Description
The achievement of customer satisfaction in service operations depends to a great extent on employee customer service behavior (CSB). In this study, 123 service providers (77% response rate) responded to a survey assessing service predisposition and job characteristics.
JOURNAL OF BUSINESS AND PSYCHOLOGY
Volume 13, No. 3, Spring 1999
CUSTOMER SERVICE BEHAVIOR:
THE INTERACTION OF SERVICE
PREDISPOSITION AND JOB
CHARACTERISTICS
Steven G. Rogelberg
Bowling Green State University
Janet L. Barnes-Farrell
University of Connecticut
Victoria Creamer
International Business Machines
ABSTRACT: The achievement of customer satisfaction in service operations de-
pends to a great extent on employee customer service behavior (CSB). In this
study, 123 service providers (77% response rate) responded to a survey assessing
service predisposition and job characteristics. Employees also completed a be-
haviorally based CSB measure developed after interviewing and surveying cus-
tomers (n = 96). Analyses suggest that job characteristics, alone, accounted for a
significant amount of CSB variance. Job characteristics did not moderate the
relationship between service predisposition and CSB. In contrast to previous
work, service predisposition and CSB were not correlated. A service provider
typology is presented to explain these findings.
INTRODUCTION
One way for a company to differentiate itself from others in a com-
petitive service environment is to advocate and provide excellent cus-
This article is based on the first author's doctoral dissertation, completed at the University
of Connecticut. Portions of this paper were presented at the 10th annual conference of the
Society for Industrial and Organizational Psychology held in Orlando, Florida (May, 1995).
We wish to thank Peter Bachiochi, Harold Goldstein, Milt Hakel, Jim Holzworth,
Charles Lowe, Steven Mellor, and Ann Marie Ryan for their wisdom and assistance in
these research efforts.
Address correspondence to Steven Rogelberg, Department of Psychology, Bowling
Green State University, Bowling Green, Ohio 43403. Electronic mail may be sent via In-
ternet to [email protected].
421 O 1999 Human Sciences Press, Inc.
422 JOURNAL OF BUSINESS AND PSYCHOLOGY
tomer service (Bateson, 1989; Zeithaml, Parasuraman & Berry, 1990).
The rationale for this strategy is that excellent customer service leads to
customer satisfaction which, in turn, increases the customer's desire to
use the supplier's services in the future (Cronin & Taylor, 1992), thus
increasing company revenue (Burke & Borucki, 1995; Johnson, Ryan, &
Schmit, 1994; Rogelberg & Creamer, 1994).
The achievement of customer satisfaction in service operations de-
pends to a great extent on employee customer service behavior (CSB). In
fact, the prominence of the service deliverer has led many researchers to
conclude that: 1) employees remain the key to success at the 'moments
of truth' (Carlzon, 1987; Normann, 1984); 2) the manner in which the
service is delivered is critical in consumers' evaluations of service qual-
ity (Czeipiel, Solomon & Supprenant, 1985); and 3) service employees
run the service operation, market the service and are equated by cus-
tomers with the service (Lovelock, 1981). Unfortunately, not all em-
ployees are customer service oriented: employees differ in the extent and
frequency with which they correctly identify and fulfill customer needs
in a timely, ethical and courteous way. These differences in employee
CSB are typically attributed to dispositional variables associated with
the service provider (i.e., personality characteristics) and variables asso-
ciated with the job or the organizational context in which the service
provider carries out his/her job.
Research has identified a number of service provider personality
traits that are related to external customer service ratings (Hogan,
Hogan & Busch, 1984; Rosse, Miller & Barnes, 1991; Paajanen, 1991;
Fogli & Whitney, 1991; Kwan, Horn & Kristen, 1993). Taken together,
these personality composites (e.g., likeability, adjustment, and
sociability) are commonly labeled service predisposition (e.g., Hogan's
Service Orientation Index). While service predisposition lends insight
into stable individual differences among employees in terms of their pro-
pensity to exhibit customer service behaviors, work climate has also
been shown to affect customer service behavior.
