The Relevance of Transformational Leadership to Nontraditional Accounting Services: Information...
ABSTRACT: This paper examines whether supervisors in nontraditional accounting services (vs. supervisors in financial auditing and tax services) are more likely to acquire transformational leadership skills, and, whether such skills impact subordinates' role clarity and role outcomes. Transformational
Keywords: transformational leadership; task analyzability; transactional leadership; information systems assurance; business consulting; role clarity; job satisfaction; organizational commitment; job performance.
Data Availability: Subject to promised confidentiality, data are available from the author.
I. INTRODUCTION
Striving for greater efficiency and effectiveness in performing audit work, public accounting firms have historically employed objective, computational procedures (1) for managing professional services (Cushing and Loebbecke 1986; Bamber et al. 1989). However, economic competition in recent years has motivated CPAs to diversify the types of professional services provided (Melancon 1998; Elliott 2000). Development and implementation of new services, by definition, relates to innovation and novelty, and is characterized by less reliance on established rules and procedures. Whether traditional management techniques historically employed by CPAs are sufficient for managing emerging services in information systems assurance and business consulting is questionable.
For example, Summers et al. (2000) found that public accounting consultants and auditors differ in regards to innovativeness. Professionals working in consulting services were more prolific in proposing original and creative solutions to problems, less concerned with immediate efficiency, and less concerned with certainty, rules, and norms. The authors found that incongruence between personality type (adaptive vs. innovative) and professional services (auditing vs. consulting services) was associated with greater role conflict. An implication of the Summers et al. (2000) study is that the most effective management tools for these two areas are likely to differ since consulting and auditing services attract and employ different personality types.
This paper examines whether transformational leadership is a viable supplement to objective, computational procedures (task analyzability) and traditional supervision (i.e., transactional leadership) in managing nontraditional accounting services such as information systems assurance and business consulting. Transformational leaders provide leadership that goes beyond structuring of tasks (e.g., specifying a sequence of steps for carrying out assignments and responsibilities) and transacting with subordinates (e.g., offering rewards and penalties regarding their compliance with supervisor expectations). Transformational leadership includes leader behaviors that broaden and elevate subordinates' interests, activate their higher-order needs, and motivate subordinates to transcend their own self-interests for the good of the organization, resulting in performance beyond original expectations (Burns 1978; Bass 1985; Yukl 1989). The dynamic nature of information systems services and business consulting provides an innovative environment where transformational leadership should be an effective management tool. If so, the extant literature (as discussed below) suggests that transformational leadership is likely to be associated with positive role outcomes for subordinates, i.e., higher levels of role clarity, job satisfaction, organizational commitment, and job performance.
The motivation for this study is closely aligned with the AICPA's Vision Project. One of the Vision Project's outcomes was recognition of the need for CPAs to develop the leadership skills necessary to influence, inspire, and motivate others to achieve results (AICPA 2000). While not identified as such, the CPA Vision Project's definition of leadership skill is essentially the definition of transformational leadership.
This paper first provides a review of transformational leadership theory and proposes specific hypotheses regarding transformational leader behaviors in nontraditional accounting services. Then role theory is used to articulate specific links between transformational leader behavior, role clarity, and role outcomes (job satisfaction, organizational commitment, and job performance). The remaining sections discuss research methods (participants and instruments), statistical analyses and results, and implications for future research.
II. THEORETICAL BACKGROUND AND HYPOTHESES
Prior leadership research in the accounting profession has often investigated the effectiveness of two leader behaviors: initiating structure and consideration (2) (Pratt and Jiambalvo 1981, 1982; Jiambalvo and Pratt 1982; Kida 1984; Kelley and Margheim 1990; Pasewark et al. 1994; Otley and Pierce 1995). Both constructs represent leader efforts that are transactional in nature, targeting specific role behavior by subordinates and supporting satisfactory working relationships between subordinates and supervisors. However, Burns (1978), followed by Bass (1985), noted that only a subset of leadership skills can be characterized by superior/subordinate exchanges (or transactions) that promise subordinates rewards for compliance and penalties for noncompliance with supervisors' proposals, requests, and expectations. Beyond this transactional view of leadership are leader behaviors that are transformational in nature.
Transformational leadership includes six potential leader behaviors: articulating a vision, providing an appropriate model, fostering the acceptance of group goals, expecting higher performance, providing individualized support, and offering intellectual stimulation (Bass and Avolio 1990; Podsakoff et al. 1990). These potential leader behaviors are theorized to affect subordinates' behavior by first transforming subordinates' beliefs and attitudes. Transformational leadership is a process directed toward influencing changes in attitudes and assumptions held by organizational members, and building commitment for organizational goals and objectives (Yukl 1989). Thus, transformational leadership is a shared process, involving the actions of leaders across multiple organizational levels.
Theory suggests that transformational leadership is most likely to occur in organizations during periods of substantive changes. In other words, transformational leadership is one way for organizations to respond to increased economic competition and higher expectations for innovative products (Niehoff et al. 1990; Keller 1992; Howell and Avolio 1993). With a declining demand for audit services, CPA firms are under increasing economic pressure to provide new services to their clients (Melancon 1998). A review of the AICPA web site (http://www.aicpa.org) shows an aggressive marketing of new products and services, including e-business support, knowledge management, database technologies, information systems security, and new financial services. Economic conditions suggest that the opportunity exists for transformational leaders to emerge in the CPA profession.
CPAs who demonstrate transformational leadership may also choose to engage in transactional behavior, i.e., leader behavior that provides rewards and recognition contingent on subordinates obtaining specific performance levels, and intervenes when standards are not met (Bass 1985). For example, supervisors engaging in transactional behavior may provide structure and display consideration that indicates which subordinate performances will be rewarded (Hater and Bass 1988). Given that transformational leaders also engage in transactional behavior, the theoretical propositions developed below assume that the, predicted associations will be observed after controlling for the effects of transactional leader behavior.
Antecedents to Transformation Leadership
Most of the research on transformational leadership has focused on measuring the strength of the association between transformational leader behaviors and subordinate outcomes (measures of leader effectiveness). Very little research has examined antecedents to transformational leadership. Lowe et al. (1996) hypothesized that transformational leadership would be more prevalent at higher levels of management. However, based on their meta-analytic review of the literature, Lowe et al. (1996) found statistical evidence that lower levels of management, rather than higher levels of management, tended to have higher transformational leadership scores.
Overlooked in prior studies is whether the occurrence of transformational leadership differs across organizational functions and departments. Theory suggests that transformational leader behavior is more likely to occur in functional areas that are dynamic in nature and undergoing substantive change (Hater and Bass 1988; Yukl 1989; Niehoff et al. 1990; Burke and Church 1992). In the accounting profession, information systems assurance and business consulting services are very open-ended and continually evolving. Although financial auditing services are subject to intense economic competition, the product line in financial auditing is relatively more stable and is more like a commodity with easily substituted producers. Also, as argued by Summers et al. (2000), audit services are standard-intensive and guided by established rules (generally accepted auditing standards and accounting principles). Since the functional service areas of public accounting are likely to differ in the rate of organizational change an d adherence to established rules and procedures, the occurrence of transformational leadership behavior is likely to differ across those areas. In particular, the continuing evolving practice of information systems assurance and business consulting, accompanied with less standard-intensive rules and procedures, suggests that effective leaders in this nontraditional functional area are more likely to demonstrate transformational leadership. Conversely, "supervisors" in financial auditing would have fewer incentives for acquiring transformational leader behaviors. These arguments suggest the following hypothesis.
H1: Supervisors in the functional area of information systems assurance and business consulting services will demonstrate higher levels of transformational leadership, compared to supervisors in the functional area of financial auditing.
Transformational Leadership Outcomes
Even if empirical evidence is found supporting H1, a critical question remains: in the accounting profession, does transformational leadership provide employees with information that clarifies their organizational role and generates positive role-related outcomes? Prior organizational behavior research has examined the association of transformational leadership with numerous organizational outcomes. The current study identified four critical role-related outcomes relevant to transformational leadership theory and previously examined in behavioral accounting research--role clarity, job satisfaction, organizational commitment, and job performance. The potential relationship between these outcomes and transformational leadership is described below.
The extant literature extensively documents that transformational leadership is positively associated with higher levels of role clarity (Niehoff et al. 1990), dimensions of job satisfaction (Deluga 1988; Niehoffet al. 1990; Podsakoff et al. 1990; Seltzer and Bass 1990; Yammarino and Bass 1990; Bycio et al. 1995; Sosik 1997), organizational commitment (Niehoff et al. 1990; Bycio et al. 1995; Barling et al. 1996), and job performance (Avolio et al. 1988; Howell and Avolio 1993; Bycio et al. 1995; Kirkpatrick and Locke 1996; Sosik 1997; Sosik et al. 1997). However, this literature has generally not considered the causal relationships of these variables, i.e., which variables are theoretical antecedents for the others. For example, Podsakoff et al. (1996) meticulously examined the effect of six transformational leader behaviors on several organizational outcomes, including role ambiguity (reverse scoring of role clarity), general job satisfaction, organizational commitment, and in-role performance. In particular , the authors examined the moderating effect that leadership substitutes (such as organizational formalization) have on the relationship between transformational leader behaviors and organizational outcomes. However, omitted in Podsakoff et al.'s (1996) analysis is consideration of whether any of these organizational variables are antecedents for each other. To assess the relative effect of transformational leadership on role-related outcomes, statistical models must include other predictor variables relevant to role-related outcomes; otherwise, the effect of transformational leadership is likely to be overstated.
