with the performance mea-
surement process may also involve senior staff adopting
an empowering approach facilitating dialogue and infor-
mation sharing rather than being overly commanding or
directive (c.f., Bisbe, Batista-Foguet, & Chenhall, 2007;
Simons, 1990). Research also suggests that involving
organizational members in the development and operation
of the PMS may facilitate the expressive role as they are
best placed to judge whether their values and beliefs are
suitably re?ected in the PMS (c.f. Wouters & Wilderom,
2008).
To understand how to engage organizational members
in the performance measurement process, we ?rst draw
on Huy’s (1999) work on ‘playfulness.’ Huy (1999) refers
to playfulness as the ability of the organization to create
a context promoting experimentation and tolerating mis-
takes. We adapt this conceptualization to refer to the orga-
nization’s ability to create an environment for developing
and operating the PMS in which mistakes are tolerated
and experimentation is encouraged. Such an environment
would be characterized by jokes, laughter, and playful
activities legitimizing trial and error (Huy, 1999) as part
of the PMS’s operation. This is consistent with the notion
that fun and playfulness are an important part of organiza-
tional life and part of attempts to engage with the human
side of organizing (Bolton & Houlihan, 2009).
The dynamic of playfulness appears particularly impor-
tant in enabling freedom of expression. The process under-
pinning the development of the PMS would need to have
aspects of ‘playfulness’ in order for staff to feel comfortable
enough to provide suggestions as to the properties of the
PMS. This is consistent with research focusing on the
way in which fun and playful activities can create a sense
of involvement (Bolton & Houlihan, 2009). Playful and
game-like qualities in accounting systems can also encour-
age organizational members to be more open and to
engage in more creative, innovative and experimental
behaviors (Cooper, Hayes, & Wolf, 1981). It would appear,
then, that the presence of ‘‘playfulness’’ in the develop-
ment of a PMS would promote ‘‘display freedom’’, and thus
the expression of values and beliefs in the PMS.
In addition to Huy (1999), we also draw on research in
evaluation, organizational communication and manage-
ment accounting that helps to shed light on how to engage
organizational members in the PMS. Research in evalua-
tion suggests formal evaluation techniques can aid the
expression of values by providing a medium through
which individuals feel comfortable expressing their per-
ceptions, which often involves modifying conventional
frameworks to suit the skills and culturally familiar idioms
and styles of expression of participants (Howes, 1992).
These modi?ed formats can then provide memory triggers
whereby individuals are able to recount and reconstruct
key events and experiences as part of the formal evaluation
process (Howes, 1992). Evaluation that is responsive to
and promotes the expression of the values of participants
tends to be formalized in a more open and narrative style,
which helps to invite re?ection, rather than in a proposi-
tional style, which limits dialogue by being more declara-
tive about ?ndings and their meaning for participants
(Abma, 1997).
Research also suggests that in order for a PMS to enable
organizational members to express values and beliefs, it
needs to be accessible. This is because organizational
members are unlikely to engage with and adequately
express their values and beliefs through a PMS that they
cannot understand. Prior research indicates that user
engagement with formal systems can be aided by present-
ing information in a language familiar to the operator
rather than the language and terms of technical staff
(Adler & Borys, 1996).
2
This is because overly formal and
technical language can provide an expressive privilege to
(often) managerial voices and thus makes it dif?cult for a
wider variety of organizational members to express their
views and opinions (Deetz, 1992, 1995). The use of more
every-day and natural language in formal systems can help
organizational members to access their content and re?ect
upon its signi?cance for them (Abma, 1997; Deetz, 1992).
This can also increase access to formal systems by limiting
the need for training in specialized techniques or abstract
2
Adler and Borys (1996) refer to the use of familiar language as part of
their discussion of ‘internal transparency’ within their description of
enabling and coercive bureaucracy. This work has, of course, been
developed further in management accounting research to refer to enabling
and coercive controls (see for example, Ahrens & Chapman, 2004; Wouters
& Wilderom, 2008).
4 R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
concepts (Hall, 2014). Management control process can be
made more accessible to organizational members by, for
example, calculating variances using categories relating to
operational work, integrating budgeting with operational
planning, and by presenting simple information in easily
digestible formats so staff can process them quickly
(Ahrens & Chapman, 2004). Research also suggests PMS
can be made more accessible by eliminating accounting
jargon and using plain language instead of technical terms
(Rowe, Birnberg, & Shields, 2008).
Method
For this study, the second author was a full-time volun-
teer at Voluntary Service Overseas (VSO), an international
development NGO. The author’s volunteer placement was
as a ‘‘Monitoring and Evaluation Advisor’’ responsible for
helping the mental health programme to develop ‘‘appro-
priate M&E systems to monitor progress of the VSO mental
health programme/project.’’
3
The author became aware of
the volunteer placement during a research visit to Sri Lanka
in January 2009 that formed part of a wider study the
authors were conducting on VSO (Chenhall, Hall, & Smith,
2013). The author subsequently applied to become a VSO
volunteer through the standard VSO volunteer recruitment
processes. Accordingly, during this time the author’s
position changed from a researcher studying VSO as an ‘out-
sider’, to a participant observer or ‘insider’ who was
immersed in the real-time ?ow of activities in the mental
health programme (c.f. Ahrens & Mollona, 2007; Jönsson &
Lukka, 2007). The author completed the volunteer place-
ment in two visits to Sri Lanka in 2009, the ?rst during
two weeks in April and the second during July–September.
4
During the placement the author was based in the VSO Sri
Lanka of?ce, located in a suburb of Colombo, where most
of his time was spent. He also travelled to other locations
in Sri Lanka to visit various mental health facilities where
other volunteers were working. In total, the author spent
61 days as a participant–observer, where, in addition to his
work as a volunteer, he observed the everyday goings-on
at the of?ce over the course of the placement, participated
in lunches and conversations with of?ce staff and volun-
teers, as well as other activities such as social gatherings,
meetings, and weekends away.
The majority of the placement concerned the develop-
ment of a new system for use in the monitoring and eval-
uation of a mental health development programme
operated by VSO in Sri Lanka (which came to be known
by the acronym ‘LEAP’). Compared with interview-based
studies, participant observation can make more credible
claims toward studying accounting in ‘action’ (Ahrens &
Mollona, 2007). It can help researchers to gain the trust
of organisational members, which can provide access to
information often unavailable to outsiders (Rowe et al.,
2008). This can provide the opportunity to collect more
signi?cant and subtle data as the researcher can see what
organizational members actually say and do and what
really matters to them (Jönsson & Lukka, 2007) and to
get a closer look at people’s facial and bodily responses
(Boedker & Chua, 2013). Participant observation is particu-
larly important to this study in order to observe organiza-
tional members in situ and see their responses (verbal and
otherwise) to the PMS ‘close up.’ To this end, there was an
explicit agreement that the volunteer placement would
serve the dual purposes of helping the mental health pro-
gramme to develop a PMS as well as provide an opportu-
nity for the author to gather empirical data for research
purposes.
Overall, we adopted an interventionist approach. The
author’s participation was as follows: he facilitated the
involvement of programme staff, volunteers and partners’
in the development of LEAP, organized regular meetings
where staff and volunteers’ feedback on the tool was
obtained and then incorporated into each subsequent ver-
sion, and coordinated completion of LEAP at each of the
mental health facilities. During these processes he spoke
with organizational members (individually and in groups)
about their work and the on-going development of LEAP. In
this way, the author participated in a change process with
people working in the organization that lead to the crea-
tion of a new accounting practice, participating on equal
terms with other members of the team and sharing
responsibility for the development of LEAP (c.f., Jönsson
& Lukka, 2007). The author’s involvement was democratic
(Jönsson & Lukka, 2007) in the sense it was the organiza-
tional members who developed the indicators in LEAP
and took the primary role in scoring the indicators and pro-
viding narrative descriptions during its completion at each
mental health facility, albeit with help and support from
the author. Our involvement at VSO did not set out to test
or develop a particular accounting innovation or practice
(Wouters & Wilderom, 2008) but was designed to gather
empirical data in order to theorize about how PMSs were
developed and operated, particularly in a setting with lim-
ited resources, expertise and time-pressure. As such, we
did not enter the ?eld with an idea of the expressive role,
so our analysis of the data in relation to the expressive role
of PMS took place entirely after the participant observation
had ended (Ahrens & Chapman, 2004).
Formal data collection consisted of ?eld notes, meet-
ings, interviews and documents. The author took detailed
?eld notes during each day of the research period, keeping
a hand-written diary providing a description of events as
they unfolded, which is important in preserving the
traceability of the research process (Jönsson & Lukka,
2007). Each evening after work, the hand written notes
were turned into an ‘expanded account’ (Spradley, 1980)
using an electronic word processing system. Once a day’s
entry was completed it was not edited (Jönsson & Lukka,
2007). The expanded account totalled 127 pages of text.
Additionally, seven meetings concerning the development
and implementation of the LEAP system were digitally
recorded and transcribed, representing a total of over
14 h of meetings. Further, we draw on documents from
the LEAP development and implementation process,
comprising seven iterations of the LEAP report template,
as well as supporting documentation. We also draw on
3
Source: Placement outline for Monitoring and Evaluation Advisor.
4
During the research period the ?rst and third authors were ‘outsiders’
who remained involved in on-going research at VSO but did not participate
directly in the work in Sri Lanka.
R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx 5
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
interviews with staff and volunteers involved with the VSO
Sri Lanka programme of?ce (seven interviews), observa-
tions of of?ce meetings (?ve meetings in total), and over
100 documents relating to VSO Sri Lanka’s mental health
project and the VSO Sri Lankan programme of?ce more
generally. We also sought to provide a longitudinal per-
spective to the study by conducting interviews and collect-
ing further documentation at the completion of the mental
health programme in February 2013. We conducted an
additional three interviews (consisting of an interview
with the Mental Health Programme Manager, a Mental
Health Programme Volunteer, and the Health Programme
Manager at VSO Head Of?ce) and collected documents
and reports pertaining to the overall evaluation of the pro-
ject. Table 1 provides an overview of the formal data
collection.
5
With respect to data analysis, we followed the approach
outlined by Eisenhardt (1989), which has been utilized in
several prior accounting studies (see for example, Ahrens
& Chapman, 2004; Chenhall et al., 2010; Free, 2007). This
involved arranging the different types of data (?eld notes,
meeting transcripts, documents) chronologically and iden-
tifying common themes and unique insights and also areas
of disagreement. We also used archival records to elabo-
rate and re?ne our understanding of important issues that
arose in ?eld notes, meetings and interview discussions.