Individuals may not be encouraged to behave in a service oriented
manner unless an organizational climate exists that is supportive of cus-
tomer service. In 1980, Schneider, Parkington, and Buxton identified a
positive relationship between employees' perceptions of organizational
practices and procedures and customer perceptions of service quality. In
1995, Schneider and Bowen replicated these results. Specifically, they
found that employees' perceptions of work facilitators, supervision, orga-
nizational career facilitation, socialization, management commitment,
systems support and logistics support were all positively related to cus-
tomers' perceptions of service quality. Subsequent research by Jones
(1991, 1992), Tornow and Wiley (1991), Ulrich, Halbrook, Meder,
Stuchlik and Thorpe (1991) and Schmit and Allscheid (1995) has pro-
S. G. ROGELBERG, J. L. BARNES-PARRELL, AND V. CREAMER
vided additional support for the findings of Schneider and his col-
leagues. Taken together, it appears that an organization interested in
achieving customer satisfaction needs to establish an organizational cli-
mate for service and an organizational climate for employee well being.
The Present Study
Previous research has generally examined the independent relation-
ship of one domain of variables (e g., job characteristics or personality)
with CSB, but has failed to examine the interactive effects of both types
of variables on CSB. This is surprising given the preponderance of
thought advocating the position that worker behavior is a function of the
interaction between organizational context and individual characteris-
tics (Aronoff & Wilson, 1985; Endler & Magnusson, 1976; Ostroff, 1993;
Pervin & Lewis, 1978, Terborg, 1981). While research has examined the
interactionist perspective for general job performance (Barrick & Mount,
1993; Colarelli, Dean, & Konstans, 1987; Ostroff, 1993), extant research
has not examined the interactionist perspective with respect to the spe-
cific domain of customer service behavior. The following study examines
the interactive effects of service predisposition and job characteristics on
CSB. Specifically, we posed the question: Is the relationship between
service predisposition and CSB moderated by job characteristics that
encourage or hinder the display of CSB?
The sample used for this study is quite different from the samples
studied in past research on CSB. Previous research has typically exam-
ined service providers who interacted face to face with customers on a
short term basis (e.g., bank tellers; Schneider, Parkington, & Buxton,
1980; Schneider & Bowen, 1985). In this study, employment consultants
providing recruiting and/or job applicant screening services to customers
were examined. These service providers interacted with customers, gen-
erally via telephone, a number of times over a period of time (one week
to one month). In addition, these individuals generally worked indepen-
dently of others, with the possible exception of their supervisors. There-
fore, besides examining CSB using an interactionist perspective, this
study serves to replicate previous research on job characteristics and
service predisposition using a type of service provider that has not been
examined to date.
METHOD
Participants
A questionnaire was distributed to employees of an organization (24
sites) that provides employment services (e.g., recruiting, testing, and
423
JOURNAL OF BUSINESS AND PSYCHOLOGY
screening of job applicants). Participation in the study was voluntary
and responses to the survey were anonymous. A total of 123 service pro-
viders (77% response rate) responded to the questionnaire. Of those re-
sponding, the following demographic characteristics were reported: 37%
reported being male; with regard to race, 76.8% of the sample was Cau-
casian, 15% African-American, and 8% Asian, Hispanic or Native Amer-
ican. In addition, 88% of the service providers sampled had been en-
gaged in customer service work for over one year. Eight respondents
who reported not interacting with customers were dropped from all fur-
ther analyses.
Employee Questionnaire
Service providers completed a questionnaire designed to assess
three classes of variables: 1) service predisposition; 2) job characteris-
tics; and 3) customer service behavior. In addition, employees reported
how frequently they interacted with customers and provided informa-
tion about the demographic characteristics noted above. All survey ques-
tions, excluding the true/false service predisposition questions, were an-
swered on five-point rating scales.
