Printer Friendly

Commitment and insurance agents' job perceptions, attitudes, and performance.


"Work commitment" describes a number of concepts relating to employee attachment to work related entities. Morrow (1983) has identified five such entities or facets of work commitment: commitment to one's organization, job, career or profession, union, and work values. While these forms of work commitment have received the most attention, other commitments such as those to a community have also been proposed and investigated (Steffy and Jones, 1988). It has been suggested that, since salespeople (e.g., insurance agents) clearly have multiple loyalties, multiple forms of work commitment should be considered within single investigations (e.g., Chonko, 1986; Johnston et al., 1990).

Considerable attention has been paid over the years to the qualities of the measures used to tap commitment concepts (e.g., Koslowsky, Caspy, and Lazar, 1990; Mathieu and Farr, 1991), to identifying the antecedents and consequences of these concepts (e.g., Colarelli and Bishop, 1990; Mathieu and Zajac, 1990), and recently to how these forms of work commitment might interact to influence attitudes and performance behaviors. It is the last of these topics that constitutes the focus of this article.

Blau and Boal (1987) were among the first to hypothesize how various forms of work commitment might interact to affect work related attitudes and behaviors. Their model crosses high and low levels of organizational commitment with high and low levels of job involvement, producing four types of employees. Their model predicts that job attitudes such as job satisfaction and behaviors such as turnover and absenteeism can be better understood by examining the interaction between organizational commitment and job involvement than by examining either form of work commitment separately. Positioning commitment variables as antecedents is also more consistent with the view that commitment is a processual phenomenon, affecting work related perceptions and outcomes, and not just a consequence of perceptions and attitudes (Johnston et al., 1990; Mowday, Porter, and Steers, 1982; Pinder, 1984; Wiener and Vardi, 1980). Vandenberg and Lance (1992), for example, used a structural equation model to demonstrate that organizational commitment caused satisfaction rather than satisfaction causing commitment. It is also consistent with the social information processing approach (Salancik and Pfeffer, 1978; Staw, 1986), which contends that perceptions and attitudes are highly variable and relatively unstable judgments that are easily influenced by such factors as preexisting attitudes (e.g., levels of commitment). Accordingly, this study examines the effects of commitment, separately and interactively, on the attitudes and behaviors of insurance agents.

Agent work commitment is an important issue to the insurance industry because of the difficulty and costs associated with attracting and retaining productive agents and because of the direct impact agents have on the insurer's financial success. Job turnover among agents at independent agencies has been shown to be relatively low, but the loss of an experienced agent can be catastrophic (CPCU Journal, 1991). In the life insurance industry, on the other hand, agent retention from 1988-1990 averaged 19 percent (King, 1992). Life insurance companies are currently emphasizing the allocation of resources for agent training and seasoning, focusing on retention rather than on new producer recruitment (King, 1992). Research on agent work commitment should prove useful to such retention efforts.

Because of the nature of the insurance industry, a direct replication of Blau and Boal's model using insurance agents is not appropriate, particularly because not all forms of work commitment are equally germane to the insurance industry. Union commitment, for example, is not applicable to the insurance agent domain. Organizational commitment also has less applicability to insurance agents than to other employee populations, because many insurance agents--particularly independent agents--sell numerous types of policies from several companies; thus, it is less logical to think of an agent's psychological attachment and desire to remain with a particular company. Commitment to one's profession and to one's job are much more relevant targets of study among insurance agents, and such research has explicitly been called for by Ingram, Lee, and Skinner (1989). The work values dimension of commitment was not selected for inclusion in this study because it is deemed to be relatively permanent over the course of one's life (Morrow, 1983) and thus is subject to little influence. Although information regarding how work-related values affect agent attitudes and behaviors might prove useful from a selection perspective, information on the impact of other forms of commitment has utility from both a selection and a development perspective.

A study of insurance agent commitment offers the opportunity to explore an additional commitment: that to one's community. This form of commitment, the psychological attachment of individuals to the community in which they live, may be highly applicable to insurance agents for whom residents of their own community comprise their main constituency. Given that insurance work entails personal sales and service over time, involvement in one's community may enhance contacts and generate repeat business. Finally, the concept of multiple commitments is especially relevant to insurance sales in that agents spend a great deal of time with people outside their organization (Ingram, Lee, and Skinner, 1989; Johnston et al., 1990).