We then re-organized the original transcripts around key
events and issues; and compared emerging ?ndings from
our study with existing research.
Case context
VSO is an international development NGO that works
by (mainly) linking volunteers with partner organizations
in developing countries. VSO’s head of?ce is in London,
United Kingdom, with a ‘country of?ce’ situated in most
of the developing countries in which VSO operates.
Volunteers are typically professionals with substantive
experience in their ?eld, and their placements (lasting
from three months to two years) involve taking up a spe-
ci?c position in an organization, usually working alongside
a local staff member.
6
The VSO Sri Lanka country of?ce is located in Colombo,
with 12 full-time staff, including a Country Director, three
Programme Managers, and a variety of support staff
responsible for volunteer relations, administration and
logistics. VSO Sri Lanka operated programmes in three
areas: mental health, disability, and participation and gov-
ernance, with each programme area headed by one of the
three Programme Managers. Each programme consisted
of a set of partner organizations and a group of volunteers.
The mental health programme
VSO had identi?ed mental health as ‘a neglected area
with services and resources concentrated in Colombo and
a few urban areas.’
7,8
The VSO mental health programme
had volunteers working in a variety of roles in the Sri Lankan
mental health sector, typically as trainers in contemporary
and community-based mental health care. Most volunteers
were quali?ed mental health practitioners from the United
Kingdom’s National Health Service (NHS), with some volun-
teers from the Netherlands and India. At the time this
research was undertaken, 10 volunteers were in placement
in Sri Lanka. Most partner organizations in the mental health
programme were government hospitals and rehabilitation
centres, along with two NGOs. The country of?ce worked
closely with the Sri Lankan Ministry of Health through on-
going involvement in regular meetings of the mental health
sector and having senior mental health professionals on the
programme’s advisory committee. The mental health pro-
gramme was primarily organized by two staff, a ‘Programme
Manager’ and a ‘Programme Of?cer’, who were responsible
for day-to-day programme organization, administrative
tasks such as planning, budgeting, and reporting, develop-
ment of partnerships and volunteer placements, and liaising
with current volunteers and partner organizations on a reg-
ular basis.
9
In 2008 VSO obtained external funding for the mental
health programme, with a four-year project worth approx-
imately one million euros. The project was 75% funded by
Table 1
Formal ?eldwork activity.
Participant observation 61 days, 127 pages
of typed ?eld notes
Observation and attendance at meetings
LEAP meetings 7 meetings
VSO Sri Lanka of?ce meetings 5 meetings
Interviews
VSO Sri Lanka staff 5 interviews
VSO Sri Lanka volunteers 4 interviews
VSO Head Of?ce – Health programmes
manager
1 interview
Documents
LEAP reporting templates 7 iterations of the
template
LEAP guidance documents >50 documents
Mental health programme and VSO Sri
Lanka of?ce
>100 documents
5
We also draw from prior research on VSO (including 32 interviews from
Chenhall et al. (2013)), which although not referenced directly in the
empirical material below, helped to provide the knowledge necessary to
make meaningful and valid interpretations of the practices we observed
(c.f., Alac & Hutchins, 2004).
6
VSO operates programmes in a variety of areas, including health,
education, secure livelihoods, HIV/AIDS, disability, and governance. For
further information on VSO and its operations, see www.vsointernation-
al.org (accessed 5 September 2012) and Chenhall et al. (2013).
7
Source: ‘Supporting and Developing Rights-Based Mental Health Ser-
vices in Sri Lanka’ – project description document.
8
In 2005 the Sri Lankan government issued the ‘Mental Health Policy of
Sri Lanka 2005–2015’ that identi?ed the poor state of mental health in Sri
Lanka, stating that it had ‘an estimated 2% of the population suffering from
serious mental illnesses.’ The policy was primarily directed at establishing
minimum service and staf?ng levels in each of the districts of Sri Lanka,
with a focus on promoting community-based mental health care that
catered for social as well as medical needs (Sri Lankan Mental Health Policy,
see www.searo.wo.int/LinkFiles/On_going_projects_mhp_slr.pdf, accessed
5 September 2012.
9
For further information about the mental health programme, see
www.vso.org.uk/Images/sri-lanka-mental-health-summary-mar07_tcm79-
20599.pdf, accessed 5 September 2012.
6 R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
the European Commission, with other funding from the
World Health Organization, Astra Zeneca and the Silvia
Adams Charitable Trust. There were four stated objectives
for the mental health project:
1. National Mental Health Policy implemented in six tar-
get provinces within Sri Lanka;
2. mental health workers in the six target provinces use
more client-centred and rehabilitation-focused
approaches;
3. newly trained mental health workers and non-mental
health workers are trained using interactive and practi-
cal methodologies; and
4. partners engage in more community-based rehabilita-
tion-focused approaches, in particular ensuring the
involvement of people with mental health problems
and/or their families in these approaches.
10
These objectives of the mental health project focused on
the extent and quality of mental health practices across
different provinces, as well as the involvement of bene?-
ciaries and their families. The mental health project
required very detailed reporting on budgeting and expen-
ditures according to pre-speci?ed budget codes and
reporting procedures. These reports were submitted quar-
terly to the European Commission. The project also
required VSO to provide annual and end-of-project evalua-
tions on progress in achieving the stated objectives, but the
precise content and format of this evaluation was not spec-
i?ed by the donors, meaning VSO was able to develop its
own measurement and reporting system for this purpose.
To this end, VSO proposed that evaluation of the mental
health project ‘include annual partner reviews, and mid-
term reporting by partners and volunteers on progress
between annual reviews. These involve participatory exer-
cises and workshops with the volunteers, partner’s staff
and client groups and other stakeholders.’
11
External funding provided a moderate amount of
resources with which to undertake monitoring and evalu-
ation activities as well as funding volunteer placements
focused on monitoring and evaluation. In this way, the
mental health project provides a useful empirical setting
for examining the expressive role of PMS as it is unlikely
to be dominated by external pressures, and there is the
potential for tension between the expressive and instru-
mental roles of PMS in that it should be ‘useful’. Further,
the project was undertaken in an environment with suf?-
cient ?nancial resources to enable the PMS’s development.
Developing a performance measurement system
One of the main existing reporting practices used in the
mental health programme was the ‘Volunteer Report’,
which focused on the activities of an individual volunteer
placement. Each volunteer completed it, typically at two-
month, 12-month and 24-month intervals. This reporting
format had been in place for many years, and was used
widely across volunteer placements throughout VSO. The
report was a narrative account that included sections such
as ‘the volunteer’s main activities so far’ and asked volun-
teers to report on ‘how has your job developed since you
arrived and what lessons can be learnt’ and ‘how have
you adapted personally?’
12
This report was considered
inadequate by the Programme Manager, as it did not relate
to the mental health project’s objectives outlined above.
The development of a new PMS for the mental health
programme began in August 2008, when the Mental
Health Programme Manager devised an Excel spreadsheet
centred on reporting progress toward the four objectives
of the programme (see Fig. 1 for a timeline of key events
in the development of a new PMS). The spreadsheet had
rows containing questions to gauge progress on each
objective, with each column corresponding to one of the
six provinces in Sri Lanka.
13
In attempting to complete the
spreadsheet, the volunteers and Programme Manager were
required to consider the effects of VSO’s work in each prov-
ince. This proved problematic, however, with the Pro-
gramme Manager commenting that the spreadsheet she
had devised was not very ‘user-friendly’
14
and the volun-
teers did not complete it. Little progress was made, until
April 2009, with the beginning of the researcher’s volunteer
placement, along with another mental health programme
volunteer working part-time on the process (referred to here
as Volunteer 1).
The development of the PMS took the Excel spreadsheet
as the starting point and proceeded through a total of
seven different designs (comprising different sections and
indicators) that were trialed and tested during the period
April–August 2009. Each round of this testing involved
the development of a new design, which was then dis-
cussed in face-to-face meetings with mental health volun-
teers, VSO staff and some representatives from partner
organizations. Feedback from these meetings was then
incorporated into the next design of the PMS, which was
taken to another meeting, and so on. The PMS that was
developed from this process came to be called LEAP, an
acronym representing ‘Learning, Evaluation, Assessment
for Partners.’ This PMS was to be completed for each of
the mental health facilities with which VSO had placed a
volunteer. The ?nal version of LEAP had four sections:
Capacity Building and Service Delivery, Training and
Behavior of Staff, Patient Experiences, and the Impact of
Mental Health Services on the Lives of Bene?ciaries. Each
section consisted of a set of indicators, where each indica-
tor was scored on a 5-point scale and accompanied by nar-
rative information.
15
The following sections reveal the different ways in
which a PMS can help organizational members to display
and express their beliefs and values. First, we focus on
the way in which a PMS can be developed to engage people
10
Source: ‘Supporting and Developing Rights-Based Mental Health Ser-
vices in Sri Lanka’ – project description document.
11
Source: ‘Supporting and Developing Rights-Based Mental Health Ser-
vices in Sri Lanka’ – project description document.
12
Source: Sample of Volunteer Reports.
13
See Appendix 1 which shows a sample from the spreadsheet.
14
Source: Field notes, January and April 2009.
15
See Appendix 2 which shows the complete set of indicators, and
Appendix 3 which reproduces the beginning of Section 1 and indicator 1.2
from LEAP as given in the ?nal version.
R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx 7
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
in the performance measurement process. Second, we
analyze the way in which the design and operation of the
PMS can embed (to varying degrees) and therefore help
with the display of values and beliefs held by staff and
volunteers.
Engagement, accessibility and playfulness
This section examines the way in which LEAP was
developed and operated to engage staff, volunteers and
others in the performance measurement process. Below
we analyze how engagement was facilitated in two ways:
by improving LEAP’s accessibility to users, and by making
the process of developing and using LEAP playful, fun and
enjoyable. The ?ndings reveal how the engagement of
organizational members, as facilitated by both accessibility
and playfulness, was a key part of making LEAP expressive,
and, as such, forms a key part of the expressive role of PMS.