Role theory, as first articulated by Kahn et al. (1964), provides a model for ordering the effect of leadership and related behavioral variables. The Kahn et al. (1964) role episode model describes a cyclical process between role senders (e.g., supervisors) and focal persons (e.g., subordinates). Greater role clarity depends on two processes: (1) role senders providing information regarding their expectations, and (2) focal persons perceiving and responding to the role information. Less role clarity, which Kahn et al. (1964) labeled as role ambiguity, is attributable to three sources: (1) either role information is nonexistent, (2) has not been provided by role senders, or (3) has not been perceived by focal persons.
Leadership theory suggests that transformational leaders communicate information to subordinates regarding their (the subordinates') organizational role. For example, transformational leader messages have been depicted as being visionary, providing a model that is consistent with that vision, and fostering the acceptance of group goals, while also providing individualized support (Niehoffet al. 1990; Kirkpatrick and Locke 1996; Podsakoff et al. 1996; Sosik et al. 1997). If such messages are relevant to subordinates' roles, then subordinates whose supervisors' engage in transformational leader behavior should experience greater role clarity. Transformational leader behavior appears to fit the role episode model: (1) supervisors in dynamic functional areas have expectations regarding subordinates' roles, (2) they communicate those expectations through transformational leader behaviors, and (3) subordinates of transformational leaders have the opportunity to acquire greater information regarding their organizati onal role. Thus, the second hypothesis in this study is based on an extension of role theory: subordinates of supervisors who engage in transformational leadership experience greater role clarity.
H2: Transformational leadership is positively associated with role clarity.
Figure 1 presents the theoretical model developed in this study. The model identifies a theoretical link between transformational leadership, role clarity, and role outcomes, as well as a leader substitute--task analyzability. Each path in the model is identified by the relevant hypothesis tested.
As in other recent research (Gerstner and Day 1997; Major et al. 1995), this study labels variations in role information as role clarity, indicating that higher levels of role information support subordinates' understanding of their organizational role. However, a preponderance of prior studies labels this phenomenon as role ambiguity (Jackson and Schuler 1985), emphasizing the negative consequences when subordinates possess lower levels of role information. Regardless of the labeling, the relationship between subordinates' possession of role information and role outcomes is consistent: higher levels of role information are associated with higher levels of job satisfaction, organizational commitment, and job performance (Jackson and Schuler 1985). Behavioral accounting research has tended to support these findings. Role ambiguity (less role information) has been negatively associated with job satisfaction (Senatra 1980; Rebele and Michaels 1990; Gregson et al. 1994; Pasewark and Strawser 1996) and job perform ance (Rebele and Michaels 1990; Gregson et al. 1994). Although no accounting study has examined the association between role information and organizational commitment, studies have found that organizational commitment positively varies with job satisfaction (Gregson 1992; Poznanski and Bline 1997). As discussed in the "Measures" section below, the current study builds on the work of Ketchand and Strawser (1998) by considering three organizational commitment dimensions: affective commitment, high sacrifice commitment, and low alternatives commitment. The pattern of results from prior accounting studies suggests that role clarity (i.e, higher levels of role information) tend to be positively associated with role outcomes.
H3a: Role clarity is positively associated with job satisfaction, organizational commitment (as measured on three dimensions), and job performance.
While role theory suggests that transformational leadership is likely to have an indirect effect on role outcomes (i.e., transformational leadership is positively associated with role clarity, which is positively associated with role outcomes), transformational leadership theory offers arguments supporting a direct relationship between transformational leader behavior and role outcomes. Regarding job satisfaction, transformational leaders provide subordinates with alternative ways of looking at critical issues. Such encouragement of innovation is posited to be associated with higher levels of job satisfaction (Niehoff et al. 1990). Also, transformational leader behaviors are expected to enhance subordinates' work attitudes. Supervisor behavior that makes subordinates tasks more interesting should have a positive impact on general job satisfaction (Podsakoff et al. 1990).
There are several transformational leadership dimensions that support subordinates' obtaining higher levels of organizational commitment. For example, transformational leaders offer subordinates individualized support (such as "shows respect for subordinate feelings" and "behaves in a thoughtful manner") and foster acceptance of group goals (such as "encouraging employees to be team players" and "develops team attitude and spirit"). The inspirational aspects of transformational leadership are most likely to enhance the more emotion-based facets of organizational commitment (Bycio et al. 1995).
Finally, transformational leaders inspire subordinates to accomplish more difficult objectives, approach problems from new perspectives, and strive for performance levels that are higher than they (the subordinates) initially expected (Howell and Avolio 1993). Also, the inspirational aspects of transformational leadership are expected to lead to more original solutions (Sosik 1997). Bass (1985) suggested that the intellectual stimulation offered by transformational leaders is needed most when groups have to solve ill-structured problems, rather than well-structured problems. This assertion seems particularly relevant for the field of information systems assurance and business consulting where problems are likely to be more novel, less well structured, and less reliant on professional standards for solutions (Summers et al. 2000, 1). However, in general, transformational leader behaviors are expected to have a direct and positive effect on role outcomes.
H3b: Transformational leadership is positively associated with job satisfaction, organizational commitment (as measured on three dimensions), and job performance, after controlling for the effect of role clarity.
Control Variables: Task Analyzability and Transactional Leadership
As noted in this paper's introduction, prior accounting research has examined whether objective, computational procedures (task analyzability) and transactional exchanges between supervisors and subordinates are effective management tools. Bamber et al. (1989) found that lower task analyzability was associated with higher role ambiguity for audit seniors at "structured firms." Also, in studies of information systems development, higher task analyzability (less uncertainty concerning user requirements) was associated with more effective software development processes (Nidumolu 1996). Although the strength of the association may differ across functional areas, higher levels of task analyzability should be associated with greater role clarity (less role ambiguity).
As previously noted, transactional leader behavior provides subordinates with rewards and recognition contingent upon subordinates' obtaining specific performance levels, and intervenes when standards are not met. As in transformational leadership, transactional leader behaviors also communicate role expectations from supervisors to their subordinates. Certainly, transactional leader behavior disciplines subordinates when subordinate role performance does not meet supervisor expectations. One study (Podsakoff et al. 1993) found empirical evidence that contingent rewards from supervisors were negatively (positively) related to role ambiguity (role clarity). The preceding arguments suggest that both task analyzability and transactional leadership are expected to positively correlate with role clarity.
H4: Task analyzability and transactional leadership are positively associated with role clarity.
Currently, there is no theory to identify conditions under which task analyzability or transactional leadership would be directly associated with role outcomes, and whether the association would be negative or positive. Therefore, although the model presented in Figure 1 identifies a direct link between these two constructs and role outcomes, the link is identified as a research question (RQ1) and is essentially included in the model as a control. This study includes other control variables that act as a proxy for leader substitutes. For example, the statistical analyses include control variables measuring differences in type of CPA firm (e.g., Big 5, other national firms, regional firms, etc.), as well as office size (i.e., number of professionals in the office). (3) Theoretical predictions regarding the effect of these control variables are not developed.
III. RESEARCH METHODS
The following sections describe participant demographics and measurement scales sources, metrics, and reliability coefficients. The data for this field study were collected through a survey mailed to members of the American Institute of Certified Public Accountants (AICPA) who had identified an interest in both (1) management information systems, and (2) consulting (computer/business systems). Measurement scales used in this study were previously tested and reported in the extant literature on leadership, role development, and role-related outcomes.
Participants
Initial selection was limited to AICPA members who (1) worked for public accounting firms with at least 20 AICPA members, and (2) were identified as employees (nonpartners in the firms). (4) In total, 1,150 AICPA members were identified as potential participants. (5) Several surveys (n = 45) were returned marked "no longer at this firm" or "no longer in public accounting." Thus, 1,105 is used as the baseline for potential respondents.
The survey was designed and implemented based on procedures recommended in Dillman (2000). Potential participants received a cover letter explaining the purpose of the study, a multi-page survey questionnaire, (6) and a postage-paid return envelope. As recommended in Dillman (2000), follow-up procedures included a one-week reminder postcard to all potential participants, and a four-week replacement questionnaire to all nonrespondents. (7) A total of 416 completed surveys were returned, representing a 37.6 percent response rate (416 out of 1,105 potential respondents). (8) Table 1 presents the demographic background for the 416 participants.
Public accounting employees often work for more than one supervisor during a fiscal year. Participants in this study were asked to evaluate a referent supervisor: the supervisor they had worked for the most over the prior 12 months. Since supervisors may also act as mentors to subordinates (Tepper 1995), and public accounting mentors appear to provide role clarifying information (Viator 2001), participants were asked to identify whether they consider their referent supervisor to be a mentor, and whether any one else in the firm had been a mentor to them in the prior 12 months.
Measures
Pretests using eight practicing CPAs from regional and Big 5 firms, and ten accounting doctoral students with public accounting experience, were conducted to ensure that each scale item was applicable and reflected wording commonly used in the profession. In some cases, minor wording changes were made based on feedback received. All scale items, with the exception of job performance as noted below, were scored on a five-point Likert scale from "strongly disagree" to "strongly agree." Appendices A and B list relevant questionnaire items, along with their means, standard deviations, and measurement model parameter estimates from confirmatory factor analysis. The following subsections provide additional information regarding measurement scales for (1) leader behaviors, (2) role clarity, (3) task analyzability, (4) job satisfaction, (5) organizational commitment, (6) job performance, and (7) other demographic variables.