To summarize, this study examines three forms of commitment--job involvement, professional commitment, and community commitment--and their relationship to insurance agent perceptions, attitudes, and performance. The perceptions, attitudes, and performance behaviors selected for inclusion in this study have been used repeatedly in predicting work related outcomes in a variety of employee populations within the management literature (see recent reviews, e.g., Morrow, Mullen, and McElroy, 1990). Perceptions studied include job design (i.e., task autonomy, identity, variety, significance, and feedback), perceived job overqualification, and perceived opportunities for job mobility within the same community. Attitudes surveyed include satisfaction with work, pay, and co-workers, and probability of remaining in the insurance profession. These outcomes are also of interest because they offer the possibility of change via organizational intervention (see Johnston et al., 1990). Finally, the performance behaviors we examine are the percent of income derived from insurance renewals, annual earned income, and the percent of total income from insurance business sources. These performance measures were selected to determine whether these forms of commitment, either separately or interactively, have any "bottom line" implications for agents.

Literature Review

Previous research has recognized the importance of job perceptions and attitudes among insurance salespeople. For example, researchers have examined agent motivation and performance (e.g., Oliver, 1974), agent satisfaction and performance (Hafer and McCuen, 1985), the impact of leadership behaviors on agent attitudes (Dubinsky et al., 1988), agent citizenship behavior and performance evaluations (MacKenzie, Podsakoff, and Fetter, 1991), and job characteristics and organizational outcomes (Tyagi, 1985). Little research exists, however, on the relationship of the various forms of work commitment to agent perceptions, attitudes, and performance.

Job Involvement

A popular topic in organizational behavior research is job involvement, the extent to which an individual psychologically identifies with his or her job. Job perceptions, including job design variables, or how people describe various aspects of their work, overqualification, and opportunities for job mobility have been subjects of frequent interest. With respect to job design, job scope has been positively linked to job involvement (Mathieu and Farr, 1991; Rabinowitz and Hall, 1977), while research on other job design variables has been equivocal (Blau, 1987; Parasuraman and Alutto, 1984; Sekaran, 1989). This suggests that some job design variables are related to job involvement while others are not.

Another managerially relevant perception is overqualification; that is, the extent to which individuals feel their abilities are not fully utilized (Khan and Morrow, 1991). Insurance agents are susceptible to this phenomenon given that opportunities for advancement and horizontal job expansion are necessarily limited by the direct sales nature of the insurance industry. Agents who do not perceive themselves as overqualified are hypothesized to exhibit higher levels of job involvement. Job involvement and perceived job mobility have consistently demonstrated an inverse relationship (Farrell and Rusbult, 1981; Rusbult and Farrell, 1983).

The relationships between job involvement and job attitudes are mixed. While the causal direction is not clear, numerous studies have shown that job involvement and job satisfaction are positively related (Blau and Boal, 1987; Mathieu and Farr, 1991; Mathieu and Kohler, 1990). The relationship between job involvement and intention to remain in the same career or profession has been shown to be not statistically significant in nonselling samples (Blau, 1985, 1989). However, an examination of job involvement and intent to remain revealed a positive relationship among industrial salespeople (Ingram and Lee, 1990).

The relationship between job involvement and indicators of job performance is also somewhat murky. Of five studies investigating this relationship, four demonstrated a positive relationship (Brooke, Russell, and Price, 1988; Mathieu and Farr, 1991; Parasuraman and Alutto, 1984; Cron and Slocum, 1986), and one indicated no relationship (Hollenbeck and Williams, 1986). Although tangential to performance, it is perhaps relevant that both Weiner and Vardi (1990) and Ingram et al. (1989) found a positive relationship between job involvement and selling effort.

Professional Commitment

Professional commitment emphasizes the importance of a profession in one's life. Because of the steady increase in the number of formally recognized professional occupations (Cherniss and Kane, 1987), interest in managing professional commitment continues to increase (Morrow and Wirth, 1989). Maintaining professional commitment is of special interest in professions like insurance sales, where only a few field representatives or agents work together and others in the community in the same line of work serve as competitors. Insurance agent professionalism is also being touted as a solution to consumers' negative perception of agents and the insurance industry and to the increasing amount of lawsuits, investigations, and adversarial rhetoric that the insurance industry receives (Knowles, 1990).