Accessibility
Making the PMS accessible to staff, volunteers and
partner representatives was particularly challenging. The
Health Programme Manager described a typical view of
monitoring and evaluation processes as follows:
I think there’s a lot of mystique around M&E [monitor-
ing and evaluation] which perhaps doesn’t necessarily
have to be there. I think people think ‘oh, M&E is some-
thing which clever people that crunch data and statis-
tics and things do.’ [. . .] I think providing people with
systems whereby they can actually do it very simply
is important but I think that can be a bit of a challenge
sometimes.
This ‘mystique’ was addressed by designing LEAP so it
was understandable to people without a background in
monitoring and evaluation. This was signi?cant because
the vast majority of people involved in the mental health
programme, including the volunteers, did not have such a
background. LEAP avoided jargon, used non-technical
language and expressed indicators so they related directly
to the activities of the mental health programme (see
Appendix 2).
A further way accessibility was promoted was by ensur-
ing LEAP was ‘as simple as possible.’
16
This was particularly
important because volunteers and staff who came to work
on the mental health project held somewhat negative views
about the use of PMS. The Programme Of?cer explained:
The majority of volunteers. . .are psychiatric nurses and
occupational therapists and psychologists and psychiat-
ric social workers. They are health professionals and
when they see this kind of a tool [LEAP], if they don’t
have a lot of background in monitoring and evaluation
and project management and so on [. . .] I think they
don’t want to take it up.
This quote highlights the way in which the volunteers’
motivation for working on the mental health programme
was driven by their work as health professionals where
tools like LEAP are viewed as a potential obstacle to their
work. In this context PMS are viewed as a distraction or
even an incursion on much more valuable activities; a sit-
uation observed in the wider NGO sector (e.g., Chenhall
et al., 2010).
The importance of LEAP being as simple and less oner-
ous as possible was evident in the following excerpt, taken
from a meeting in August 2009, where volunteers contrast
the ?rst and ?nal drafts of LEAP:
Volunteer 2: When this [LEAP] was ?rst shown to us
some months ago it was like, my god, what on earth
has VSO thrown up [. . .] this is too big [. . .] I feel this
[the revised version] is workable [. . .]
Volunteer 1: We’ve listened to your feedback.
Volunteer 2: It’s not just that, it’s that, I’m a volunteer,
I’m surrounded by paper work more than I was in the
UK [. . .] Finally I understand the value of reporting for
the future [. . .] it’s more user friendly for me, so a round
of applause for the LEAP team, yeah!
[General applause]
Volunteer 7: And also like the guff that we end up ?lling
in in England, you know, I’ve not been in the statutory
sector in England [the NHS) very long but I’ve never
yet had any input into it at all. You know, you ring up,
Fig. 1. Timeline of PMS developments.
16
Source: Draft 1 of LEAP reporting template.
8 R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
I can’t even remember the name of the wretched data-
base team, about why something doesn’t work, and it’s
just like, well, you can’t do it like that, you can’t put in
the information you want, you’ve got to put in the infor-
mation we want, you know, and to have something
that’s what we actually need and is useful and is things
that we’re doing and talking about that is rather
refreshing because that’s not been my experience at
home.
Volunteer 5: I think the thing I like about it is, I ?nd
paperwork absolutely demoralizing, but this is less of
an imposition.
17
As with the Excel spreadsheet completed in August
2008, the ?rst version of LEAP was criticized for being
too big and complicated. In contrast, the ?nal version of
LEAP, which was the result of extensive efforts to simplify
and make it clearer, is viewed as user-friendly and more
workable. The lack of user input Volunteer 7 perceives in
the NHS (‘you can’t do it like that’) is clearly a cause of
much frustration, evident in her view of the operators as
the ‘wretched database team.’ In effect, by (seemingly)
not even considering the views of the volunteer, this lack
of user input can be seen as limiting and controlling the
types of values and beliefs that can be displayed and
included in the PMS (c.f. Huy, 1999). In contrast to the
NHS case, she views her involvement in LEAP as ‘refresh-
ing’ and considers its content as related directly to her
work (‘things that we’re doing and talking about’), evident
in another conversation where Volunteer 7 stated that
LEAP is ‘actually rooted in our work.’
18
This is further sup-
ported by the statement of Volunteer 5 who describes LEAP
as ‘less of an imposition’, revealing the way in which it could
bring a sense of hope and encouragement (rather than
‘demoralizing’) by not encroaching on the work the volun-
teers came to Sri Lanka to pursue.
Playfulness
Games and other techniques to encourage a sense of
playfulness were also a feature of the LEAP development
process. One example is from a pivotal moment in the
development of LEAP that occurred at a quarterly Sector
Group Meeting in July 2009. At this meeting, which
involved VSO staff, volunteers and representatives of part-
ner organization, LEAP was of?cially introduced for the
?rst time. This did not involve a formal presentation, but
was organized around a brief introduction followed by a
game. At the very start of the activity, all 50 participants
were asked to stand up and, then, on the count of three,
to jump in the air and shout ‘LEAP.’ This prompted much
laughter from participants.
The game used to introduce LEAP was called ‘Making
Rice’, and was the suggestion of Volunteer 1 who had been
exposed to a similar game in her volunteer placement. The
game involved participants being organized into groups of
approximately ?ve people, and then being provided with
20 min to answer ?ve questions concerning the making
of rice, where each question was designed to mimic a stage
in a typical evaluation process, such as monitoring, evalu-
ation and impact.
19
Responses from each group were col-
lected and placed on a ?ipchart, and the session ended
with analogies being drawn between the stages in the eval-
uation of making rice and the stages in the evaluation of the
mental health programme.
The game itself prompted much discussion, joking and
laughter throughout and was also a source of amusement
at several moments later in the meeting – for example,
one participant joked that the researcher and Volunteer 1
should conduct a rice tour to research how to cook the best
rice and that this should involve its own evaluation. That
these jokes were made at all indicates LEAP was indeed
perceived as accessible and non-threatening by partici-
pants, which is important in enabling the expression of
beliefs and values (Huy, 1999). The ability of the game to
generate a sense of playfulness relates, at least in part, to
the selection of a context that was culturally and cogni-
tively appropriate. Rice is a staple food in Sri Lanka and
is often used to celebrate important events – hence select-
ing this as the context can be seen to symbolize the signif-
icance of the LEAP process. In addition, as participants
would actually have experienced the cooking of rice them-
selves, they were all able to participate in the activity,
helping to reduce mystique or feelings of exclusion from
evaluation activities such as LEAP.
Several other aspects of the process sought to generate
encouragement and a sense of hope. For example, the ‘LEAP’
acronym was chosen deliberately to symbolize the way in
which evaluation can help to generate improvements in
the lives of persons with mental illness, which was rein-
forced through the selection of an image attempting to cap-
ture visually the progress the project was trying to achieve,
as shown in Fig. 2. The use of such images can encourage
hope and motivation by giving ?esh to the vision of a pro-
ject or goal (Huy, 1999; Smith & Stewart, 2011).
LEAP
Where are
we now?
How do we
get here?
Challenges?
Fig. 2. Diagram used in presentation at Sector Group Meeting and on
cover page of LEAP reporting template and all guidance documents.
17
Source: LEAP meeting, August 2009.
18
Source: Field notes, July 2009.
19
For example, question 2 was ‘What do you do to check if the rice is
cooking properly?’, which referred to the monitoring stage, and question 4
was ‘What is the result of well-cooked rice?’, which portrayed the impact
stage.
R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx 9
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
Playfulness was also evident in the informal nature of
the discussions held with staff (e.g., doctors, nurses) at
each mental health facility in the process of completing
the LEAP reports. For example, these discussions would
begin with an ‘ice-breaker’ activity, the discussion space
was designed so that participants sat in a circle without
any tables and chairs between them, and VSO volunteers
were intermingled amongst staff rather than in separate
groups. Volunteer 6 made the following comment about
these discussions:
It’s [the LEAP discussions] not like a serious thing [. . .]
like in an exam where you have to give an answer, peo-
ple are able to tell a joke and say, you know, what they
feel like and they are able to express themselves freely.
These informal arrangements provided a relatively safe
and protective environment in which staff could share
ideas and have the freedom to participate in the discussion
and answer questions (Huy, 1999).
In this section we examined how the engagement of
staff, volunteers and others in the performance measure-
ment process was facilitated by making LEAP accessible
to users (e.g., using simplifying language for indicators)
and by creating a sense of playfulness (e.g., the use of the
‘making rice’ game) when developing the LEAP template
and throughout its completion at each mental health facil-
ity. Accessibility and playfulness were both important in
engaging organizational members in the development
and operation of LEAP and thus formed a key part of the
expressive role of PMS. The involvement and engagement
of volunteers and staff also helped to facilitate the inclu-
sion of indicators in LEAP focused on their values and
beliefs, primarily related to the importance of the bene?-
ciary. In the next section we focus on the way in which
the indicators that were included in LEAP helped organiza-
tional members to express their values and beliefs through
the PMS.
Expressing values and beliefs through the PMS
In this section we ?rst examine the way in which the
expressive role of PMS was helped by the inclusion of a
focus on bene?ciaries in LEAP. We then focus on tensions
that arose in enabling the expressive role in both the
design of, and scoring process for, the indicators in LEAP.
Focusing on the bene?ciary
A strong belief in the importance of the bene?ciary was
evident throughout the mental health programme. For
example, programme documentation regularly expressed
the importance of a rights-based approach to mental
health and of having mental health services centred on
clients and service users.
20
Volunteers often expressed the
belief that their motivation for volunteering was to ‘make
a difference’ in the lives of staff and patients they were
working with in their placements.
21
The Programme
Manager stated the programme was ‘ultimately about the
bene?ciary group, this is why we are here.’
22
A belief in
the importance of bene?ciaries was also evident in changes
to VSO’s approach to its work, with its increased focus on
‘impact’, that is, whether and how the work of VSO contrib-
uted to changes at the bene?ciary level (see, for example,
VSO, 2004).
The Volunteer Report was viewed as ‘quite good at
looking at activity but less strong in terms of [capturing]
overall impact.’
23
The programme manager commented that
the existing volunteer reports were ‘still placement speci?c’
and, in the context of developing LEAP, she stated a PMS
should enable her and others ‘to see the bene?ts [of the pro-
gramme] for them [the bene?ciaries].’
24
Volunteers also
expressed frustration at not being able to see how their work
was having an impact.