Service Predisposition. A fourteen item subset of Hogan and Hogan's
(1995) 87-item service orientation index was used to measure service
predisposition. Hogan and Hogan (1995, pg. 64) report that the 14-item
measure (e.g., "I usually notice when I am boring people"), as a whole,
was correlated .71 with the 87-item measure and that test-retest re-
liability over a four-week period was .78. In turn, the 87-item measure
has been shown to be positively related to customer service behavior in a
number of contexts (Hogan, Hogan & Busch, 1984; Rosse, Miller &
Barnes, 1991).
Job Characteristics. Because the employee questionnaire used was de-
signed primarily for management feedback purposes rather than re-
search purposes, there were constraints on the number and type of
items we could include in the measure. Eight items assessing work char-
acteristics were included in the survey. A principal axis factor analysis
of responses to these items revealed a "clean" three-factor structure.
Specifically, the eigenvalues for the first three factors all exceeded 1.0;
no commonality estimates were under .55; after a varimax rotation all
factor loadings were found to be above .6; and no items cross-loaded.
Additional information is available from the first author. Based on the
results of the factor analysis, three subscales were formed. The first sub-
scale, labeled Autonomy, contained the following three items: "I am em-
powered to carry out my job responsibilities in the way I think best"; "I
am accountable for the results of my work"; and "I am encouraged to
424
S. G. ROGELBERG, J. L. BARNES-FARRELL, AND V. CREAMER
take reasonable risks in order to effectively fulfill my job respon-
sibilities". The second subscale, labeled Requisite Time, contained three
questions that are quite similar to Beehr, Walsh and Taber's (1976) role
overload scale. The three items were: "Unnecessary work in my job
makes it hard for me to effectively service my customers"; "I sometimes
feel a lack of control over my job because too many customers demand
service at the same time"; and "I have enough time to serve my cus-
tomer effectively". The final subscale, labeled Requisite Resources, con-
tained the following two questions: "I feel like I have the necessary
equipment to provide high quality service to my customers"; and "I have
enough information to do my job well".
Customer Service Behavior. The design of our study also required an
individual-level measure of CSB. While, ideally customers should rate
each respective service provider, very rarely are we able to ascertain
customer ratings of service providers by "name". Typically, customers
complete a survey containing a series of questions assessing service/
product quality followed by a question as to what work location provided
the service. Customer responses are then aggregated by work location
and inferences are made about individual CSB. Individual CSB ratings
can come more readily from one of two sources: the supervisor and/or
the service provider themselves. Unfortunately, supervisors are gener-
ally not good sources of information concerning individual CSB for two
reasons: 1) large spans of control reduce the amount of direct informa-
tion that they are likely to have regarding individual employees; and 2)
supervisors are usually not part of the service encounter. As a result, a
well-designed employee self-report measure may be the most viable
source of information concerning individual customer service behavior.
This position is supported by the research that has found a significant
relationship (r=.60 and above) between employees perceptions of ser-
vice quality and customers perceptions of service quality (Schneider,
Parkington, & Buxton, 1980; Schneider & Bowen, 1985). Furthermore,
research has found that an employee self report measure (the SOCO)
assessing selling orientation was related to sales performance (Saxe &
Weitz, 1982). Therefore, while self-report measures tend to be inflated
(Beatty, Schneier, & Beatty, 1977), they still are correlated with impor-
tant outcomes and can serve as an important source of information
when well constructed.
The present study introduces a self-report measure called the be-
havioral self assessment (BSA). The BSA asks employees to rate their
performance on certain critical customer service behaviors. Discussion
of this scale must begin with an elaboration of the careful process used
to identify the key service behaviors on which the measure was based.