We found no studies that explicitly evaluate the relationship between professional commitment and job perceptions. However, as in the case of job involvement, agents who report lower feelings of overqualification are hypothesized to demonstrate higher levels of professional commitment. Stronger relationships have been observed in conjunction with job attitudes. Professional commitment has consistently demonstrated a positive relationship to job satisfaction (Blau, 1989; McCloskey and McCain, 1987; Norris and Niebuhr, 1983). Professional commitment and intent to remain in the same career or profession have been shown repeatedly to be inversely related (Blau, 1985, 1988, 1989). Lastly, job performance was not found to have been explicitly examined with respect to professional commitment.

Community Commitment

Little research exists on the antecedents and consequences of community commitment. One might hypothesize that community commitment is salient to insurance agents because members of the community constitute their main constituency. Stated differently, service to community members is the focal point of an insurance agent's work. Achieving and maintaining a high profile in the community through participation in civic affairs, for example, may enhance contacts, customer confidence, and referrals. Attachment to actual or potential customers may even substitute for feelings of satisfaction others receive via conventional co-worker contact. Finally, since insurance agents are more individually responsible for their success than workers in other occupations, career planning is an important behavior for agents. Steffy and Jones (1988) found that individuals with higher levels of community commitment engaged in more career planning. Consequently, community commitment might affect agents' willingness to serve members of the community and, subsequently, their job satisfaction and performance behaviors. The lack of any prior research on this newly articulated concept makes specific prediction difficult. But the nature of the insurance agent's work provides a unique setting in which to test the efficacy of this commitment construct.

Finally, if interactive models of commitment are correct, one would predict the presence of multiple forms of commitment to intensify the impact of commitment on attitudes, behaviors, and performance. Professional commitment, job involvement, and community commitment are all related to an insurance agent's work. Indeed, these three concepts are deemed to be three different facets of work commitment. Consequently, an interactive model, such as that proposed by Blau and Boal (1987), posits that the more psychologically attached one is to one's work (i.e., the more commitment one feels), the more positive will be one's perceptions, attitudes, and, ultimately, performance.

Research Propositions

Based on the above literature review, four general propositions serve to guide this research.

1. Insurance agents with higher levels of job involvement will report more favorable job perceptions and attitudes and more positive performance behaviors than insurance agents with lower levels of job involvement.

2. Insurance agents with higher levels of professional commitment will report more favorable job perceptions and attitudes and more positive performance behaviors than insurance agents with lower levels of professional commitment.

3. Insurance agents with higher levels of community commitment will report more favorable job perceptions and attitudes and more positive performance behaviors than insurance agents with lower levels of community commitment.

4. In line with Blau and Boal's (1987) interactive model of commitment, it is expected that these three forms of commitment will interact to affect job perceptions and attitudes and performance behaviors. Specifically, an interactive model would predict that higher levels and a variety of commitments will result in more positive job perceptions, attitudes, and behaviors.



Data were collected from 221 insurance agents attending continuing education seminars in a U.S. rural midwestern state. Seminars were held at seven sites within the state, and agents met in groups ranging from 25 to 40 participants. Questionnaires were distributed at each seminar site, and respondents were given the option to complete and return the survey during the seminar or return it by mail later. Of the 221 questionnaires distributed, usable data were obtained from 166, yielding a response rate of 75 percent. The sample was primarily male (79 percent) with a mean age of 45 years. The average agent in the sample had attended some college and currently worked in a small community of between 5,000 and 10,000 residents. Fifty-two percent of the responding agents classified themselves as independent agents representing multiple insurance companies, with the remainder being either primarily life or captive/exclusive agents who represent a single insurance company. Since these demographic variables were not of substantive interest, they served as covariates in the analysis.