25
As such, the existing Volunteer
Reports did not provide volunteers and staff with the free-
dom to display and express the importance of the bene?-
ciary in their work (c.f., Huy, 1999). In contrast to the
Volunteer Report, the ?rst (and subsequent) drafts of LEAP
sought to include a stronger focus on bene?ciaries. In partic-
ular, indicators were developed concerning patient behavior
(such as whether patients provide input into their treat-
ment), as well as a separate section on ‘the impact of mental
health services on the lives of bene?ciaries’, which included
indicators such as the length of stay, readmissions, commu-
nity visits, and discharges (see Appendix 2 for further detail).
The inclusion of indicators focused on bene?ciaries in
LEAP helped to express values and beliefs by prompting
discussion of values amongst staff and volunteers during
the completion of LEAP at the different mental health facil-
ities. This was because completing LEAP typically involved
volunteers and staff from the respective mental health
facility meeting in face to face discussions to review
progress and complete the scoring for each of the indica-
tors.
26
As many of the indicators in LEAP concerned bene?-
ciaries, this meant most of the time spent completing LEAP
related to discussions concerning patients, providing a set-
ting in which staff and volunteers could discuss and re?ect
on values related to bene?ciary care. The following four
examples serve to illustrate this process: the ?rst two exam-
ples show how LEAP prompted organizational members to
recall particularly impressive activities and outcomes for
patients, whereas the third and fourth examples show how
LEAP provided opportunities for organizational members
20
Source: End of Project documentary (see www.youtube.com/
watch?v=UZcQ9a0bLm0&feature=youtu.bevideo, accessed 14 March
2013) and End of Project report entitled ‘Sharing skills and experiences:
good practices in mental health’, seehttp://www.vsointernational.org/
Images/sharing-skills-and-experiences-good-practices-in-mental-health-
sri-lanka_tcm76-38911.pdf, accessed 16 March 2013).
21
Additional motivations expressed by volunteers related to the chance
to experience life in a different country, to travel and meet new people, and
to gain experience in the international development sector with a view to
gaining employment after their placement (Source: Field notes).
22
Source: Field notes, April 2009.
23
Source: Field notes, April 2009.
24
Source: Field notes, April 2009.
25
Source: Field notes, January and April 2009.
26
These discussions typically lasted between a half-day and one and a
half days and could involve discussions with different groups of staff and
patients at each mental health facility. For example, those involved in
completing LEAP could speak to doctors, nurses, social workers, support
staff and patients in separate groups. Each group discussion could take at
least one hour. Source: LEAP guidance document and ?eld notes.
10 R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
to display and register failures and disappointments as they
relate to patient progress and treatment.
The ?rst example concerns the completion of LEAP at a
large mental health hospital in Colombo. This involved
three volunteers, one hospital staff member and the
researcher in a discussion lasting approximately two
hours. During the discussion of the indicators in Section
2, particularly the indicator on violence and aggression
training, the hospital staff member was prompted to offer
the following re?ection:
So the big thing is that we have a target to train all the
staff in the management of violence and aggression, we
have trained nearly half [of the staff]. . .now we can see
in the hospital, there is a big difference with the staff,
now no violence, the violence has stopped, it has
reduced, it’s very very nice [. . .] When I came to this
hospital in 1998, every staff, the minor staff, had a big
stick to attack the patients. . . now no sticks. . .It chan-
ged, it changed.
The comments it is a ‘big thing’ and a ‘big difference’
illustrates how LEAP provided an opportunity for the staff
member to express the importance of the change in the
behavior of staff he had observed. Discussing the indicator
on violence and aggression training allowed himto register
the signi?cance of the change since his time at the hospital,
particularly as he, along with the volunteers, had been lar-
gely responsible for developing and operating the training
for staff in violence and aggression management that had
helped to change the patient experience at the hospital.
A second example concerns the completion of LEAP at a
women’s mental health facility located in Colombo. This
involved three staff at the facility (the matron, a doctor
and a nurse), a volunteer, and the researcher in a discus-
sion prompted by the cover page of the LEAP report, which
asked participants to identify and record the ‘three biggest
achievements’, the ‘three biggest challenges’ and the ‘three
biggest priorities for the coming year.’
27
Many achieve-
ments were raised during this discussion including, as
above, changes in the way staff were acting toward patients.
The doctor was the most vocal in expressing the signi?cance
of the changes and in particular re?ected on the importance
to her of changes in the behavior of patients she had
observed. Speci?cally, she commented the patients now
‘have ambitions. . .enjoy themselves’ and ‘they have wishes
and preferences.’ She further commented that in the past
the patients were ‘institutionalized to a high degree’ but
now had begun to show ‘more expressions’ with ‘even acute
patients saying hello.’
28
As the indicators in LEAP were
focused on changes in the experiences of bene?ciaries, it
prompts the doctor to register and express the importance
of the positive changes in the behavior of the patients she
had witnessed. By providing opportunities to re?ect on sig-
ni?cant achievements they had observed and/or were
involved in, LEAP helped to provide for uplifting moments
where participants could experience a sense of hope and
encouragement (c.f., Huy, 1999). It also provided a setting
in which values related to the importance of bene?ciary care
were espoused and discussed amongst staff at the hospital,
particularly values concerning the treatment of patients as
individuals with their own wishes and preferences. These
two examples also show that this is possible because the
LEAP process actually involved the staff at each facility in
completing LEAP thereby providing them with the opportu-
nity to express their values and beliefs.
Along with the ability to highlight achievements, LEAP
provided participants with the chance to display disap-
pointments and failures related to patient treatment and
progress. The third example occurred during the comple-
tion of LEAP at a rehabilitation centre located in a central
province of Sri Lanka. This involved a staff member, a vol-
unteer, and the researcher in a discussion lasting approxi-
mately three hours. The story of Marsha, which Volunteer
7 relates below, arose during discussions over the comple-
tion of the indicators in Section 4 of LEAP related to
changes to the lives of bene?ciaries:
Marsha was a really bright young woman, in her early
thirties, with a diagnosis of bipolar affective disorder,
a history of mental illness in the family [. . .] she’d been
here [to the rehabilitation centre] three times [. . .], she
was chief editor of the little newsletter that we’ve got
going [at the rehabilitation centre], she was always up
for speaking at a meeting or singing or anything. She’s
always interested, a real bright spark, and I really
wanted her to go and learn something or do something
[. . .] but it wasn’t to be [. . .] her father was fretting and
he rolled up one day [. . .] the house in Colombo he’d
exchanged for lifetime care of Marsha and her brother
in this private long stay institution [. . .] I said ‘what’s
it like? Do they go out there?’ And he [Marsha’s father]
said ‘oh yes, they go out once a year.’ [. . .] It was com-
pletely crap [. . .] It was awful, it was awful.
In recalling this experience Volunteer 7 is able to dis-
play her fondness and affection for Marsha (‘she was a real
bright spark’) and express her sadness and disappointment
at her being taken away to a long stay institution by her
father (‘it was awful, it was awful’). This shows how the
LEAP process prompted the volunteer to recall a signi?cant
moment from her work where, in this instance, a patient’s
life had changed for the worse. This story was particularly
meaningful because a key aim of the mental health pro-
gramme was to move mental health care away from the
type of ‘long stay institution’ Marsha had been taken to
and toward the rehabilitation model of care available at
the rehabilitation centre. LEAP provided a setting where
the values of staff and volunteers concerning the funda-
mental importance of the potential rehabilitation of
patients (rather than a lifetime of institutionalization)
could be expressed and discussed. The opportunity to
recall a signi?cant event was helped by Volunteer 7 being
provided suf?cient time and space to actually tell the story,
which was exempli?ed when she commented toward the
end of telling the Marsha story that she was ‘ranting’, but
was reassured by those present that the recalling of the
experience was valuable and that she should continue.
Importantly, the telling of this story was not strictly
27
Source: LEAP Reporting template document.
28
Source: Field notes.
R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx 11
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
necessary for scoring the indicators in Section 4 and com-
pleting the LEAP report.
The fourth example occurred during a three hour meet-
ing with the mental health volunteers, VSO mental health
programme staff and the researcher during July 2009.
The purpose of the meeting was to discuss the near-?nal
draft of the LEAP report, and to review each indicator and
garner feedback and any ?nal suggestions for change. The
following statement from Volunteer 2 arose during the
review of indicator 2.3, which concerns whether staff at
the mental health facilities informed patients about their
diagnosis, treatment and side effects of medication:
They [the hospital staff] don’t discuss anything with the
clients, at [a hospital] they’re [the patients] all schizo-
phrenics as far as they’re [the staff] concerned [. . .] if
they’re [staff] handing out tablets on the ward the
patient should be at least reminded ‘this is your epi-
lepsy tablet’ [. . .] they’re [the staff] not doing that, they
[the patients] just have two pink ones [tablets], and a
yellow one and a green one, and off you [the patient]
go. . .
Echoing the sentiment of those present during this dis-
cussion, Volunteer 2 recalls how current practices at the
mental health facilities do not involve the type of staff
behavior volunteers believe should be occurring. As with
Marsha’s story, discussing treatment with patients, think-
ing of them as more than a diagnosis (not ‘all schizophren-
ics’) and giving patients information regarding their
medication (‘this is your epilepsy tablet’) were central pil-
lars of the approach to mental health care volunteers and
others working in the programme valued and believed in
very strongly. Importantly, it is the review of, and subse-
quent discussion surrounding, indicator 2.3 that prompts
the volunteer to recall this experience. This allows her
and others present to express the disappointment of the
current practices they had been observing in their volun-
teer placements and to discuss how this connects to their
values surrounding proper patient care and treatment.
Designing indicators to express values and beliefs
In this section we examine how the ability of a PMS to
enable the expression of values and beliefs can depend, at
least in part, on the way in which particular indicators are
seen to capture and relate to sometimes quite speci?c val-
ues and beliefs. We present three examples to illustrate
this process.
One example of where the design of the indicators in
LEAP was able to capture speci?c values and beliefs con-
cerned the development of indicator 1.3 (focused on keep-
ing records in the patient ?le). The following excerpt, from
the above-mentioned three-hour LEAP meeting in July
2009, arose in a discussion regarding indicator 2.2 on the
extent to which quali?ed staff actively seek out, listen
and respond to the preferences of patients:
Volunteer 8: Could it [indicator 2.2] be something along
the lines of actively seek out, listen and record?