The specific service behaviors used for the self-report measure were
425
JOURNAL OF BUSINESS AND PSYCHOLOGY
Table 1
Criticality Study Results
Service Factor
Providing hiring managers with pre-interview information
Knowledge of the hiring company and employment process
Processing and handling of paperwork
Communication with the hiring manager
Providing information to job candidates
Handling of verifications and medical testing
Screening of job applicants
Tracking of job applicants in employment process
Handling of job candidate logistics and expenses
Total (n = 45)
mean (s.d.)
1.51 (.84)
1.62 (.76)
1.66 (.72)
1.70 (.84)
1.73 (.86)
1.77 (.83)
1.78 (.97)
1.86 (.75)
2.23 (.93)
Note. Means range from very critical (1) to not at all critical (5). Lower means reflect
greater criticality.
compiled after conducting and analyzing 50 critical incident interviews.
Randomly selected customers were asked to generate specific examples
of service behaviors that were either exemplary or poor. Two content
experts, then, independently sorted the 120 non-redundant critical inci-
dents (e.g., "post-hire paperwork was processed accurately") into nine
categories based on their conceptual and logical similarity (85% agree-
ment). The nine emergent categories, hereafter referred to as key ser-
vice factors, are shown in Table 1.
To further assure the validity and representativeness of the nine
key service factors, 46 additional randomly selected customers com-
pleted a survey (response rate = 43%) that asked them to rate the crit-
icality (five point rating scale, a value of one indicating "very critical") of
each service factor for achieving customer satisfaction. All criticality
means were less than 2.5 with standard deviations less that one. In
addition, customers did not identify any additional service behaviors re-
lated to their perceptions of service quality. Therefore, we concluded
that the nine key service factors captured the domain of important ser-
vice behaviors for this organization. Note, however, that these nine fac-
tors may be endemic to the host organization and the services it pro-
vides. A summary of the findings can be found in Table 1.
The behavior self-assessment measure that employees completed
represented the nine key service factors identified above. Subject matter
experts identified one specific and representative behavior from each ser-
vice factor. Employees, in turn, rated their performance on these nine
specific behaviors (e.g., "I communicated the candidate's status in the
employment process"). To facilitate honest responses and minimize the
426
S. G. ROGELBERG, J. L. BARNES-FARRELL, AND V. CREAMER
leniency bias that is typical of self-ratings of performance, two strategies
were used. First, service providers were encouraged to respond as accu-
rately and honestly as possible to the measure. In addition, we utilized a
strategy adapted from work on leniency reduction by Farh and Werbel
(1986): Employees were told that that their customers responded to these
same questions via a customer satisfaction survey and that we wanted to
see if their self-assessments were in line with the customer ratings.
RESULTS
Summary Statistics
All experimental variables were examined for identification of ex-
cessive amounts of missing values, and the identification of skewed
variables that would tend to result in a violation of the assumptions
underlying the statistical tests employed in this study. Frequency distri-
butions suggested minimal missing values. Furthermore, skewness in-
dices did not exceed a value of one for any of the measured variables.
A summary of the means and standard deviations for all major vari-
ables assessed can be found in Table 2. Note that higher values reflect
more positive scores/perceptions on the various measures. The observed
service predisposition mean of 9.03 (s.d. = 1.47) was slightly below the
10.36 (s.d. = 2.52) norm reported by Hogan and Hogan (1995, pg. 65).
Average perceptions of the job characteristics ranged from a low of 3.21
for the requisite time scale to a high of 4.12 for the autonomy scale. On
average, CSB ratings were quite high at 4.40, with some variability
(s.d. = .40). However, it is worth noting that customers were indepen-
dently surveyed by the host organization (shortly after collecting this
study^s data) to assess customer satisfaction. Customer satisfaction rat-
ings, using the same items found in the CSB, were quite high (£ = 4.12,
s.d. = .82), and thus consistent with the observed CSB self-report ratings
(Rogelberg, 1995).