Leader Behaviors
This study used Podsakoff et al.'s (1990) leader behavior inventory to assess both transformational and transactional leader behaviors. These scales are designed to measure six key dimensions of transformational leadership identified in the research literature, as well as one dimension for transactional leadership. Podsakoff et al. (1996) confirmed the psychometric properties of the transformational leadership scales and demonstrated that the hypothesized six-factor model fit the data significantly better than models with fewer factors. The six transformational leadership dimensions are: (1) articulating a vision; (2) providing an appropriate model; (3) fostering the acceptance of group goals; (4) high performance expectations; (5) providing individualized support; and (6) intellectual stimulation.
The final measurement model, using structural equation modeling (SEM), yielded six separate transformational leadership factors (one for each subscale), a single transactional leadership factor, and a second-order transformation leadership factor (defined by the six separate transformational leadership factors).
Because the issues raised in this study relate to a single construct for transformational leadership, the second order factor was used in tests of hypotheses. As reported in Appendix A, all measurement model parameter estimates were significant and only one observation variable had a significant cross-loading (i.e., "has a clear understanding of where we are going" loaded on both articulating a vision and providing an appropriate model). The Cronbach alpha coefficients for all seven leader behavior subscales were acceptable and ranged from 0.73 to 0.91, exceeding the minimum 0.70 cut-off recommended by Nunnally (1978).
Role Clarity
The survey included a modified version of the six-item Rizzo et al. (1970) scale. The current study does not reverse score scale-items and thus reports the measurement as "role clarity" (Gerstner and Day 1997; Major et al. 1995). The scale has been used in prior accounting studies that have reverse scored scale-items to report role ambiguity (Senatra 1980; Bamber et al. 1989; Rebele and Michaels 1990; Gregson et al. 1994; Viator 2001). The Cronbach alpha coefficient was 0.89, indicating acceptable reliability.
Task Analyzability
This study used a modified version of Withey et al.'s (1983) scale for task analyzability. The scale is based on Perrow's (1967) conceptualization of technology and has achieved acceptable reliability in other studies (Rice 1992; Abernethy and Brownell 1997). Although the scale correlates highly with the Daft and Macintosh (1981) scale used by Bamber et al. (1989), Withey et al. (1983) found their proposed scale contained higher face validity and demonstrated higher internal consistency. The Cronbach alpha coefficient obtained in the current study was 0.78, which compares favorably to the 0.72 Cronbach alpha obtained by Bamber et al. (1989).
Job Satisfaction
Job satisfaction was measured using six items from Rusbult and Farrell (1983), and previously included in an accounting-based organizational commitment study by Ketchand and Strawser (1998). The scale contained items relating to satisfaction with current assignments and responsibilities, as well as overall satisfaction with employer and type of work. The Cronbach alpha coefficient obtained in the current study was 0.89.
Organizational Commitment
Following the work of Ketchand and Strawser (1998), this study used Meyer and Allen's (1984) multidimensional measure of organizational commitment. The scales have been extensively evaluated in nonaccounting organizations (Meyer et al. 1989; Allen and Meyer 1990; Meyer et al. 1991). The eight-item affective commitment scale asked questions relating to participants' "sense of belonging to the firm" and "feeling emotionally attached to the firm." The Cronbach alpha for affective commitment was 0.91, indicating acceptable reliability. The eight items measuring continuance commitment contained four items regarding high sacrifice commitment (e.g., "too much in my life would be disrupted if I decided I wanted to leave my firm now") and four items regarding low alternatives commitment (e.g., "I feel I have too few options to consider leaving this firm"). The initial Cronbach alphas were relatively low (0.69 and 0.67, respectively), indicating moderate internal consistency. Further analysis indicated that the Cronbac h alphas could be improved to acceptable levels by deleting one item in each scale. (9) Thus the final measurement of continuance commitment contained six items: three high-sacrifice items and three low-alternative items, with Cronbach alphas of 0.79 and 0.71, respectively. See Appendix B for items included in the final measurement model and relevant parameter estimates.
Job Performance
The 13-item job performance scale was primarily adopted from Choo (1986), but also included two supplementary items from Rebele and Michaels (1990): "developing practical solutions to problems encountered on an engagement or project" and "comprehending and demonstrating an interest in clients' business." Each item was measured using a five-point Likert scale ranging from "unsatisfactory" to "outstanding." Although self-reported measures of job performance should be interpreted with caution, such measures have been used frequently in accounting studies (Choo 1986; Rebele and Michaels 1990; Rasch and Harrell 1990; Kalbers and Fogarty 1995; Viator 2001). The Cronbach alpha coefficient obtained in the current study was 0.87.
Demographic Variables
The statistical models used variables related to participant functional areas and demographic background that are expected to vary with transformational leadership, role clarity, and role outcomes. Dummy variables identifying participant demographics include gender (0 = male, and 1 = female) and participant organizational level (manager, senior manager, and partner, where senior accountant/consultant is the comparison category). Continuous variables relating to participant demographics included age and years in public accounting.
Participants were classified into four practice areas. ISA/BC (participants working in information systems assurance and business consulting) represents the participant group of primary interest. These participants indicated that their primary practice area was either information systems auditing, information systems consulting, or business consulting. (10) Other practice areas identified by dummy variables include tax services and other-functional areas. The remaining participants worked in financial audit services and served as the comparison group.
Participant firms were classified as other-national, regional, and local (with Big 5 as the comparison category). Participant office size was classified into five groups with the comparison category being offices with greater than 200 professionals. The four dummy variables representing subsequently smaller offices included office_<200, office_<100, office <50, and office_<25.
The presence of informal mentors has been found to be associated with less role ambiguity (higher role clarity) (Viator 2001). Therefore, this study included dummy variables related to mentoring. Referent supervisor mentor identifies participants who viewed their supervisor as a mentor; and, other mentor indicates participants who viewed other persons at the firm as being mentors to them. In the regression analysis of transformational leadership, only referent supervisor mentor is relevant to the analysis and included in the model since the existence of "other mentor" is not expected to affect the level of transformational leadership provided by referent supervisors.
IV. STATISTICAL ANALYSES AND RESULTS
Linear regression analysis was used to test the first hypothesis using one-tailed test of significance for Hl. The model presented in Figure 1, which has multiple dependent variables with potentially high covariation, suggests that structural equation modeling (SEM) is a more appropriate tool for testing H2 through H4, and RQ1. SEM was conducted with the Mplus modeling program (Muthen and Muthen 1998). All analyses used the maximum likelihood method and eliminated insignificant paths based on [alpha] = 0.05.
Manipulation Check
A key assumption of this study is that ISA/BC participants (information systems assurance and business consulting participants) work in an environment where fewer tasks are executable by objective, computational procedures, compared to participants from traditional financial audit services. If this assumption is true, then a potential observation is for ISA/BC participants to report levels of task analyzability that are substantially less than that reported by participants working primarily in financial audit services. Table 2 presents the results of a regression analysis that tests this assumption while controlling for other variables that may influence the level of task analyzability.
The beta coefficient for task analyzability for ISA/BC was negative and statistically significant ([beta] = -2.180, p < 0.001), supporting the assertion that the ISA/BC group perceived their work as having substantially lower levels of task analyzability, compared to participants working primarily in financial audit services. Participants in other areas (tax services and other functional areas) also perceived less task analyzability compared to those in financial audit services. However, a statistical contrast of betas indicated that participants in ISA/BC reported lower levels of task analyzability compared to participants in tax services and other functional areas (p < 0.01, and p < 0.05, respectively).
Some control variables were also statistically significant. Participants at other-national, regional, and local firms (vs. participants from Big 5 firms) indicated lower levels of task analyzability, although only the beta coefficient for regional firms was statistically significant. The negative beta coefficients suggest that participants at Big 5 firms have access to methodologies that provide more objective, computational procedures in analyzing client needs and performing client service. Two variables with significantly positive beta coefficients were years in public accounting and referent supervisor mentor, indicating that longer tenure in public accounting and having supervisors who provide mentoring guidance are associated with higher levels of task analyzability.
Test of Hypotheses: Regression Analysis
Hypothesis 1 states that supervisors in the functional area of ISA/BC will demonstrate higher levels of transformational leader behavior, compared to supervisors in the functional area of financial auditing. Table 2 presents the results of regression analysis for the dependent variable transformational leadership. The variable ISA/BC had a positive and statistically significant beta coefficient ([beta] = +2.058, p <0.05), indicating support for H1. Also, a statistical contrast of betas indicated that ISA/BC participants reported higher levels of transformational leadership compared to tax services participants (p < 0.10).
The regression analysis of transformational leadership identified other statistically significant coefficients. Variables positively associated with transformational leadership included referent supervisor mentor ([beta] = +12.518, p < 0.001) and local firm ([beta] = +4.710, p < 0.05). Of particular interest is the mentorship variable, whose statistical significance suggests that supervisor-mentors in public accounting are strongly associated with transformational leadership. This association implies that functions performed by mentors go beyond the two basic functions attributed to mentors: career support and psychosocial support (Kram 1988). The results suggest that mentors may also provide support that transforms subordinates' perceptions of their work environment and organizational role.