Volunteer 7: Well, except that recording is maybe a
different one [indicator] and maybe that if we’re
interested in recording then maybe we need that as a
separate indicator of some kind [. . .]
Volunteer 8: I mean just something basic, like each
patient has a ?le is a start [. . .]
Programme Manager: But what they record is not really
relevant or appropriate information, it’s just basic stuff
on medical stuff, but not on social elements, [. . .]
Volunteer 2: And even that is very badly recorded [. . .]
Volunteer 8: You know there’s a few people [patients]
inevitably that are regularly upset or whatever but little
of that goes into the record, all that goes into the record
is the extra dose of tracomazine [a drug].
Volunteer 3: Do you need to add another indicator here
about recording?
Researcher: Each patient has a ?le with basic informa-
tion? [. . .]
Programme Manager: In some places they do but that
won’t give us the indicator that is progression, my con-
cern is [. . .] it’s just basic information based on medical
diagnosis just now we’re talking about their life situa-
tion, their priorities, their preferences and family situa-
tion and the psychiatric social workers also need to be
included so if we are putting an indicator we need to
make it something that’s really capturing all elements
of the patient’s life and not just basic information.
The discussion of indicator 2.2 prompts the volunteers
to re?ect on values related to appropriate forms of care
and treatment for patients. In particular, Volunteer 7 sug-
gests that an additional indicator concerning the recording
of information about patients might be required (as none
of the indicators in LEAP at that time captured recording
of patient information). Importantly, the discussion is
focused not only on the absence of record keeping per se,
but that the type of record keeping the indicator would
attempt to measure needed to be consistent with the spe-
ci?c values and beliefs espoused by the volunteers and
staff on the mental health programme. A core value of vol-
unteers and staff concerned shifting practice away from
the ‘medical stuff’ and to concentrate attention on the ‘life
situation’ of each patient. In the context of record keeping
in the patient ?le this meant that a variety of information
re?ecting the patients’ experiences should be recorded, not
just the medical information such as the ‘dose of tracom-
azine.’ This was considered by staff and volunteers to be
a central tenet of the approach they believed in – to think
of, and therefore treat, mental health patients as a human
being, rather than as a medical diagnosis. LEAP thus pro-
vided a setting where they could discuss values concerning
the treatment of patients and how best to re?ect these val-
ues in the indicators (e.g., whether there is a patient ?le vs.
a ?le capturing holistic patient information). The indicator
developed from this discussion, which assessed whether
the patient ?le included demographic, assessment, treat-
ment and progress information from different staff (see
Indicator 1.3 in Appendix 2), was seen to capture and thus
express these beliefs and values appropriately.
A further example relates to the development of an
indicator to capture the extent to which patients have a
say in their care and treatment. During the same meeting
in July 2009, volunteers reviewed the indicators used in
12 R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
Section 3 on patient experiences, and, as above, felt the
existing indicators were insuf?cient:
Volunteer 7: You’d [the patient] be having a bit more
choice about lots of things. There’s something about
patients [whereby they] don’t feel they’ve got a right
to say anything. It’s kind of, ‘‘I’m a mental patient, what
do I know?’’ kind of a notion that is reinforced by the
service [at the mental health facility]
Volunteer 8: Maybe something [for the indicator]
around patients actually feel[ing] able to state what
their needs are.
Volunteer 7: Yes.
Volunteer 8: Yeah, being able to state what they want.
Volunteer 7: Yeah [. . .]
Volunteer 2: Which is breaking that culture of long
institutional care in different organizations, isn’t it?
[. . .]
Volunteer 8: So there’s something also about patients
feeling able to refuse. . .
Volunteer 7: To say what they want and what they
don’t want [. . .]
Volunteer 5: For example, [the patients are] expected to
stay in bed whether or not they are asleep.
Volunteer 7: So I think that expresses it really well,
what they want and what they don’t want.
In this excerpt, the volunteers are able to discuss and
directly engage with values relating to the importance of
treating patients as having their own individual preferences
rather than treatment being driven by the dictates or pref-
erences of the mental health care facility. This is evident
in how the volunteers comment that the indicator should
capture whether patients are ‘able to state what their needs
are’, ‘what they want’ and being ‘able to refuse.’ This leads to
a suggestion that an indicator be developed to capture the
extent to which patients are able to say what they want
and what they don’t want, which subsequently became
indicator 3.3. That indicator 3.3 used the precise wording
as suggested by the volunteers meant it was able to express
‘really well’ volunteers’ values concerning more extensive
involvement of patients in their own care and treatment.
In contrast to the previous two examples, the indicators
related to community-based mental health care provide an
illustration of where the design of the indicators made it
more dif?cult for LEAP to function in an expressive role.
LEAP included a section on ‘Engagement in the community’
(see Section 4.2 in Appendix 2) that consisted of indicators
to address visits to the community by patients, and any
employment, training and work activities outside of the
hospital/rehabilitation centre. These indicators were
viewed as problematic, however, because they were seen
to promote a conception of mental health care at odds with
volunteers’ values and beliefs. For example, Volunteer 1
commented that the indicators in Section 4 should be
framed in a way that helped ‘government services to think
beyond day trips to the community’,
29
indicating they were
lacking in their ability to help volunteers to express to others
their beliefs about the importance of particular forms of
community-based mental health care.
These indicators were also criticized because they
focused on the activities of patients who remained in hos-
pital and not on those patients who had been rehabilitated.
This issue was raised during a meeting with volunteers in
July 2009:
Volunteer 7: If the bene?ciaries have really bene?ted
they might not be in the institution any more.
Volunteer 8: How are you going to follow that up?
Volunteer 2: If you capture this [patient experiences in
Section 3] they’ll always be there. The statistics on the
other one [Section 4], you might catch them [going] in
and out [of the hospital] but the effectiveness is not
going to be there.
Volunteer 8: And that quality.
Volunteer 2: Yeah, that’s the essence, yeah.
30
In this way the indicators are viewed as focusing on
patients who are ‘in’ the facility, and as there was no mech-
anism to ‘follow up’ patients who had left, volunteers were
unable to express the importance of focusing on patients
who had ‘really bene?ted.’ Furthermore, it shows that the
indicators are lacking in their ability to provide a sense of
hope and encouragement by excluding the very activities
seen by volunteers to be the ‘essence’ of the project.
However, the collection of data on community mental
health services would have been very expensive and
time-consuming, involving travel to different areas in the
province to collect the necessary data and follow up with
patients. This was because ‘volunteers didn’t have the
knowledge to complete’ such information, evident in such
information being missing from the Excel spreadsheet.
31
This example shows how although it was more ef?cient
not to develop indicators addressing community mental
health services in a more comprehensive manner, this made
it dif?cult for those indicators to enable organizational
members to express their values and beliefs about the
importance of community-based care.
Scoring the indicators
From above we saw that scoring the indicators and
completing the LEAP report provided a setting where
organizational members could discuss values concerning
the treatment of patients, which was particularly evident
in discussions over signi?cant achievements and setbacks
in attempts to improve patient care. Below we examine
two examples where the scoring process is more prob-
lematic. The ?rst example concerns the ability of the
indicators to capture variations in performance signi?cant
to organizational members. The second example relates
to the involvement of patients in the scoring process.
Both examples also illustrate how the scoring process
for the indicators can result in a situation where the
29
Source: comment from Volunteer 1 on 3rd draft of LEAP reporting
template, May 2009.
30
LEAP meeting, July 2009.
31
Source: Mental Health project baseline spreadsheets, August 2008.
R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx 13
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
instrumental and expressive roles of PMS appear
irreconcilable.
The following statement is from Volunteer 5 and
occurred during the above-mentioned meeting to com-
plete LEAP at a large mental health hospital in Colombo.
The statement was prompted during discussions focused
on scoring indicator 1.6 – the extent to which patients have
a say in their treatment:
On [the] forensic [ward], there are very few care plans
but there are some, and some of those do allow people
to make some choices, one man for instance, on the care
plan we did this week, had wanted to read a newspaper.
They have no newspapers up there [in the forensic
ward], they get ward three’s second hand ones [. . .]
they’re planning now to setup a reading group so that
people don’t smoke the newspapers, and all of that, so
that one man did have a choice [. . .] I don’t know where
you put that [. . .] there was another [patient] actually
who wanted to work in horticulture and it was agreed
that he could but of course the system doesn’t allow
for that ‘cause the walls aren’t high enough and the
fence isn’t ?nished so I don’t know quite where you’d
put these little nuggets of progress like that because
there’s kind of some choice.
Volunteer 5 relates examples of the development of
care plans, whereby two patients on the forensic ward
were given opportunities to engage in activities they indi-
vidually preferred, such as ‘to read a newspaper.’ Recall,
however, that the scoring of the indicators required an
assessment of the extent to which all patients are given
opportunities to have a say in their treatment and to
assign a score on the 5-point scale. As opportunities were
only being given to a few patients, the indicator was
assigned a low score, prompting Volunteer 5 to query
how the indicator would capture these ‘little nuggets of
progress.’ It is important to note that this is not merely
a concern over an incomplete or ill-speci?ed indicator,
as it was the recognition and celebration of these nuggets
of progress that gave meaning to the volunteers’ work
and gave the hope and encouragement (c.f., Huy, 1999)
necessary to sustain them through what were often extre-
mely dif?cult and trying circumstances where progress of
any kind was a rare commodity. This was particularly
acute on the forensic ward, which was the area of the
hospital identi?ed by volunteers as being in most need
of improvement.
32
This example shows how in this situation the sensitivity
of the scoring system used in LEAP was not able to capture
and express the signi?cance of the change in practice
Volunteer 5 had observed on the forensic ward. In this
example it resulted in a meaningful discrepancy between
the value the volunteer attributed to the changed practice
and the way the score on indicator 1.6 represented
that change, and meant the indicator did not help
Volunteer 5 to express her values and beliefs. Although
throughout the study we saw little further evidence of
such meaningful discrepancies, this example does suggest
that the sensitivity of indicators in expressing the meaning
of activities and achievements is likely to be very impor-
tant in enabling the expressive role of PMS.