Table 2
Summary Statistics and Intel-correlations of Measures
Measure
1. Service Predisposition
2. Autonomy
3. Requisite Time
4. Requisite Resources
5. CSB
*p
The achievement of customer satisfaction in service operations depends to a great extent on employee customer service behavior (CSB). In this study, 123 service providers (77% response rate) responded to a survey assessing service predisposition and job characteristics.
JOURNAL OF BUSINESS AND PSYCHOLOGY
Volume 13, No. 3, Spring 1999
CUSTOMER SERVICE BEHAVIOR:
THE INTERACTION OF SERVICE
PREDISPOSITION AND JOB
CHARACTERISTICS
Steven G. Rogelberg
Bowling Green State University
Janet L. Barnes-Farrell
University of Connecticut
Victoria Creamer
International Business Machines
ABSTRACT: The achievement of customer satisfaction in service operations de-
pends to a great extent on employee customer service behavior (CSB). In this
study, 123 service providers (77% response rate) responded to a survey assessing
service predisposition and job characteristics. Employees also completed a be-
haviorally based CSB measure developed after interviewing and surveying cus-
tomers (n = 96). Analyses suggest that job characteristics, alone, accounted for a
significant amount of CSB variance. Job characteristics did not moderate the
relationship between service predisposition and CSB. In contrast to previous
work, service predisposition and CSB were not correlated. A service provider
typology is presented to explain these findings.
INTRODUCTION
One way for a company to differentiate itself from others in a com-
petitive service environment is to advocate and provide excellent cus-
This article is based on the first author's doctoral dissertation, completed at the University
of Connecticut. Portions of this paper were presented at the 10th annual conference of the
Society for Industrial and Organizational Psychology held in Orlando, Florida (May, 1995).
We wish to thank Peter Bachiochi, Harold Goldstein, Milt Hakel, Jim Holzworth,
Charles Lowe, Steven Mellor, and Ann Marie Ryan for their wisdom and assistance in
these research efforts.
Address correspondence to Steven Rogelberg, Department of Psychology, Bowling
Green State University, Bowling Green, Ohio 43403. Electronic mail may be sent via In-
ternet to [email protected].
421 O 1999 Human Sciences Press, Inc.
422 JOURNAL OF BUSINESS AND PSYCHOLOGY
tomer service (Bateson, 1989; Zeithaml, Parasuraman & Berry, 1990).
The rationale for this strategy is that excellent customer service leads to
customer satisfaction which, in turn, increases the customer's desire to
use the supplier's services in the future (Cronin & Taylor, 1992), thus
increasing company revenue (Burke & Borucki, 1995; Johnson, Ryan, &
Schmit, 1994; Rogelberg & Creamer, 1994).
The achievement of customer satisfaction in service operations de-
pends to a great extent on employee customer service behavior (CSB). In
fact, the prominence of the service deliverer has led many researchers to
conclude that: 1) employees remain the key to success at the 'moments
of truth' (Carlzon, 1987; Normann, 1984); 2) the manner in which the
service is delivered is critical in consumers' evaluations of service qual-
ity (Czeipiel, Solomon & Supprenant, 1985); and 3) service employees
run the service operation, market the service and are equated by cus-
tomers with the service (Lovelock, 1981). Unfortunately, not all em-
ployees are customer service oriented: employees differ in the extent and
frequency with which they correctly identify and fulfill customer needs
in a timely, ethical and courteous way. These differences in employee
CSB are typically attributed to dispositional variables associated with
the service provider (i.e., personality characteristics) and variables asso-
ciated with the job or the organizational context in which the service
provider carries out his/her job.
Research has identified a number of service provider personality
traits that are related to external customer service ratings (Hogan,
Hogan & Busch, 1984; Rosse, Miller & Barnes, 1991; Paajanen, 1991;
Fogli & Whitney, 1991; Kwan, Horn & Kristen, 1993). Taken together,
these personality composites (e.g., likeability, adjustment, and
sociability) are commonly labeled service predisposition (e.g., Hogan's
Service Orientation Index). While service predisposition lends insight
into stable individual differences among employees in terms of their pro-
pensity to exhibit customer service behaviors, work climate has also
been shown to affect customer service behavior.