The positive beta coefficient for local firm indicates that participants in small CPA firms perceived their referent supervisor as providing more transformational leadership, compared to participants from Big 5 firms (the comparison category). Interestingly, participants located in smaller offices (represented by four dummy variables) indicated receiving less transformational leadership, compared to participants in offices with more than 200 professionals (the comparison category). The significant effects for local firm and office size are somewhat inversed, since participants from local firms tended to be from smaller offices (90.6 percent were from offices with less than 100 professionals). Essentially, the negative effect of being in a smaller office is offset by the positive effect of working for a local firm, possibly because supervisors at local firms have more influence over strategic issues. Other variables negatively associated with transformational leadership included participants who were managers (as compared to seniors) ([beta]= -2.934, p <0.05) and age ([beta] = -0.149, p < 0.10), indicating that older participants perceived less transformational leadership.
Test of Hypotheses: Structural Equation Modeling
Prior research indicates that the demographic variables included in this study are likely to covary with the dependent latent variables used in the structural equation model (SEM). To remove the effect of these demographic variables, each observation variable measuring a dependent latent variable was regressed on the demographic predictor variables. The residuals became input for the SEM. For example, the variable "my current work compares very well to my ideal job" (an observation variable related to the latent construct "job satisfaction") was regressed on demographic predictor variables. The residuals were then included as one measure of job satisfaction in the structural equation model.
To provide a demonstration of the effect of this procedure, Table 3 presents the results of regressing summary measures of role clarity and related role outcomes on the demographic predictor variables. All models were statistically significant. There was only one predictor variable that was statistically significant across all six regressions: referent supervisor mentor. This procedure removes the effect of mentoring and confirms conclusions from Viator (2001) that having a mentor is associated with greater role clarity. (11)
SEM was performed using data from all 416 participants. The Mplus modeling software provides two measures of model fit: the Chi-square test of model fit and the root mean square error of approximation. The Chi-square test statistic was 3375.2 with 2120 degrees of freedom, resulting in a ratio of 1.59. Good fitting models evidence ratios of 2.0 or less (Wheaton et al. 1977). The root mean square error of approximation was only 0.039 with tight 90 percent confidence bands (0.036 to 0.041) that did not exceed the target upper bound of 0.05. Thus, the overall fit statistics indicate that the proposed model fit the data reasonably well.
Univariate tests indicated that 14 structural paths in the model were insignificant. Removing these 14 paths from the model resulted in an increase in Chi-square of only 16.8, a statistical insignificant amount. Thus, the 14 paths removed were multivariately insignificant. Figure 2 presents the final model's structural path parameter estimates, including estimates of covariation among the dependent variables.
Hypotheses 2 and 4 predict that transformational leadership, task analyzability, and transactional leadership are positively associated with role clarity. The parameter estimates for all three predictor variables were positive and statistically significant (0.285, 0.709, and 0.184, respectively), providing support for both H2 and H4. It is important to note that the structural path estimates indicate an effect after controlling for the other effects in the model. In other words, transformational leadership was positively associated with role clarity, after controlling for task analyzability and transactional leadership, which were also positive and statistically significant.
Hypotheses 3a and 3b predict that role clarity and transformational leadership are positively associated with several role outcomes--job satisfaction, organizational commitment, and job performance. Figure 2 presents the relevant structural path parameter estimates for H3a and H3b. Regarding H3a, the parameter estimates for role clarity were positive and statistically significant for job satisfaction (0.463), affective commitment (0.469), and job performance (0.118), but not significant for either continuance commitment scale (p > .05). The statistical results provide general support for H3a.
Regarding H3b, the structural path parameter estimates for transformational leadership were positive and statistically significant for job satisfaction (0.456) and affective commitment (0.513), providing limited support for H3b. Transformational leadership was not directly associated with either high sacrifice commitment or job performance (p > 0.05). For the participants in this study, transformational leadership's effect on job performance was limited to an indirect effect: transformational leadership was positively associated with role clarity, which was positively associated with job performance. This result suggests that transformational leaders provide subordinates with higher levels of role clarity, which then translates into higher levels of job performance.
The SEM results indicated that transformational leadership was negatively associated with low alternatives commitment (-0.475). Further consideration of the continuance commitment scales suggests that the results may be meaningful. Both continuance commitment scales suggest an atmosphere of entrapment: either the subordinate cannot afford to leave (high sacrifice), or the subordinate has few options (low alternatives). Participants in the current study who perceived higher levels of transformational leadership tended to identify lower levels of low alternatives continuance commitment: they actually perceived having "more alternatives." This is especially interesting given that higher levels of transformational leadership were associated with higher affective commitment, such as "I would be very happy to spend the rest of my career with this firm." The combined results suggest that transformational leadership has a positive impact on the more emotionally laden aspect of organizational commitment, and simultane ously provides subordinates with a view of low sacrifice for leaving the firm. Transformational leadership in public accounting appears to result in subordinates feeling emotionally committed to the firm, but not feeling trapped there, recognizing there are viable alternatives outside of the firm.
RQ1 indicated that both task analyzability and transactional leadership may be associated with role-related outcomes, but did not offer a directional prediction. The SEM results reported in Figure 2 indicate that neither task analyzability nor transactional leadership was statistically associated with any of the five role outcomes. Thus the direct effect of task analyzability and transactional leadership to role-related outcomes appears to be nonexistent. These results support the assertion that transformational leadership (rather than classic transactional leadership or leader substitutes such as task analyzability) plays a substantive role in creating higher levels of subordinate job satisfaction and emotional commitment to the firm.
Additional Analysis
A remaining question is whether leader behavior has a differential effect across functional areas. To investigate the potential moderating effect of functional area, this study developed structural models for three separate subgroups: ISA/BC, tax services, and financial audit services. (12) Parameter estimates were tested for equivalency across the three subgroups, and resulted in an integrated model with a single estimate of model fit. Model fit statistics indicated that partitioning the data into subgroups resulted in poorer model fit. The ratio of the Chi-square test statistic to the model's degrees of freedom was essentially unchanged (1.58, compared to the original ratio of 1.59); however the root mean square error of approximation increased to 0.069, moving above the accepted target of 0.05.
Aside from resulting in a weaker fitting model, the subgroup analysis indicated no difference across subgroups in regard to the structural path parameter estimates defined by transformational leadership (H2 and H3b). (13) There were only two structural path parameter estimates that differed among the three subgroups. First, for the ISA/BC group, role clarity was positively associated with high sacrifice commitment (0.365), whereas role clarity held no association with high sacrifice commitment for the other two subgroups. This suggests that for individuals in the ISA/BC group, as their role in the organization is made clearer, they perceive a higher sacrifice associated with leaving the firm. This is a nontrivial result given that transformational leadership, as well as task analyzability and transactional leadership, were positively associated with role clarity. In other words, as supervisors of ISA/BC professionals provide role clarity to their subordinates, those professionals perceive a higher sacrifice f or leaving the firm.
Second, for the financial audit services participants, role clarity was negatively associated with low alternatives commitment (-0.367), whereas role clarity held no association with low alternatives for the other two subgroups. Once again, the positive association of leader behavior with role clarity suggests that as supervisors of financial audit services provide more role clarity, their subordinates are more likely to perceive that low alternatives are not a barrier to leaving the firm.
V. DISCUSSION OF FINDINGS AND FUTURE RESEARCH
As noted in the paper's introduction, economic pressure and competition are driving CPAs to provide more diverse services to their clients. The AICPA's Vision Project acknowledges that leadership skill (defined as the ability to influence, inspire, and motivate others to achieve results) is a critical skill needed for transforming the profession in the 21st century. Leader behaviors that are transformational in nature (e.g., provide inspiration, intellectual stimulation, and individualized support) are one organizational mechanism for responding to increased economic competition and demand for innovative products. However, increased economic competition does not ensure the emergence of transformational leadership. In fact, very little research has examined organizational and environmental conditions that facilitate the emergence of transformational leadership.
This paper proposes that the dynamic and substantive changes occurring in the public accounting profession provide an opportunity for the emergence of transformational leadership. If such leadership emerges, then it should be associated with positive organizational outcomes. As predicted, the study found that participants from the dynamic area of information systems assurance and business consulting reported higher levels of transformational leadership, compared to participants from either financial auditing or tax services. Furthermore, the paper provides some evidence that transformational leadership in public accounting is likely to enhance organizational performance. Transformational leadership was directly and positively associated with subordinates' role clarity, job satisfaction, and affective commitment (and negatively associated with low alternatives commitment), after accounting for the effects of task analyzability and transactional leadership. However, in the long term, the critical issue is wheth er transformational leader behaviors make a difference in the economic performance of CPA firms, especially with regard to growth in client services. Future research should examine whether partners and managers who demonstrate transformational leadership do, in fact, produce higher economic returns, i.e., higher revenue, higher client satisfaction, and an expanding client base.
There are several methodological issues that should be addressed in future research of transformational leadership in nontraditional accounting services. Leader behaviors are often measured by obtaining subordinates' perception of supervisors' leadership. These measures are admittedly biased by the degree to which subordinates actually perceive leader behaviors. However, several alternatives are available regarding (1) leadership outcomes, and (2) measurement of leadership outcomes. The current study examined leadership outcomes in regard to subordinates' role outcomes--role clarity, job satisfaction, organizational commitment, and job performance. As previously argued, other leadership outcomes, such as leaders' job performance, are potentially more relevant in assessing the value of transformational leader behaviors. Measurement of leaders' job performance can be obtained from sources independent of subordinates' perception of leader behaviors. At a minimum, future studies should obtain data from supervisor /subordinate dyads and match supervisors' self-reported job performance with subordinates' perception of supervisor leader behavior. A more rigorous method would obtain supervisors' actual job performance evaluations: CPA firms' rating of referent supervisors regarding development of client services, client satisfaction, and revenue growth. Obtaining such data would require firm participation and the promise of complete anonymity for participating staff, managers, and partners (vs. the assurance of confidentiality used in the current study).