The potential for tension between the instrumental and
expressive roles of PMS is further highlighted by the
involvement of patients in the scoring process. Patients
were involved by asking them a series of questions about
their experiences in the ward and/or rehabilitation centre
and then using their responses to help determine the most
appropriate score for each indicator (primarily indicators
1.4–1.6 and 3.1–3.3). Staff and volunteers expressed the
importance of involving patients directly in the process
not only because the programme was strongly focused on
patient involvement per se, but also because of a desire
to produce relatively accurate scores. That is, volunteers
and programme staff expressed concern that staff at
the mental health facilities may present a somewhat
optimistic picture of the current level of performance and
this would then overstate the performance of the pro-
gramme as a whole.
33
Patient input was therefore viewed
as a way to ‘correct’ for this potential optimism. However,
this was problematic as illustrated by a situation that arose
when completing LEAP at a rehabilitation centre in August
2009.
In the course of asking patients about their experiences,
one patient asked whether she would be able to get a job
and wanted to know what the future would hold for her.
As providing answers to these questions was well beyond
the remit of LEAP, Volunteer 1 stated involvement was
not ‘meaningful enough for the patients’ and it was
primarily directed at ‘getting what we want’, that is, the
information to complete the LEAP report.
34
In this way,
the involvement of patients does not register the signi?-
cance of being able to do something for patients who found
themselves in a troubling situation – in effect, it was viewed
as an information gathering exercise at odds with the volun-
teer’s beliefs about the importance of helping patients. This
left Volunteer 1 feeling frustrated and pessimistic about
being able to achieve meaningful change for patients (c.f.,
Huy, 1999). Thus, there were tensions associated with the
involvement of patients in completing LEAP – on the one
hand it was considered important to help develop accurate
scores, on the other hand it caused distress and made it dif-
?cult to express the importance of helping patients in a
meaningful way. The frustration and distress experienced
by Volunteer 1 highlights the potential for the con?ict
between the expressive and instrumental roles to have
adverse effects on staff and volunteers.
Collection and use of LEAP information
The ?nal version of LEAP was ?nalized in late July 2009,
after which it was completed at each mental health facility
where VSO had placed a volunteer. By September 2009, the
LEAP process had produced a large quantity of perfor-
mance information that had not existed previously. This
32
Source: The LEAP report for the hospital identi?ed the forensic ward as
the biggest priority for the coming year.
33
Source: Field notes and interviews with volunteers (2009 and 2013).
34
Source: Field notes, August 2009.
14 R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
information consisted of the scores for each of the
performance indicators for each mental health facility, an
extensive set of comments related to performance on each
indicator, as well as written details of plans that were
underway to improve performance.
The collection of quantitative performance indicators
meant that, for the ?rst time, reports were produced that
aggregated performance on each indicator across the
mental health programme. This had the effect of providing
performance information considered helpful by organiza-
tional members. In particular, the Programme Manager
highlighted the limitations of the volunteer report, stating
that the information from that report meant ‘we’ve never
been able to measure change and impact.’ In contrast,
she made the following statement about the inclusion of
indicators related to bene?ciaries in LEAP during a meeting
in September 2009:
This [LEAP] is de?nitely useful for me when you’re look-
ing at changes and making sure we make an impact [. . .]
that’s what’s been missing all this time and even though
we’ve done the programme area review [. . .] this [lack
of information] is why I’ve been struggling, the fact that
we don’t have baselines, we’ve never been able to mea-
sure change and impact.
This statement indicates how LEAP allows the
Programme Manager to look at changes and impact in
two ways. First, because LEAP collected information using
indicators that were standardized across mental health
facilities, it meant she could examine differences and
changes between the different facilities using common
indicators. Second, as noted, many of the indicators in LEAP
related to bene?ciary progress, thus providing information
on the impact of the mental health programme that had
been lacking under prior reporting systems. Echoing this
sentiment, VSO’s Health Programme Manager in London
stated that the focus on bene?ciaries in LEAP was the ‘most
advanced I have yet come across in any health programme’
and, given this, he believed it should serve as a model for
VSO’s other health programmes.
35
Discussion and conclusions
This study has examined the expressive role of PMS,
which we conceptualized as the capacity of a PMS to
facilitate the display of a variety of values and beliefs held
by organizational members. As such, our study contributes
to the nascent body of literature considering organiza-
tional members’ values, beliefs and emotions at work as
potentially productive elements of the work environment
(Boedker & Chua, 2013; Bolton & Houlihan, 2009; Dutton
et al., 2006; Huy, 1999). Additionally, with its focus on
how PMS can play an active role in helping organizational
members to express their beliefs and values as part of
their work, our study presents a counterpoint to prior
MCS research that conventionally considers values as
something to be expressed through belief system mecha-
nisms like mission, vision and values statements, and top
management communications by way of the interactive
use of accounting systems (e.g. Simons, 1995). In contrast,
our focus is on how the PMS are implicated in the commu-
nication, discussion and expression of values and beliefs by
organizational members.
Through our analysis of the development of LEAP at
VSO, it emerged that a PMS can provide a mechanism
through which the values and beliefs of organizational
members can be expressed as part of their work. We saw
how LEAP helped staff and volunteers to express and dis-
cuss the importance of the bene?ciary in meetings where
the indicators were developed and re?ned, during comple-
tion of LEAP at the various mental health facilities, and in
discussions following the completion and collation of the
LEAP data. In this way LEAP provided organizational mem-
bers with settings in which they could discuss values con-
cerning patient treatment and express the signi?cance of a
variety of events related to changes toward the type of
patient care they valued, which included both distressing
episodes and moments of progress and success. Our analy-
sis also sought to illuminate the design and operational
characteristics that would help a PMS to ful?l an expres-
sive role, which we discuss in the following sections.
Playfulness and accessibility
LEAP was characterized by trial and error and experi-
mentation (c.f. Wouters & Roijmans, 2011; Wouters &
Wilderom, 2008), developed through the use of games,
imagery, and fun to promote the capacity of participants
to feel comfortable expressing their values and beliefs in
a supportive environment. We identi?ed ‘playfulness’ –
the organization’s ability to create an environment in
which mistakes are tolerated and experimentation is
encouraged – as being an important mechanism for
enhancing the expression of values and beliefs in the
PMS. Key to this was the use of games to allow organiza-
tional members to share ideas relating to their beliefs in
a way that was non-threatening, free from judgement
and fear of reprisal, and having a high level of informality
and ?exibility associated with the PMS. In particular,
LEAP operated in such a way that there was the opportu-
nity for organizational members to have the time and
space to re?ect, experiment and offer up new ideas,
which supported the display of their values and beliefs
(Huy, 1999).
LEAP was characterized by the use of non-technical
language, the avoidance of jargon, and the expression of
indicators in a way that was ‘meaningful’ and understand-
able to organizational members (c.f. Rowe et al., 2008). It
was important to adapt the level of sophistication of the
PMS and the technical expertise required so it was accessi-
ble to users and allowed them to engage with the PMS. The
fact that the suggestions of organizational members were
used in developing the measures in the ?rst instance sup-
ported this process, re?ecting an evaluation process using
techniques and concepts readily understood by partici-
pants without the need for training (Hall, 2014). This is
in contrast to common monitoring and evaluation tools
35
Source: Email from Programme Development Advisor, Health, VSO
London, to Mental Health Programme Manager, VSO Sri Lanka, November
2009.
R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx 15
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
used in international development, such as the Logical
Framework,
36
which are replete with jargon and technical
terms that are often dif?cult to comprehend without any
prior training and exposure (see for example, Gasper,
2000; Wallace, Bornstein, & Chapman, 2007).
Overall, our analysis shows that engaging organiza-
tional members in the PMS is important for the expressive
role. This is because for a PMS to help the expression of
beliefs and values, it must actually involve people in mean-
ingful action of some kind, whether that is developing indi-
cators, providing feedback on PMS design, or considering
the implications of PMS information. Passivity or lack of
engagement provides little opportunity for the PMS to
enable staff and others to express their values and beliefs
in a meaningful way. In this way, engaging people with
the PMS re?ects a particular kind of participation where
organizational members are actively involved in negotiat-
ing, developing and, later, operating the PMS. In contrast,
engagement that is a mere ‘feel good’ exercise, akin to a
‘sham ritual’ (Ebrahim, 2003; Najam, 1996), is unlikely to
promote the expressive role of PMS.
Design of indicators
Our study showed how the content of the performance
indicators was important in enabling the expressive role of
PMS. The variety of indicators that focused on bene?ciaries
enabled staff and volunteers to express their beliefs
regarding the importance of bene?ciaries by prompting
them to discuss and re?ect on patient progress with other
staff and volunteers. These indicators provided a prompt
for volunteers and staff to have discussions about patients
– it enabled them to express their feelings and beliefs
about what has occurred, which, can be moments of
achievement and jubilation, along with particularly dis-
tressing episodes.
The opportunity to discuss both positive and negative
bene?ciary outcomes (e.g., a volunteer relaying Martha’s
story) also meant volunteers and staff were encouraged
to express a wide range of experiences and emotions with-
out fear of reprisal (c.f. Huy, 1999). This is in contrast to the
typical situation in NGOs where there is often pressure to
report only success stories (e.g., Lewis, 2006). More gener-
ally, this suggests that the capacity of a PMS to ful?l an
expressive role is likely to be hindered when indicators
are focused too narrowly, for example, where there is a
focus on only ?nancial metrics (Kaplan & Norton, 1996)
or a strong reliance on only pro?t-based measures of per-
formance (Hopwood, 1974). In addition, strong sanctions
fromfailing to achieve particular objectives or targets asso-
ciated with the PMS (e.g., Van der Stede, 2000) can create a
fear of reprisal inhibiting rather than promoting the free
expression of values and beliefs, particularly where they
relate to expressing the signi?cance of negative outcomes.
Developing a close af?nity between the speci?c values
and beliefs of organizational members and the precise
content and wording of indicators is also important in
enabling the expressive role of PMS. In this process there
are likely to be trade-offs involved in terms of time and
resources necessary in order to develop indicators. Achiev-
ing a close af?nity can be aided by drawing on the experi-
ences and suggestions of organizational members (e.g.,
volunteers) when developing indicators, as it is these staff
who are best placed to judge whether their values and
beliefs are re?ected in the PMS (c.f. Wouters & Wilderom,
2008).
The expressive and instrumental roles of PMS in organizations
Our study also sought to shed light on the way in which
the expressive role of PMS can be a complement to, or
clash with, the use of accounting information for instru-
mental purposes, namely, to improve ef?ciency. Our ?nd-
ings show that the expressive and instrumental roles of
PMS can, and at times do, complement each other (c.f.