Individuals may not be encouraged to behave in a service oriented
manner unless an organizational climate exists that is supportive of cus-
tomer service. In 1980, Schneider, Parkington, and Buxton identified a
positive relationship between employees' perceptions of organizational
practices and procedures and customer perceptions of service quality. In
1995, Schneider and Bowen replicated these results. Specifically, they
found that employees' perceptions of work facilitators, supervision, orga-
nizational career facilitation, socialization, management commitment,
systems support and logistics support were all positively related to cus-
tomers' perceptions of service quality. Subsequent research by Jones
(1991, 1992), Tornow and Wiley (1991), Ulrich, Halbrook, Meder,
Stuchlik and Thorpe (1991) and Schmit and Allscheid (1995) has pro-
S. G. ROGELBERG, J. L. BARNES-PARRELL, AND V. CREAMER
vided additional support for the findings of Schneider and his col-
leagues. Taken together, it appears that an organization interested in
achieving customer satisfaction needs to establish an organizational cli-
mate for service and an organizational climate for employee well being.
The Present Study
Previous research has generally examined the independent relation-
ship of one domain of variables (e g., job characteristics or personality)
with CSB, but has failed to examine the interactive effects of both types
of variables on CSB. This is surprising given the preponderance of
thought advocating the position that worker behavior is a function of the
interaction between organizational context and individual characteris-
tics (Aronoff & Wilson, 1985; Endler & Magnusson, 1976; Ostroff, 1993;
Pervin & Lewis, 1978, Terborg, 1981). While research has examined the
interactionist perspective for general job performance (Barrick & Mount,
1993; Colarelli, Dean, & Konstans, 1987; Ostroff, 1993), extant research
has not examined the interactionist perspective with respect to the spe-
cific domain of customer service behavior. The following study examines
the interactive effects of service predisposition and job characteristics on
CSB. Specifically, we posed the question: Is the relationship between
service predisposition and CSB moderated by job characteristics that
encourage or hinder the display of CSB?
The sample used for this study is quite different from the samples
studied in past research on CSB. Previous research has typically exam-
ined service providers who interacted face to face with customers on a
short term basis (e.g., bank tellers; Schneider, Parkington, & Buxton,
1980; Schneider & Bowen, 1985). In this study, employment consultants
providing recruiting and/or job applicant screening services to customers
were examined. These service providers interacted with customers, gen-
erally via telephone, a number of times over a period of time (one week
to one month). In addition, these individuals generally worked indepen-
dently of others, with the possible exception of their supervisors. There-
fore, besides examining CSB using an interactionist perspective, this
study serves to replicate previous research on job characteristics and
service predisposition using a type of service provider that has not been
examined to date.
METHOD
Participants
A questionnaire was distributed to employees of an organization (24
sites) that provides employment services (e.g., recruiting, testing, and
423
JOURNAL OF BUSINESS AND PSYCHOLOGY
screening of job applicants). Participation in the study was voluntary
and responses to the survey were anonymous. A total of 123 service pro-
viders (77% response rate) responded to the questionnaire. Of those re-
sponding, the following demographic characteristics were reported: 37%
reported being male; with regard to race, 76.8% of the sample was Cau-
casian, 15% African-American, and 8% Asian, Hispanic or Native Amer-
ican. In addition, 88% of the service providers sampled had been en-
gaged in customer service work for over one year. Eight respondents
who reported not interacting with customers were dropped from all fur-
ther analyses.
Employee Questionnaire
Service providers completed a questionnaire designed to assess
three classes of variables: 1) service predisposition; 2) job characteris-
tics; and 3) customer service behavior. In addition, employees reported
how frequently they interacted with customers and provided informa-
tion about the demographic characteristics noted above. All survey ques-
tions, excluding the true/false service predisposition questions, were an-
swered on five-point rating scales.