Further investigation of transformational leadership in public accounting's nontraditional service areas appears warranted given other research demonstrating that leaders can acquire transformational leadership skills through leadership training. Using a pretest-posttest experimental design, Barling et al. (1996) found that subordinates of managers receiving transformational leadership training subsequently rated their manager significantly higher on three transformational leadership dimensions (intellectual stimulation, individualized consideration, and charisma). Also, the training provided to participants was associated with positive "real world" economic outcomes: organizational units managed by persons trained in transformational leadership obtained higher levels of financial performance. Future accounting research should further examine the acquisition of transformational leader behavior and its association with growth in client services and client base. As identified by the AICPA's Vision Project, futu re client services are likely to be more diverse, and competencies needed by CPAs to provide these services will continue to evolve. For example, public accounting firms are pursuing services that assist clients in managing risks associated with development and implementation of information technology solutions, including implementation of enterprise resource planning systems (ERPs), telecommunications and network services, e-business solutions, and integrated enterprise-wide security management systems. (14) Whether the profession obtains growth in nontraditional services is dependent on having leaders (managers and partners) who can develop and successfully implement new services. Such leaders are, by definition, transformational.
(1.) "Objective, computational procedures" characterize highly analyzable tasks (Daft and Macintosh 1981). When work is less analyzable, individuals are likely to spend more time thinking about what to do and searching for solutions beyond normal procedures. Tasks that rely on established bodies of knowledge (such as general accepted auditing standards) are considered to have high task analyzability (Scott and Tiessen 1999; Summers et al. 2000)
(2.) Supervisors "initiate structure" when they define subordinates' roles in regards to goal attainment, and provide "consideration" when they demonstrate respect for subordinates' ideas and feelings (Jiambalvo and Pratt 1982, 737).
(3.) Other leader substitutes, such as professional orientation and formalization (Kerr and Jermier 1978; Podsakoff and MacKenzie 1994), were not expected to differ widely among CPAs, and were not included in the study. Survey design constraints (e.g., limitations on the number of pages included in the survey) restricted the survey to measurement scales most likely to vary across practicing CPAs.
(4.) The final sample included responses from 32 partners/directors, suggesting that the AICPA database contains some demographic data that is not current.
(5.) Funding for this study supported mailing survey questionnaires to as many as 1,200 AICPA members. The initially targeted participant pool were AICPA members who worked at large CPA firms (i.e., firms with at least 100 AICPA members) and had identified an interest in both management information systems and consulting services. Because of the relatively low numbers meeting these criteria, the pool of candidates was increased to 1,150 by including members at smaller firms (i.e., firms with at least 20 AICPA members).
(6.) Copies of the survey instrument are available from the author.
(7.) Unique identifying numbers were used to track nonrespondents. Cover letters sent to potential participants acknowledged use of the tracking numbers. However, it is possible that, in spite of promised confidentiality, some potential participants chose not to respond because of the sensitive nature of the survey items (e.g., commitment to the firm questions).
(8.) Kerlinger (1986) indicates that the expected response rate from mail surveys is 20 to 40 percent. Following Oppenheim's (1966) recommendation, several statistical tests were performed to identify differences between early respondents (n = 292) and late respondents (n = 124). Late respondents represented those surveys returned after the four-week replacement questionnaire had been mailed to all nonrespondents. Based on Chi-squared tests, no differences were identified regarding respondent organizational level, primary functional area, firm size, office size, and mentorship (p > 0.10). Based on ANOVA tests, no differences were found for levels of task analyzability, transformational leadership, role clarity, and measures of role outcomes (p > 0.10). Late respondents had a higher percentage of females (35.5 percent vs. 23.6 percent, p < 0.05) and slightly higher average experience in public accounting (11.8 years vs. 10.3, p < 0.05).
(9.) Both items deleted from thc continuance commitment scales required reverse scoring. It is possible that these two items had relatively low factor loadings because participants did not recognize thc reverse wording and failed to adjust their responses.
(10.) Participants identified as primarily working in business consulting had checked "other" as their primary work area and wrote descriptions indicating business consulting. Another 37 participants who cheeked "other" as their primary work area wrote descriptions that varied, but included work in business valuation, technical support, and other functional areas.
(11.) The current study extends Viator (2001) by focusing on (1) whether transformational leadership is more likely to occur in nontraditional accounting service areas, and (2) whether such leadership provides role information beyond that supplied by mentoring relationships. Although the effect of mentors on role clarity was replicated in the current study, the data set is substantially different from that reported in Viator (2001). In the current study, 24.0 percent of the respondents indicated an interest in ISA/BC and 51 percent were from regional and local CPA firms; whereas in the earlier study, only 8.5 percent indicated working in consulting services and 93.2 percent were from Big 6 public accounting firms. Also, in the current study vs. earlier study, participants were older (median age of 35 vs. 29), consisted of more males (72.8 percent vs. 44.8 percent), and tended to be at higher organizational levels (75.7 percent were above senior/staff vs. 60.2 percent).
(12.) The number of participants in other functional areas was too small (n = 40) to develop an SEM model for that subgroup. However, the SEM analysis on subgroups was rerun by combining participants from other functional areas (n = 40) with thc ISA/BC (n = 100). The results did not differ in substance from those reported in the text.
(13.) A limitation of the current study is that all participants, including those who work primarily in financial auditing and tax services, had indicated an interest in management information systems and consulting services. The participant pool did not include CPAs who work in traditional services and have a more limited interest in nontraditional services. The homogeneity of the participant pool may have created a bias against finding differences across functional areas. Perhaps because of this bias, the strength of the hypothesized relationships differed only slightly across functional areas.
(14.) Additional examples of emerging services arc advertised at the web sites of large public accounting firms.
REFERENCES
Abernethy, M. A., and P. Brownell. 1997. Management control systems in research and development organizations: The role of accounting, behavior, and personnel controls. Accounting, Organizations and Society 22 (3/4): 233-248.
Allen, N. J., and J. P. Meyer. 1990. The measurement and antecedents of affective, continuance and normative commitment to the organization. Journal of Occupational Psychology 63 (March): 1-18.
American Institute of Certified Public Accountants (AICPA). 2000. CPA Vision Project. Available at http://www.aicpa.org/vision/index.htm.
Avolio, B. J., D. A. Waldman, and W. O. Einstein. 1988. Transformational leadership in a management game simulation: Impacting the bottom line. Group & Organization Studies 13 (1): 59-80.
Bamber, E. M., D. Snowball, and R. M. Tubbs. 1989. Audit structure and its relation to role conflict and role ambiguity: An empirical investigation. The Accounting Review 64 (2): 285-299.
Barling, J., T. Weber, and E. K. Kelloway. 1996. Effects of transformational leadership training on attitudinal and financial outcomes: A field experiment. Journal of Applied Psychology 81(6): 827-832.
Bass, B. 1985. Leadership and Performance Beyond Expectations. New York, NY: The Free Press.
-----, and B. J. Avolio. 1990. Transformational Leadership Development: Manual for the Multifactor Leadership Questionnaire. Palo Alto, CA: Consulting Psychologists Press, Inc.
Burke, W. W., and A. H. Church. 1992. Managing change, leadership style, and intolerance to ambiguity: A survey of organization development practitioners. Human Resources Management 31(4): 301-318.
Burns, J. M. 1978. Leadership. New York, NY: Harper & Row.
Bycio, P., R. D. Hackett, and J. S. Allen. 1995. Further assessments of Bass's (1985) conceptualization of transactional and transformational leadership. Journal of Applied Psychology 80 (4): 468-478.
Choo, F. 1986. Job stress, job performance, and auditor personality characteristics. AUDITING: A Journal of Practice & Theory 5 (2): 17-34.
Cushing, R. E., and J. K. Loebbecke. 1986. Comparison of Audit Methodologies of Large Accounting Firms. Accounting Research Study No. 26. Sarasota, FL: American Accounting Association.
Daft, R. L., and N. B. Macintosh. 1981. Tentative exploration into the amount and equivocality of information processing in organizational work units. Administrative Science Quarterly 26 (June): 207-224.
Deluga, R. J. 1988. Relationship of transformational and transactional leadership with employee influencing strategies. Group & Organization Studies 13 (4): 456-467.
Dillman, D. A. 2000. Mail and Internet Surveys: The Tailored Design Method. New York, NY: John Wiley & Sons, Inc.
Elliott, R. K. 2000. Who are we as a profession--And what must we become? Journal of Accountancy 189 (2): 81-85.
Gerstner, C. R., and D. V. Day. 1997. Meta-analytic review of leader-member exchange theory: Correlates and construct issues. Journal of Applied Psychology 82(6): 827-844.
Gregson, T. 1992. An investigation of the casual ordering of job satisfaction and organizational commitment in turnover models in accounting. Behavioral Research in Accounting 4: 80-95.
-----, J. Wendell, and J. Aono. 1994. Role ambiguity, role conflict, and perceived environmental uncertainty: Are the scales measuring separate constructs for accountants? Behavioral Research in Accounting 6: 144-159.