Frumkin, 2005). For example, the development of indica-
tors for bene?ciaries provided a mechanism to enable the
exchange of views relating to individuals’ values and
beliefs about bene?ciary care (expressive role), and also
enabled the organization to capture outcomes relating to
bene?ciary care that could help with decision making
and resource allocation (instrumental role). The indicator
on the patient ?le provides a further illustration – it was
standardized and so could be used to aggregate perfor-
mance across the different mental health facilities (essen-
tial for assessing the progress of the programme and the
development of action plans), yet it was also viewed as
expressing the importance of the social model of care
valued by staff and volunteers. As such, we see that PMS
ful?lling an instrumental role can support expressive
purposes by focusing attention on areas important to the
values and beliefs held by organizational members.
Our study also reveals situations in which there is ten-
sion between the expressive and instrumental roles. This
appears to be particularly the case where the use of indica-
tors is unable to recognize the uniqueness of particular sit-
uations and the meaning attached to them by members of
the organization. This was most evident where perfor-
mance was represented in LEAP through the use of a stan-
dardized scoring method (?ve levels of performance for
each indicator), which meant the ‘nuggets of progress’ that
were very meaningful to volunteers were not re?ected
adequately in the scoring process. In this way, the transfor-
mation of qualities (such as nuggets of progress) into quan-
tities expressed on a standard scale (such as a rating of
performance on a scale) hindered the ability of organiza-
tional members (such as volunteers) to express the mean-
ing and signi?cance of some activities. As such, this
illustrates that the commensuration process inherent in
the use of standardized indicators and/or scoring processes
(c.f., Espeland & Sauder, 2007) can, in some situations,
make it dif?cult for the PMS to operate in an expressive
role. However, with careful attention to the development
of indicators, this situation was uncommon in our setting.
This analysis also has implications for prior research
stressing the importance of developing indicators to re?ect
the unique strategy and operating characteristics of
36
The Logical Framework is a commonly-used planning, monitoring and
evaluation tool in international development NGOs. For a description, see
Hall (2014), and for an example of its application, see Martinez and Cooper
(2012).
16 R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
divisions in an organization (e.g., Kaplan & Norton, 1996;
Lipe & Salterio, 2000). The development of unique indica-
tors is typically premised on instrumental grounds – they
promote better decision making, resource allocation and
performance evaluation.
37
Our study suggests an additional
(and complementary) rationale for the development of
unique indicators because they are better able to re?ect,
and thus enable, the expression of values and beliefs salient
to organizational members in different divisions. In contrast,
common measures, by de?nition, are standardized across
divisions and thus would only ful?l an expressive role where
values and beliefs are relatively homogenous throughout the
organization. This discussion may also help to explain resis-
tance to top-down PMS approaches imposing standard indi-
cators and templates (e.g., Wouters & Wilderom, 2008) as
they can fail to account for the differing values and beliefs
of organizational members in different divisions.
Effects of expressive PMS
As outlined above, our study was focused on the ability
of a PMS to help organizational members express their val-
ues and beliefs, the design and operational characteristics
of PMS that facilitate this process, and the relation
between the expressive role of PMS and other uses of
PMS in organizations. As such, in terms of the effects of
expressive PMS, our data are more limited in this regard,
so we seek to highlight aspects of our ?ndings that provide
some preliminary evidence. Given our study was focused
on the development of a PMS, the effects we observe relate
primarily to attitudes toward and use of the PMS itself
rather than behavioral and organizational outcomes
(Cavalluzzo & Ittner, 2004; Wouters & Wilderom, 2008).
The ?rst effect we observed related to changes in atti-
tudes of organizational members toward the PMS
(Wouters & Wilderom, 2008). In particular, we observed
more positive attitudes to the PMS, with organizational
members viewing the ?nal version of LEAP as less onerous,
more user-friendly and more closely linked with their
work activities in comparison to both prior versions of
LEAP and also the Excel spreadsheet. LEAP was also viewed
by staff as providing important information on progress
with the mental health programme that had been lacking
for some time and was considered the most advanced
PMS in use in VSO’s health programmes. We suggest the
expressive role of PMS helps to generate such positive atti-
tudes because it is focused on the collection of information
relating to and allowing the expression of the values and
beliefs of organizational members, which, in our setting,
was primarily concerned with patient care. The effect of
the expressive role of PMS in fostering more positive atti-
tudes toward the PMS is likely to be important in ensuring
sustained engagement with the PMS over time and thus
may help to enhance commitment to a PMS. The absence
of an expressive role could help to improve understanding
of how and why PMS do not operate effectively and/or face
resistance and tension during their implementation and
continued operation (e.g., Ittner & Larcker, 1998).
The second effect relates to the quality of data captured
by the PMS. The Excel spreadsheet developed by the Pro-
gramme Manager was not completed by volunteers and
thus did not provide any helpful information for managing
the mental health programme. In contrast, LEAP was com-
pleted for each mental health institution where VSO had
placed a volunteer, and thus provided a large quantity of
performance information that had not existed previously.
We suggest this arose because enabling organizational
members to express their values and beliefs as part of
the performance measurement process serves to elevate
the level of care and attention they devote to the produc-
tion of high quality information for the PMS, helping to
ensure the information in PMS is more complete and less
prone to errors.
Relatedly, the third effect we observed concerned the
use of performance measurement information in planning
and decision-making. In our study we saw that informa-
tion from LEAP was, toward the end of the project, forming
a more central part of regular programme planning and
review processes, providing information making it easier
to identify and review more systematically areas of good
and bad performance in the mental health programme.
We also saw some evidence of the way in which the
expressive use of PMS can promote learning and explora-
tion of alternatives (Huy, 1999), where organizational
members used the information from LEAP to learn from
prior experiences and develop action plans exploring a
variety of different ideas and alternatives, drawing on the
suggestions of volunteers, VSO staff and staff at the mental
health institutions. These ?ndings are consistent with the
expectation that the expression of values and beliefs can
generate higher levels of energy and motivation amongst
organizational members (Frumkin, 2005; Huy, 1999). In
particular, our ?ndings suggests that where a PMS can
enable the expression of values and beliefs, the resulting
energy and motivation can help with planning and review
processes, particularly in settings where organizational
members come together to discuss progress and action
plans as part of organizational processes.
Suggestions for further research
Most generally, our study suggests that future research
should devote more attention to examination of the
expressive role of PMS, particularly as organizations are
increasingly being seen as having both instrumental and
expressive purposes (Denis et al., 2007; Schultz et al.,
2000). In our study the primary focus was on explicating
the expressive role of PMS and examining the effects on
individuals’ use of and attitudes toward the PMS itself. As
such, future research could investigate the likely behav-
ioral and organizational effects of expressive PMS. At the
behavioral level, this could include examining whether
and how the expressive use of PMS relates to individual
work behaviors, where, for example, we would expect
the expressive use of PMS to generate higher levels of
intrinsic motivation and organizational commitment and
lower levels of role con?ict and resistance amongst
37
As Lipe and Salterio (2000) show, this may not eventuate if managers
?xate on common measures to the exclusion of unique measures, but
subsequent research reveals this can be overcome (e.g., Banker, Chang, &
Pizzini, 2004; Libby, Salterio, & Webb, 2004).
R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx 17
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
organizational members (Burney & Widener, 2007;
Frumkin, 2005; Hall, 2008; Huy, 1999). The expressive role
of PMS could also relate to greater learning and exploration
and reduced emotional dissonance that can accompany
values con?icts (c.f., Huy, 1999). At the organizational
level, research suggests that enabling organizational mem-
bers to express their values and beliefs can help organiza-
tions to achieve their objectives (c.f., Frumkin, 2005; Huy,
1999). Although this indicates the expressive use of PMS
may enhance organizational performance, this would
require careful consideration of other, potentially offset-
ting negative effects from the expressive use of PMS, such
as greater time and effort of organizational members in
developing and using the PMS.
Our empirical setting was characterized by organiza-
tional members having relatively homogenous and
strongly held values and beliefs. Future research could
examine the expressive role of PMS in situations where
the values and beliefs of organizational members are more
heterogeneous and/or where the strength of those beliefs
varies. In these situations, it may be more dif?cult for the
PMS to incorporate a wide variety of indicators adequately
addressing those beliefs and/or the beliefs of organiza-
tional members may con?ict in ways making them dif?cult
to accommodate within the context of a single PMS. In
these situations, careful attention to the development of
workable compromises in the design and operation of the
PMS (c.f. Chenhall et al., 2013) is likely to be very impor-
tant for enabling the expressive role of PMS.
In our case setting, progress toward realizing employ-
ees’ values and beliefs could, in some cases, be achieved
in ways that were consistent with more instrumental
objectives. Future research could consider situations in
which the values of organizational members are in more
serious con?ict with more instrumental concerns. Finally,
as our study shows that PMS can ful?l an expressive role,
future research could explore whether and how other ele-
ments of MCS can potentially operate in ways that enable
organizational members to express their values and
beliefs.
Acknowledgements
The support of CIMA General Charitable Trust is grate-
fully acknowledged. We would also like to thank Brendan
O’Dwyer, Kate Ruff, Mike Shields, Kim Lang?eld-Smith,
Susan O’Leary, Clinton Free, Emily Barman, Kerry Jacobs,
Martin Messner, Allan Hansen, Binh Bui, Kari Lukka, Suresh
Cuganesan, participants at the AOS Workshop on Account-
ing, Non-government Organizations and Civil Society, Lon-
don, December 2012, the 6th Workshop on Management
Accounting as Social and Organizational Practice, Copenha-
gen, April 2013, the Accounting and Finance Association of
Australia and New Zealand (AFAANZ) Conference, July
2013, the Global Management Accounting Research Sym-
posium (GMARS), June 2014, and seminar participants at
the University of Innsbruck, University of Exeter, Erasmus
University, University of Technology, Sydney, University
of New South Wales, HEC Paris, the University of Western
Australia, Macquarie University, and Grif?th University for
their helpful comments and suggestions.
Appendix A. Supplementary material
Supplementary data associated with this article can be
found, in the online version, athttp://dx.doi.org/10.1016/
j.aos.2014.11.002.
References
Abma, T. A. (1997). Playing with/in plurality: Revitalizing realities and
relationships in Rotterdam. Evaluation, 3, 25–48.