Service Predisposition. A fourteen item subset of Hogan and Hogan's
(1995) 87-item service orientation index was used to measure service
predisposition. Hogan and Hogan (1995, pg. 64) report that the 14-item
measure (e.g., "I usually notice when I am boring people"), as a whole,
was correlated .71 with the 87-item measure and that test-retest re-
liability over a four-week period was .78. In turn, the 87-item measure
has been shown to be positively related to customer service behavior in a
number of contexts (Hogan, Hogan & Busch, 1984; Rosse, Miller &
Barnes, 1991).
Job Characteristics. Because the employee questionnaire used was de-
signed primarily for management feedback purposes rather than re-
search purposes, there were constraints on the number and type of
items we could include in the measure. Eight items assessing work char-
acteristics were included in the survey. A principal axis factor analysis
of responses to these items revealed a "clean" three-factor structure.
Specifically, the eigenvalues for the first three factors all exceeded 1.0;
no commonality estimates were under .55; after a varimax rotation all
factor loadings were found to be above .6; and no items cross-loaded.
Additional information is available from the first author. Based on the
results of the factor analysis, three subscales were formed. The first sub-
scale, labeled Autonomy, contained the following three items: "I am em-
powered to carry out my job responsibilities in the way I think best"; "I
am accountable for the results of my work"; and "I am encouraged to
424
S. G. ROGELBERG, J. L. BARNES-FARRELL, AND V. CREAMER
take reasonable risks in order to effectively fulfill my job respon-
sibilities". The second subscale, labeled Requisite Time, contained three
questions that are quite similar to Beehr, Walsh and Taber's (1976) role
overload scale. The three items were: "Unnecessary work in my job
makes it hard for me to effectively service my customers"; "I sometimes
feel a lack of control over my job because too many customers demand
service at the same time"; and "I have enough time to serve my cus-
tomer effectively". The final subscale, labeled Requisite Resources, con-
tained the following two questions: "I feel like I have the necessary
equipment to provide high quality service to my customers"; and "I have
enough information to do my job well".
Customer Service Behavior. The design of our study also required an
individual-level measure of CSB. While, ideally customers should rate
each respective service provider, very rarely are we able to ascertain
customer ratings of service providers by "name". Typically, customers
complete a survey containing a series of questions assessing service/
product quality followed by a question as to what work location provided
the service. Customer responses are then aggregated by work location
and inferences are made about individual CSB. Individual CSB ratings
can come more readily from one of two sources: the supervisor and/or
the service provider themselves. Unfortunately, supervisors are gener-
ally not good sources of information concerning individual CSB for two
reasons: 1) large spans of control reduce the amount of direct informa-
tion that they are likely to have regarding individual employees; and 2)
supervisors are usually not part of the service encounter. As a result, a
well-designed employee self-report measure may be the most viable
source of information concerning individual customer service behavior.
This position is supported by the research that has found a significant
relationship (r=.60 and above) between employees perceptions of ser-
vice quality and customers perceptions of service quality (Schneider,
Parkington, & Buxton, 1980; Schneider & Bowen, 1985). Furthermore,
research has found that an employee self report measure (the SOCO)
assessing selling orientation was related to sales performance (Saxe &
Weitz, 1982). Therefore, while self-report measures tend to be inflated
(Beatty, Schneier, & Beatty, 1977), they still are correlated with impor-
tant outcomes and can serve as an important source of information
when well constructed.
The present study introduces a self-report measure called the be-
havioral self assessment (BSA). The BSA asks employees to rate their
performance on certain critical customer service behaviors. Discussion
of this scale must begin with an elaboration of the careful process used
to identify the key service behaviors on which the measure was based.