Hater, J. J., and B. M. Bass. 1988. Superiors' evaluation and subordinates' perceptions of transformational and transactional leadership. Journal of Applied Psychology 73 (4): 695-702.
Howell, J. M., and B. J. Avolio. 1993. Transformational leadership, transactional leadership, locus of control, and support for Innovation: Key predictors of consolidated-business-unit performance. Journal of Applied Psychology 78(6): 891-902.
Jackson, S. E., and R. S. Schuler. 1985. A meta-analysis and conceptual critique of research on role ambiguity and role conflict in work settings. Organizational Behavior and Human Decision Processes 36: 16-78.
Jiambalvo, J., and J. Pratt. 1982. Task complexity and leadership effectiveness in CPA firms. Accounting Review (October): 734-750.
Kahn, R. L., D. M. Wolfe, R. P. Quinn, J. D. Snoek, and R. A. Rosenthal. 1964. Occupational Stress: Studies in Role Conflict and Ambiguity. New York, NY: Wiley.
Kalbers, L. P., and T. J. Fogarty. 1995. Professionalism and its consequences: A study of internal auditors. AUDITING: A Journal of Practice & Theory (14): 64-86.
Keller, R. T. 1992. Transformational leadership and the performance of research and development project groups. Journal of Management 18 (3): 489-501.
Kelley, T., and L. Margheim. 1990. The impact of time budget pressure, personality, and leadership variables on dysfunctional auditor behavior. AUDITING: A Journal of Practice & Theory 9 (2): 21-42.
Kerlinger, F. N. 1986. Foundation of Behavioral Research. Third edition. Fort Worth, TX: Harcourt, Brace Jovanovich Publishers.
Kerr, S., and J. M. Jermier. 1978. Substitutes for leadership: Their meaning and measurement. Organizational Behavior and Human Performance 22: 375-403.
Ketchand, A. A., and J. R. Strawser. 1998. The existence of multiple measures of organizational commitment and experience-related differences in a public accounting setting. Behavioral Research in Accounting 10: 109-137.
Kida, T. E. 1984. Performance evaluation and review meeting characteristics in public accounting firms. Accounting, Organizations and Society 9 (2): 137-147.
Kirkpatrick, S. A., and E. A. Locke. 1996. Direct and indirect effects of three core charismatic leadership components on performance and attitudes. Journal of Applied Psychology 81(1): 36-51.
Kram, K. E. 1988. Mentoring at Work: Developmental Relationships in Organizational Life. Lanham, MD: University Press of America, Inc.
Lowe, K. B., K. G. Kroeck, and N. Sivasubramaniam. 1996. Effectiveness correlates of transformational and transactional leadership: A meta-analytic review of the MLQ literature. Leadership Quarterly 7 (3): 385-425.
Major, D. A., S. W. J. Kozlowski, G. T. Chao, and P. D. Gardner. 1995. A longitudinal investigation of newcomer expectations, early socialization outcomes, and the moderating effects of role development factors. Journal of Applied Psychology 80 (3): 418-431.
Melancon, B. C. 1998. The changing strategy for the profession, the CPA and the AICPA: What this means for the education community. Accounting Horizons 12 (4): 397-406.
Meyer, J. P., and N. J. Allen. 1984. Testing the "side-bet" of organizational commitment: Some methodological considerations. Journal of Applied Psychology 69 (August): 372-378.
-----, S. V. Paunonen, I. R. Gellatly, R. D. Goffin, and D. N. Jackson. 1989. Organizational commitment and job performance: It's the nature of the commitment that counts. Journal of Applied Psychology 74 (December): 152-156.
-----, D. R. Bobocel, and N.J. Allen. 1991. Development of organizational commitment during the first year of employment: A longitudinal study of pre- and post-entry influences. Journal of Management 17 (December): 717-733.
Muthen, L. K., and B. O. Muthen. 1998. Mplus: The Comprehensive Modeling Program for Applied Researchers. Los Angeles, CA: Muthen & Muthen.
Nidumolu, S. R. 1996. Standardization, requirements uncertainty and software project performance. Information & Management 31: 135-150.
Niehoff, B. P., C. A. Enz, and R. A. Grover. 1990. The impact of top-management actions on employee attitudes and perceptions. Group & Organization Studies 15 (3): 337-352.
Nunnally, J. C. 1978. Psychometric Theory. New York, NY: McGraw-Hill Book Company.
Oppenheim, A. N. 1966. Questionnaire Design and Attitude Measurement. New York, NY: Basic Books, Inc.
Otley, D. T., and B. J. Pierce. 1995. The control problem of public accounting firms: An empirical study of the impact of leadership style. Accounting, Organizations and Society 20 (5): 405-420.
Pasewark, W. R., J. R. Strawser, and J. E. Wilkerson, Jr. 1994. An empirical examination of the relationships among leader behaviors, audit team performance, and staff satisfaction. Advances in Accounting 12:143-166.
-----, and -----. 1996. The determinants and outcomes associated with job insecurity in a professional accounting environment. Behavioral Research in Accounting 8: 91-113.
Perrow, C. 1967. A framework for the comparative analysis of organizations. American Sociological Review (April): 194-208.
Podsakoff, P. M., S. B. MacKenzie, R. H. Moorman, and R. Fetter. 1990. Transformational leader behaviors and their effects on followers' trust in leader, satisfaction, and organizational citizenship behaviors. Leadership Quarterly 1 (2): 107-142.
-----, B. P. Niehoff, S. B. MacKenzie, and M. L. Williams. 1993. Do substitutes for leadership really substitute for leadership? An empirical examination of Kerr and Jermier's situational leadership model. Organizational Behavior and Human Decision Processes 54: 1-44.
-----, and S. B. MacKenzie. 1994. An examination of the psychometric properties and nomological validity of some revised and reduced substitutes for leadership scales. Journal of Applied Psychology 79 (5): 702-713.
-----, -----, and W. H. Bommer. 1996. Transformational leader behaviors and substitutes for leadership as determinants of employee satisfaction, commitment, trust, and organizational citizenship behaviors. Journal of Management 22 (2): 259-298.
Poznanski, R. J., and D. M. Bline. 1997. Using structural equation modeling to investigate the causal ordering of job satisfaction and organizational commitment among staff accountants. Behavioral Accounting Research 9: 154-171.
Pratt, J., and J. Jiambalvo. 1981. Relationships between leader behaviors and audit team performance. Accounting. Organizations and Society 6 (2): 133-142.
-----, and -----. 1982. Determinants of leader behavior in an audit environment. Accounting, Organizations and Society 7 (4): 369-379.
Rasch, R. H., and A. Harrell. 1990. The impact of personal characteristics on the turnover behavior of accounting professionals. AUDITING: A Journal of Practice & Theory 9 (Spring): 90-102.
Rebele, J. E., and R. E. Michaels. 1990. Independent auditors' role stress: Antecedent, outcome, and moderating variables. Behavioral Research in Accounting 2: 124-153.
Rice, R. E. 1992. Task analyzability, use of new media, and effectiveness: A multi-site exploration of media richness. Organizational Science 3 (4): 475-500.
Rizzo, J. R., R. J. House, and S. I. Lirtzman. 1970. Role conflict and ambiguity in complex organizations. Administrative Science Quarterly 15: 150-163.
Rusbult, C. E., and D. Farrell. 1983. A longitudinal test of the investment model: The impact of job satisfaction, job commitment, and turnover of variations in rewards, costs, alternatives, and investments. Journal of Applied Psychology 68 (August): 429-438.
Scott, T. W., and P. Tiessen. 1999. Performance measurement and managerial teams. Accounting, Organizations and Society 24: 263-2 85.
Seltzer, J., and B. Bass. 1990. Transformational leadership: Beyond initiation and consideration. Journal of Management 16 (4): 693-703
Senatra, P. T. 1980. Role conflict, role ambiguity, and organizational climate in a public accounting firm. The Accounting Review 54: 594-603.
Sosik, J. J. 1997. Effects of transformational leadership and anonymity on idea generation in computer-mediated groups. Group and Organization Management 22 (4): 460-487.
-----, B. J. Avolio, and S. S. Kahai. 1997. Effects of leadership style and anonymity on group potency and effectiveness in a group decision support system environment. Journal of Applied Psychology 82 (1): 89-103.
Summers, S. L., J. T. Sweeney, and C. M. Wolk. 2000. Problem-solving style and fit in consulting and auditing. Journal of Information Systems 14(1): 1-15.
Tepper, B. J. 1995. Upward maintenance tactics in supervisory mentoring and nonmentoring relationships. AcademyofManagementJournal38 (4): 1191-1205.
Viator, R. E. 2001. The association of formal and informal public accounting mentoring with role stress and related job outcomes. Accounting, Organizations and Society 26 (1): 73-93.
Wheaton, B., B. Muthen, D. Alwin, and G. Summers. 1977. Assessing reliability and stability in panel models. In Sociological Methodology, edited by D. Heise. San Francisco, CA: Jossey Boss.
Withey, M., R. L. Daft, and W. H. Cooper. 1983. Measures of Perrow's work unit technology: An empirical assessment and a new scale. Academy of Management Journal 26 (1): 45-63.
Yammarino, F. J., and B. Bass. 1990. Transformational leadership and multiple levels of analysis. Human Relations 43 (10): 975-995.
Yukl, G. A. 1989. Managerial leadership: A review of theory and research. Journal of Management 15 (2): 251-289.