Ahrens, T., & Chapman, C. S. (2004). Accounting for ?exibility and
ef?ciency: A ?eld study of management control systems in a
restaurant chain. Contemporary Accounting Research, 21, 271–301.
Ahrens, T., & Mollona, M. (2007). Organisational control as cultural
practice—A shop ?oor ethnography of a Shef?eld steel mill.
Accounting, Organizations and Society, 32, 305–331.
Alac, M., & Hutchins, E. (2004). I see what you are saying: Action as
cognition in fMRI brain mapping practice. Journal of Cognition and
Culture, 4(3), 629–662.
Argyres, N., & McGahan, A. M. (2002). An interview with Michael Porter.
Academy of Management Executive, 16, 43–52.
Banker, R., Chang, H., & Pizzini, M. J. (2004). The balanced scorecard:
Judgmental effects of performance measures linked to strategy. The
Accounting Review, 79(1), 1–23.
Berry, A. J. (2005). Accountability and control in a cat’s cradle. Accounting,
Auditing and Accountability Journal, 18(2), 255–297.
Bisbe, J., Batista-Foguet, J., & Chenhall, R. H. (2007). De?ning management
accounting constructs: A methodological note on the risks of
conceptual misspeci?cation. Accounting, Organizations and Society,
32, 789–820.
Boedker, C., & Chua, W. F. (2013). Accounting as an affective technology: A
study of circulation, agency and entrancement. Accounting,
Organizations and Society, 38, 245–267.
Bolton, S. C., & Houlihan, M. (2009). Are we having fun yet? A
consideration of workplace fun and engagement. Employee Relations,
31(6), 556–568.
Burney, L., & Widener, S. K. (2007). Strategic performance measurement
systems, job-relevant information, and managerial behavioural
responses-role stress and performance. Behavioral Research in
Accounting, 19, 43–69.
Cavalluzzo, K. S., & Ittner, C. D. (2004). Implementing performance
measurement innovations: Evidence from government. Accounting,
Organizations and Society, 29(3–4), 243–267.
Chenhall, R. H., Hall, M., & Smith, D. (2010). Social capital and
management control systems: A case study of a non-government
organization. Accounting, Organizations and Society, 35(8), 737–756.
Chenhall, R. H., Hall, M., & Smith, D. (2013). Performance measurement,
modes of evaluation and the development of compromising accounts.
Accounting, Organizations and Society, 38(4), 268–287.
Cooper, D. J., Hayes, D., &Wolf, F. (1981). Accounting inorganized anarchies:
Understanding and designing accounting systems in ambiguous
situations. Accounting, Organizations and Society, 6(3), 175–191.
Cushen, J. (2013). Financialization in the workplace: Hegemonic
narratives, performative interventions and the angry knowledge
worker. Accounting, Organizations and Society, 38, 314–331.
Dart, J., & Davies, R. (2003). A dialogical, story-based evaluation tool: The
most signi?cant change technique. American Journal of Evaluation, 24,
137–155.
Deetz, S. (1992). Democracy in an age of corporate colonization:
Developments in communication and the politics of everyday life.
Albany: SUNY Press.
Deetz, S. (1995). Transforming communication, transforming business:
Stimulating values negotiation for more responsive and responsible
workplaces. International Journal of Values-Based Management, 8,
255–278.
Denis, J.-L., Langley, A., & Rouleau, L. (2007). Strategizing in pluralistic
contexts: Rethinking theoretical frames. Human Relations, 60(1),
179–215.
Dent, J. F. (1991). Accounting and organizational cultures: A ?eld study of
the emergence of a new organizational reality. Accounting,
Organizations and Society, 16(8), 705–732.
Dutton, J. E., Worline, M. C., Frost, P. J., & Lilius, J. (2006). Explaining
compassion organizing. Administrative Science Quarterly, 51, 59–96.
Ebrahim, A. (2003). Accountability in practice: Mechanisms for NGOs.
World Development, 31, 813–829.
Eisenhardt, K. M. (1989). Building theories from case study research.
Academy of Management Review, 14(4), 532–550.
18 R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002
Espeland, W., & Sauder, M. (2007). Rankings and reactivity: How public
measures recreate social worlds. American Journal of Sociology, 113(1),
1–40.
Etzioni, A. (1961). A comparative analysis of complex organizations. New
York: Free Press.
Ezzamel, M., Willmott, H., & Worthington, F. (2008). Manufacturing
shareholder value: The role of accounting in organizational
transformation. Accounting, Organizations and Society, 33(2–3),
107–140.
Free, C. (2007). Supply chain accounting practices in the UK retail sector:
Enabling or coercing co-operation? Contemporary Accounting
Research, 24(3), 1–24.
Frumkin, P. (2005). On being nonpro?t: A conceptual and policy primer.
Harvard University Press.
Gasper, D. (2000). Evaluating the ‘logical framework approach’ towards
learning-oriented development evaluation. Public Administration and
Development, 20(1), 17–28.
Gordon, C. W., & Babchuk, N. (1959). A typology of voluntary associations.
American Sociological Review, 24(1), 22–29.
Greene, J. C. (1999). The inequality of performance measurements.
Evaluation, 5, 160–172.
Hall, M. (2008). The effect of comprehensive performance measurement
systems on role clarity, psychological empowerment and managerial
performance. Accounting, Organizations and Society, 33(2–3), 141–163.
Hall, M. (2014). Evaluation logics in the third sector. International Journal
of Voluntary and Nonpro?t Organizations, 25, 307–336.
Hirschheim, R., & Klein, H. K. (1994). Realizing emancipatory principles in
information systems development: The case for ETHICS. MIS
Quarterly, March, 83–109.
Hopwood, A. G. (1974). Accounting and human behaviour. London:
Haymarket Publishing.
Howes, M. (1992). Linking paradigms and practice: Key issues in the
appraisal, monitoring and evaluation of British NGO projects. Journal
of International Development, 4, 375–396.
Huy, Q. N. (1999). Emotional capability, emotional intelligence, and
radical change. Academy of Management Review, 24(2), 321–345.
Ittner, C. D., & Larcker, D. F. (1998). Innovations in performance
measurement: Trends and research implications. Journal of
Management Accounting Research, 10, 205–238.
Jönsson, S., & Lukka, K. (2007). There and back again: doing
interventionist research in management accounting. In C. S.
Chapman, A. G. Hopwood, & M. D. Shields (Eds.), Handbook of
management accounting research (pp. 373–397). Elsevier.
Kaplan, R. S., & Norton, D. P. (1996). Balanced Scorecard: Translating
strategy into action. Boston: Harvard Business School Press.
Lewis, D. (2006). The management of non-governmental development
organisations. London: Routledge.
Libby, T., Salterio, S., & Webb, A. (2004). The balanced scorecard: The
effects of assurance and process accountability on managerial
judgment. The Accounting Review, 79(4), 1075–1094.
Lipe, M. G., & Salterio, S. E. (2000). The balanced scorecard: Judgmental
effects of common and unique performance measures. The Accounting
Review, 75(3), 283–298.
Martinez, D., & Cooper, D. (2012). Transforming and mobilizing ?uid
accounting inscriptions: The Logical Framework in nongovernmental
organizations. Working paper.
Mason, D. E. (1996). Leading and managing the expressive dimension:
Harnessing the hidden power source of the nonpro?t sector. San
Francisco: Jossey Bass Publishers.
Najam, A. (1996). NGO accountability: A conceptual framework.
Development Policy Review, 14(1), 339–353.
Adler, P. S., & Borys, B. (1996). Two types of bureaucracy: Enabling and
coercive. Administrative Science Quarterly, 41, 61–89.
Ouchi, W. G. (1979). A conceptual framework for the design of
organizational control mechanisms. Management Science, 25,
833–848.
Pratt, M. G., & Rafaeli, A. (1997). Organizational dress as a symbol of
multilayered social identities. Academy of Management Journal, 40(4),
862–898.
Rowe, C., Birnberg, J. G., & Shields, M. D. (2008). Effects of organizational
process change on responsibility accounting and managers’
revelations of private knowledge. Accounting, Organizations and
Society, 33(2–3), 164–198.
Schultz, M., Hatch, M. J., & Larsen, M. H. (2000). Introduction: Why the
expressive organization? In M. Schultz, M. J. Hatch, & M. H. Larsen
(Eds.), The expressive organization (pp. 1–10). Oxford: Oxford
University Press.
Shaw, G. G. (2000). Planning and communicating using stories. In M.
Schultz, M. J. Hatch, & M. H. Larsen (Eds.), The expressive organization
(pp. 182–195). Oxford: Oxford University Press.
Simons, R. (1990). The role of management control systems in creating
competitive advantage: New perspectives. Accounting, Organizations
and Society, 15(1/2), 127–143.
Simons, R. (1995). Levers of control. Boston: Harvard Business School
Press.
Smith, A. C. T., & Stewart, B. (2011). Organizational rituals: Features,
functions and mechanisms. International Journal of Management
Reviews, 13, 113–133.
Spradley, J. P. (1980). Participant observation. Rinehart and Winston: Holt.
Trice, H., & Beyer, J. (1993). The cultures of work organizations. Englewood
Cliffs, NJ: Prentice-Hall.
Van der Stede, W. A. (2000). The relationship between two consequences
of budgetary controls: budgetary slack creation and managerial short-
term orientation. Accounting, Organizations and Society, 25, 609–622.
Voluntary Services Overseas (2004). Focus for change: VSO’s strategic plan.
Accessed 10.08.10.
Wallace, T., Bornstein, L., & Chapman, J. (2007). The aid chain: Coercion and
commitment in development NGOs. Rugby, UK: Practical Action
Publishing.
Wouters, M., & Roijmans, D. (2011). Using prototypes to induce
experimentation and knowledge integration in the development of
enabling accounting information. Contemporary Accounting Research,
28(2), 708–736.
Wouters, M., & Wilderom, C. (2008). Developing performance-
measurement systems as enabling formalization: A longitudinal
?eld study of a logistics department. Accounting, Organizations and
Society, 33(4–5), 488–516.
R.H. Chenhall et al. / Accounting, Organizations and Society xxx (2014) xxx–xxx 19
Please cite this article in press as: Chenhall, R. H., et al. The expressive role of performance measurement systems: A ?eld study of a mental
health development project. Accounting, Organizations and Society (2014),http://dx.doi.org/10.1016/j.aos.2014.11.002

doc_221623273.pdf