The specific service behaviors used for the self-report measure were
425
JOURNAL OF BUSINESS AND PSYCHOLOGY
Table 1
Criticality Study Results
Service Factor
Providing hiring managers with pre-interview information
Knowledge of the hiring company and employment process
Processing and handling of paperwork
Communication with the hiring manager
Providing information to job candidates
Handling of verifications and medical testing
Screening of job applicants
Tracking of job applicants in employment process
Handling of job candidate logistics and expenses
Total (n = 45)
mean (s.d.)
1.51 (.84)
1.62 (.76)
1.66 (.72)
1.70 (.84)
1.73 (.86)
1.77 (.83)
1.78 (.97)
1.86 (.75)
2.23 (.93)
Note. Means range from very critical (1) to not at all critical (5). Lower means reflect
greater criticality.
compiled after conducting and analyzing 50 critical incident interviews.
Randomly selected customers were asked to generate specific examples
of service behaviors that were either exemplary or poor. Two content
experts, then, independently sorted the 120 non-redundant critical inci-
dents (e.g., "post-hire paperwork was processed accurately") into nine
categories based on their conceptual and logical similarity (85% agree-
ment). The nine emergent categories, hereafter referred to as key ser-
vice factors, are shown in Table 1.
To further assure the validity and representativeness of the nine
key service factors, 46 additional randomly selected customers com-
pleted a survey (response rate = 43%) that asked them to rate the crit-
icality (five point rating scale, a value of one indicating "very critical") of
each service factor for achieving customer satisfaction. All criticality
means were less than 2.5 with standard deviations less that one. In
addition, customers did not identify any additional service behaviors re-
lated to their perceptions of service quality. Therefore, we concluded
that the nine key service factors captured the domain of important ser-
vice behaviors for this organization. Note, however, that these nine fac-
tors may be endemic to the host organization and the services it pro-
vides. A summary of the findings can be found in Table 1.
The behavior self-assessment measure that employees completed
represented the nine key service factors identified above. Subject matter
experts identified one specific and representative behavior from each ser-
vice factor. Employees, in turn, rated their performance on these nine
specific behaviors (e.g., "I communicated the candidate's status in the
employment process"). To facilitate honest responses and minimize the
426
S. G. ROGELBERG, J. L. BARNES-FARRELL, AND V. CREAMER
leniency bias that is typical of self-ratings of performance, two strategies
were used. First, service providers were encouraged to respond as accu-
rately and honestly as possible to the measure. In addition, we utilized a
strategy adapted from work on leniency reduction by Farh and Werbel
(1986): Employees were told that that their customers responded to these
same questions via a customer satisfaction survey and that we wanted to
see if their self-assessments were in line with the customer ratings.
RESULTS
Summary Statistics
All experimental variables were examined for identification of ex-
cessive amounts of missing values, and the identification of skewed
variables that would tend to result in a violation of the assumptions
underlying the statistical tests employed in this study. Frequency distri-
butions suggested minimal missing values. Furthermore, skewness in-
dices did not exceed a value of one for any of the measured variables.
A summary of the means and standard deviations for all major vari-
ables assessed can be found in Table 2. Note that higher values reflect
more positive scores/perceptions on the various measures. The observed
service predisposition mean of 9.03 (s.d. = 1.47) was slightly below the
10.36 (s.d. = 2.52) norm reported by Hogan and Hogan (1995, pg. 65).
Average perceptions of the job characteristics ranged from a low of 3.21
for the requisite time scale to a high of 4.12 for the autonomy scale. On
average, CSB ratings were quite high at 4.40, with some variability
(s.d. = .40). However, it is worth noting that customers were indepen-
dently surveyed by the host organization (shortly after collecting this
study^s data) to assess customer satisfaction. Customer satisfaction rat-
ings, using the same items found in the CSB, were quite high (£ = 4.12,
s.d. = .82), and thus consistent with the observed CSB self-report ratings
(Rogelberg, 1995).
Table 2
Summary Statistics and Intel-correlations of Measures
Measure
1. Service Predisposition
2. Autonomy
3. Requisite Time
4. Requisite Resources
5. CSB
*p