Table 1
Participant Demographics by Organizational Level
Participant Organizational Level
Consultant/
Senior
Accountant Manager
Gender
Male 59 117
Female 42 41
Total participants 101 158
Primary Functional Area
IS/BC 21 43
Financial auditing 44 60
Tax 27 42
Other functional area 9 13
Firm Size
Big 5 39 56
Other national 14 20
Regional 41 73
Local 7 9
Office Size
(Number of Professionals)
> 200 28 50
101 to 200 19 21
51 to 100 21 38
26 to 50 23 28
< 25 10 21
Median Age 30 34
Median Years: Public
Accounting Experience 5 8
Referent Supervisor is Mentor 69.3% 60.8%
Participants with NO Mentor 15.8% 20.2%
Participant
Organizational Level
Senior Directors/ Total
Manager Partners Participants
Gender
Male 99 28 303
Female 26 4 113
Total participants 125 32 416
Primary Functional Area
IS/BC 31 5 100
Financial auditing 52 8 164
Tax 33 10 112
Other functional area 9 9 40
Firm Size
Big 5 62 10 167
Other national 2 1 37
Regional 50 16 180
Local 11 5 32
Office Size
(Number of Professionals)
> 200 53 12 143
101 to 200 16 1 57
51 to 100 28 10 97
26 to 50 18 6 75
> 25 10 3 44
Median Age 37 46 35
Median Years: Public
Accounting Experience 12 15 10
Referent Supervisor is Mentor 52.8% 59.4% 60.3%
Participants with NO Mentor 22.4% 25.0% 20.2%
Table 2
Regression Analysis of Task Analyzability and Transformational
Leadership
(n = 416)
Transformational
Task Analyzability (a) Leadership (a)
Predictor Variables [beta] [beta]
Primary Functional Area
ISA/BC (b) -2.180 (****) 2.058 (**)
Tax -1.119 (****) -0.309
Other functional areas -1.005 (**) -0.397
Organizational Level
Manager 0.098 -2.934 (**)
Senior manager -0.037 -2.002
Partner -1.447 (**) 1.965
Gender
Female 0.261 0.476
Firm Size
Other national -0.737 -0.651
Regional -0.754 (*) -0.655
Local -0.740 4.710 (**)
Office Size (Number of
Professionals)
101 to 200 -0.414 -3.874 (**)
51 to 100 -0.605 -3.025 (*)
26 to 50 -1.179 (**) -4.284 (**)
< 25 -0.178 -3.825 (*)
Other Demographics
Age -0.040 (*) -0.149 (*)
Years in public accounting 0.074 (**) 0.041
Referent supervisor mentor 0.851 (****) 12.518 (****)
Adjusted [R.sup.2] 0.174 0.340
Model: F-statistic 5.89 (****) 12.96 (****)
(****), (***), (**), (*)p < .001, p < .01, p < .05, and p < .10,
respectively.
(a)Positive beta coefficients indicate higher levels of task
analyzability and transformational leadership.
(b)Test of beta coefficients reflects directional one-tailed test.
TABLE 3
Regression Analysis of Role Clarity and Related Role Outcomes
(n = 416)
High Low
Role Job
Clarity (a) Satisfaction (a)
Predictor Variables [beta] [beta]
Organizational Level
Manager -0.089 -0.004
Senior Manager -0.057 0.432
Partner -0.596 0.471
Gender
Female 0.073 0.626
Firm Size
Other National 0.034 -0.877
Regional -0.047 -0.186
Local 0.780 2.073 (**)
Office Size
(Number of Professionals)
101 to 200 -0.995 0.213
51 to 100 -0.866 -0.081
26 to 50 -1.484 (*) -0.347
< 25 -1.488 (*) -0.146
Other Demographics
Age -0.023 0.022
Years in Public
Accounting 0.045 0.026
Referent Supervisor
Mentor 1.825 (****) 3.004 (****)
Other Mentor 0.729 0.944 (**)
Adjusted [R.sup.2] 0.086 0.151
Model: F-statistic 2.39 (***) 4.51 (****)
High
Affective Sacrifice
Commitment (a) Commitment (a)
Predictor Variables [beta] [beta]
Organizational Level
Manager 0.937 0.097
Senior Manager 2.534 (***) 0.819 (**)
Partner 4.205 (***) 1.376 (**)
Gender
Female 1.123 (*) 0.430
Firm Size
Other National -0.599 1.133 (**)
Regional 0.100 0.504
Local 2.914 (**) 0.844
Office Size
(Number of Professionals)
101 to 200 0.702 0.427
51 to 100 0.557 -0.068
26 to 50 0.530 -0.312
< 25 0.589 0.281
Other Demographics
Age -0.005 0.007
Years in Public
Accounting -0.034 0.005
Referent Supervisor
Mentor 3.961 (****) -0.500 (*)
Other Mentor 1.957 (***) -0.048
Adjusted [R.sup.2] 0.181 0.070
Model: F-statistic 5.59 (****) 1.91 (**)
Low
Alternative Job
Commitment (a) Performance (a)
Predictor Variables [beta] [beta]
Organizational Level
Manager -0.138 0.435
Senior Manager 0.207 2.703 (***)
Partner 0.110 3.284 (**)
Gender
Female 0.038 1.491 (**)
Firm Size
Other National 0.839 -3.290 (***)
Regional 0.181 -0.718
Local -0.069 0.300
Office Size
(Number of Professionals)
101 to 200 0.046 -1.695 (*)
51 to 100 -0.011 -0.627
26 to 50 -0.203 -1.767
< 25 0.520 -1.116
Other Demographics
Age 0.034 0.005
Years in Public
Accounting 0.057 (*) -0.004
Referent Supervisor
Mentor -0.773 (***) 1.019 (*)
Other Mentor -0.245 0.458
Adjusted [R.sup.2] 0.128 0.125
Model: F-statistic 3.73 (****) 3.61 (****)
(****), (***), (**), (*) p < .001, p < .01, p < .05, and p < .10,
respectively.
(a)Positive beta coefficients indicate higher levels of the dependent
variables.
APPENDIX A
Questionnaire Items' Means, Standard Deviations, and Measure Model
Parameter Estimates
Measure
Model
Parameter
Estimates:
Leader
Behaviors
Means
Questionnaire Items (Std. Dev.) Vision
Leadership: My Referent Supervisor...
(Articulates Vision)
...paints an interesting picture of 3.598 1.000
the future for our practice (0.856)
area.
...has a clear understanding of 3.744 .517
where we are going. (0.814)
...inspires others with his/her 3.374 1.247
plans for the future. (0.929)
...is able to get others committed 3.267 1.201
to his/her dream of the future. (0.875)
...is always seeking new 3.895 .906
opportunities for our practice (0.878)
area and for the firm.
(Provides Model)
...leads by "doing" rather than 3.598
simply by "telling." (0.999)
...provides a good model to follow. 3.733
(0.909)
Leadership: My Referent Supervisor...
(Fosters Acceptance of
Group Goals)
...fosters collaboration 3.601
among work groups. (0.861)
...develops a team attitude 3.567
and spirit among his/her (0.921)
subordinates.
...encourages employees to be 3.757
"team players." (0.848)
...gets each engagement team 3.565
(or, project team) to work (0.716)
together for the same goal.
(High Performance)
...shows us that he/she 4.231
expects a lot from us. (0.697)
...insists on only the best 3.943
performance. (0.722)
...will not settle for second 3.802
best. (0.773)
(Provides Individualized Support)
...acts without considering my 3.377
feelings. (R) (0.966)
...shows respect for my personal 3.639
feelings. (0.891)
...behaves in a manner that is 3.569
thoughtful of my personal needs. (0.896)
...acts in a manner that doesn't 3.600
consider my personal feelings. (0.964)
(R)
Leadership: My Reference Supervisor...
(Intellectual Stimulation)
...has provided me with new ways of 3.418
looking at probles which used to (0.925)
puzzle me.
...has stimulated me to think about 3.477
old problems in new ways. (0.849)
...has ideas that have forced me to 3.547
rethink some of my own ideas. (0.785)
(Transactional Leadership)
...always gives me positive 3.421
feedback when I perform well. (1.003)
...gives me special recognition 3.491
when my work is very good. (1.006)
...personally compliments me when I 3.554
do outstanding work. (0.987)
...frequently does not acknowledge 3.435
my good performance. (R) (1.001)
Measure Model Parameter
Estimates: Leader Behaviors
Group
Questionnaire Items Model Goals
Leadership: My Referent Supervisor...
(Articulates Vision)
...paints an interesting picture of
the future for our practice
area.
...has a clear understanding of .363
where we are going.
...inspires others with his/her
plans for the future.
...is able to get others committed
to his/her dream of the future.
...is always seeking new
opportunities for our practice
area and for the firm.
(Provides Model)
...leads by "doing" rather than 1.000
simply by "telling."
...provides a good model to follow. 1.108
Leadership: My Referent Supervisor...
(Fosters Acceptance of
Group Goals)
...fosters collaboration 1.000
among work groups.
...develops a team attitude 1.232
and spirit among his/her
subordinates.
...encourages employees to be 1.110
"team players."
...gets each engagement team .828
(or, project team) to work
together for the same goal.
(High Performance)
...shows us that he/she
expects a lot from us.
...insists on only the best
performance.
...will not settle for second
best.
(Provides Individualized Support)
...acts without considering my
feelings. (R)
...shows respect for my personal
feelings.
...behaves in a manner that is
thoughtful of my personal needs.
...acts in a manner that doesn't
consider my personal feelings.
(R)
Leadership: My Reference Supervisor...
(Intellectual Stimulation)
...has provided me with new ways of
looking at probles which used to
puzzle me.
...has stimulated me to think about
old problems in new ways.
...has ideas that have forced me to
rethink some of my own ideas.
(Transactional Leadership)
...always gives me positive
feedback when I perform well.
...gives me special recognition
when my work is very good.
...personally compliments me when I
do outstanding work.
...frequently does not acknowledge
my good performance. (R)
Measure Model Parameter
Estimates: Leader Behaviors
High Individual
Questionnaire Items Performance Support
Leadership: My Referent Supervisor...
(Articulates Vision)
...paints an interesting picture of
the future for our practice
area.
...has a clear understanding of
where we are going.
...inspires others with his/her
plans for the future.
...is able to get others committed
to his/her dream of the future.
...is always seeking new
opportunities for our practice
area and for the firm.
(Provides Model)
...leads by "doing" rather than
simply by "telling."
...provides a good model to follow.
Leadership: My Referent Supervisor...
(Fosters Acceptance of
Group Goals)
...fosters collaboration
among work groups.
...develops a team attitude
and spirit among his/her
subordinates.
...encourages employees to be
"team players."
...gets each engagement team
(or, project team) to work
together for the same goal.
(High Performance)
...shows us that he/she 1.000
expects a lot from us.
...insists on only the best 1.160
performance.
...will not settle for second 1.518
best.
(Provides Individualized Support)
...acts without considering my 1.000
feelings. (R)
...shows respect for my personal 1.001
feelings.
...behaves in a manner that is 1.017
thoughtful of my personal needs.
...acts in a manner that doesn't 1.150
consider my personal feelings.
(R)
Leadership: My Reference Supervisor...
(Intellectual Stimulation)
...has provided me with new ways of
looking at probles which used to
puzzle me.
...has stimulated me to think about
old problems in new ways.
...has ideas that have forced me to
rethink some of my own ideas.
(Transactional Leadership)
...always gives me positive
feedback when I perform well.
...gives me special recognition
when my work is very good.
...personally compliments me when I
do outstanding work.
...frequently does not acknowledge
my good performance. (R)
Measure Model Parameter
Estimates: Leader Behaviors
Intellectual
Questionnaire Items Stimulation Transactional
Leadership: My Referent Supervisor...
(Articulates Vision)
...paints an interesting picture of
the future for our practice
area.
...has a clear understanding of
where we are going.
...inspires others with his/her
plans for the future.
...is able to get others committed
to his/her dream of the future.
...is always seeking new
opportunities for our practice
area and for the firm.
(Provides Model)
...leads by "doing" rather than
simply by "telling."
...provides a good model to follow.
Leadership: My Referent Supervisor...
(Fosters Acceptance of
Group Goals)
...fosters collaboration
among work groups.
...develops a team attitude
and spirit among his/her
subordinates.
...encourages employees to be
"team players."
...gets each engagement team
(or, project team) to work
together for the same goal.
(High Performance)
...shows us that he/she
expects a lot from us.
...insists on only the best
performance.
...will not settle for second
best.
(Provides Individualized Support)
...acts without considering my
feelings. (R)
...shows respect for my personal
feelings.
...behaves in a manner that is
thoughtful of my personal needs.
...acts in a manner that doesn't
consider my personal feelings.
(R)
Leadership: My Reference Supervisor...
(Intellectual Stimulation)
...has provided me with new ways of 1.000
looking at probles which used to
puzzle me.
...has stimulated me to think about 1.042
old problems in new ways.
...has ideas that have forced me to .726
rethink some of my own ideas.
(Transactional Leadership)
...always gives me positive 1.000
feedback when I perform well.
...gives me special recognition .932
when my work is very good.
...personally compliments me when I .927
do outstanding work.
...frequently does not acknowledge
my good performance. (R) .920
APPENDIX B
Questionnaire Items' Means, Standard Deviations, and Measure Model
Parameter Estimates
Measure Model
Parameter Estimates
Means Task
Questionnaire Items (Std. Dev.) Analyzability
(Task Analyzability)
The firm has well-developed 3.741 1.000
methods for analyzing client (0.873)
needs and providing service to
clients.
There is a body of professional 3.990 1.519
literature that provides guidance (0.901)
in doing my work.
There is an understandable 3.797 1.486
sequence of steps that can be (0.800)
followed in carrying out my
assignmens and responsibilities.
In doing my work, I rely on 3.724 1.563
procedures and practices (0.870)
established by the firm.
(Role Clarity)
I feel certain about how much 3.685
authority I have at my firm. (0.895)
Clear, planned goals and 3.400
objectives exist for my job. (0.985)
I know that I have allocated my 3.487
time properly at my firm. (0.854)
I know what my responsibilities 3.548
are at my firm. (0.921)
I have been provided clear 3.449
explanations of what has to be (0.899)
done at my firm.
Measure
Model
Parameter
Estimates
Role
Questionnaire Items Clarity
(Task Analyzability)
The firm has well-developed
methods for analyzing client
needs and providing service to
service to clients.
There is a body of professional
literature that provides guidance
in doing my work.
There is an understandable
sequence of steps that can be
followed in carrying out my
assignmens and responsibilities.
In doing my work, I rely on
procedures and practices
established by the firm.
(Role Clarity)
I feel certain about how much 1.000
authority I have at my firm.
Clear, planned goals and 1.295
objectives exist for my job.
I know that I have allocated my .862
time properly at my firm.
I know what my responsibilities 1.312
are at my firm.
I have been provided clear 1.185
explanations of what has to be
done at my firm.
Measure Model
Parameter
Estimates
Means Affective
Questionnaire Items (Std. Dev.) Commitment
(Organizational Commitment)
I do not feel a strong sense of 3.610 1.000
belonging to my firm. (R) (1.098)
I do not feel "emotionally 3.448 1.085
attached" to this firm. (R) (1.124)
This firm has a great deal of 3.424 .985
personal meaning to me. (0.978)
I would be very happy to spend 3.426 1.022
the rest of my career with this (1.025)
firm.
I enjoy discussing my firm with 3.786 .748
people outside it. (0.818)
I really feel as if this firm's 3.294 .771
problems are my own. (0.997)
I do not feel like "part of the 3.529 1.036
family" at this firm. (R) (1.034)
I think I could easily become as 3.068 .888
attached to another firm as I am (1.056)
to this one. (R)
One of the major reasons I 2.815
continue to work for this firm (1.162)
is that leaving would require
considerable personal sacrifice.
For example, another firm may
not match the benefits I have.
It would be hard to leave my firm 2.827
now, even if I wanted to. (1.074)
Too much in my life would be 2.888
disrupted if I decided I wanted (1.080)
to leave my firm now.
It wouldn't be too costly for me 2.900
to leave my firm in the near (1.013)
future. (R)
Right now, staying with my firm 2.605
is a matter of necessity as much (1.070)
as desire.
I feel I have too few options to 2.121
consider leaving this firm. (0.995)
One of the few negative 2.102
consequences of leaving this (1.061)
firm the scarcity of available
alternatives.
I am not afraid of what might 3.153
happen if I quit my job without (1.235)
having another one lined up. (R)
Measure Model Parameter
Estimates
High Low
Questionnaire Items Sacrifice Alternatives
(Organizational Commitment)
I do not feel a strong sense of
belonging to my firm. (R)
I do not feel "emotionally
attached" to this firm. (R)
This firm has a great deal of
personal meaning to me.
I would be very happy to spend
the rest of my career with this
firm.
I enjoy discussing my firm with
people outside it.
I really feel as if this firm's
problems are my own.
I do not feel like "part of the
family" at this firm. (R)
I think I could easily become as
attached to another firm as I am
to this one. (R)
One of the major reasons I 1.000
continue to work for this firm
is that leaving would require
considerable personal sacrifice.
For example, another firm may
not match the benefits I have.
It would be hard to leave my firm 1.692
now, even if I wanted to.
Too much in my life would be 1.471
disrupted if I decided I wanted
to leave my firm now.
It wouldn't be too costly for me --
to leave my firm in the near
future. (R)
Right now, staying with my firm 1.000
is a matter of necessity as much
as desire.
I feel I have too few options to .764
consider leaving this firm.
One of the few negative .726
consequences of leaving this
firm the scarcity of available
alternatives.
I am not afraid of what might --
happen if I quit my job without
having another one lined up. (R)
Measure Model
Parameter
Estimates
Means Job
Questionnaire Items (Std.Dev.) Satisfaction
All things considered, I 3.543 1.000
am extremely satisfied with my (0.924)
current assignments and
responsibilities.
Knowing what I now know, if 3.606 .876
I had to decide all over again (0.994)
whether to pursue this type of
work, I would certainly do it.
If a good friend of mine told 3.690 .876
me that he/she was interested (0.905)
in a job like mine (with my
employer), I would recommend it.
My current work compares very 3.125 1.141
well to my ideal job. (0.981)
My current assignments and 3.384 .824
responsibilities measure up very (0.920)
well to the sort of job I
wanted when I chose this career.
In general, I like my work 3.693 .896
very much. (0.835)
(Job Performance)
Maintaining expected quantity 3.801
of work. (0.809)
Maintaining expected quality 4.101
of work. (0.669)
Accepting respo