Independent Variables

Three forms of work commitment constituted the independent variables of interest in this study. The first, job involvement, was operationalized using Kanungo's (1982) ten-item job involvement scale. Using a five-point Likert scale (1 = strongly disagree, 5 = strongly agree), respondents indicated the extent to which they disagree or agree with statements such as, "I am very much personally involved in my job," "Most of my personal life goals are job-oriented," and "I consider my job to be very central to my existence." Scores on the individual items were totaled and then divided by the number of items in order to convert responses back to the scale's original metric. This scale demonstrated good internal reliability, as evidenced by a coefficient alpha of 0.84.

The second independent variable in this study was professional commitment, which was measured using an instrument developed by Aranya, Kushnir, and Valency, (1986). This instrument adapts the fifteen-item seven-point Likert scale employed by Mowday, Steers, and Porter (1979) in their Organizational Commitment Scale, with the word "profession" substituted for the word "organization." Although there is some debate about whether items reflecting an intent to remain in the organization or profession should be retained in this measurement approach, the value of being able to make comparisons with studies using the full version of the measure is felt to outweigh recent criticism (Mathieu and Zajac, 1990; Morrow, 1993). Sample items include "I talk up my profession to my friends as a great field to work in," "I really care about the fate of my profession," and "I find that my values and the values of my profession are very similar." As with job involvement, the arithmetic mean of the scale responses serves as a single indicator of one's level of professional commitment. Coefficient alpha for this scale was 0.90.

The third independent variable in this study was community commitment, which was also measured using an instrument based on Mowday, Steers, and Porter's (1979) Organizational Commitment Scale. We substituted the word "community" for the word "organization" in order to gauge one's sense of community commitment. Coefficient alpha was 0.90 for this scale. Descriptive information on each of these measures is presented in Tables 1 and 2.

Evidence supporting the independence of the three forms of work commitment is provided by the relatively low intercorrelations among the measures (i.e., rs = 0.40, 0.27, and 0.28). Further support for the independence of the construct was supplied by a factor analytic assessment, available from the authors on request. A principal components analysis using a varimax rotation yielded a seven-factor solution accounting for 62.8 percent of the total variance. The first three factors clearly corresponded to the commitment forms specified; the remaining emergent factors indicate that some commitment measures may be multidimensional.


Dependent Variables: Job Perceptions

Job design variables. The Job Diagnostic Survey (JDS) developed by Hackman and Oldham (1975) was used to measure respondents' job perceptions. This instrument measures six job characteristics: autonomy, identity, variety, task significance, feedback, and dealing with others. Dealing with others was excluded from this analysis because of reliability problems, presumably emanating from confusion among insurance agents over whether "others" refers to co-workers or customers.

The remaining five job design perceptions are important in that they collectively measure the extent to which a job is thought to embody positive attributes by job incumbents. These perceptions, in turn, are linked to employee attitudes and behavior. Although the JDS has its critics, Griffin TABULAR DATA OMITTED (1982) notes that it has been the most widely used measure in task design research. Each of the five job design perceptions used in this study was measured using a three-item scale and a seven-point response format. Coefficient alphas ranged from 0.60 for the job variety scale to 0.83 for the identity scale.

Overqualification. The degree to which respondents feel overqualified for their current position was measured using a five-item scale developed by Khan and Morrow (1991). On a five-point Likert scale, respondents agree or disagree with statements such as "My job frequently provides me with new challenges" (reverse coded), "My job provides me with many opportunities to learn new things" (reverse coded), and "My talents are not fully utilized on my job." Coefficient alpha on this scale for this sample was 0.75.

Perceptions of mobility within community. A single item was used to measure respondents' perceptions of how mobile they are within their own community. The participants were asked, "If you decided to leave your present job, how would you describe your chances of getting another job in your community?" Response options ranged from 1, very high (I am 95-100 percent sure I would find another job), to 6, very low (I am 95-100 percent sure I would not be able to find a job).

Dependent Variables: Job Attitudes

Satisfaction. Smith, Kendall, and Hulin's (1969) Job Descriptive Index (JDI) was used to operationalize three facets of satisfaction: satisfaction with work, satisfaction with pay, and satisfaction with co-workers. Two other facets measured by the JDI, satisfaction with supervisor and satisfaction with opportunities for promotion, were not used due to their inappropriateness for insurance agents, who are, in many instances, self-employed.

The JDI employs a list of adjectives and a "yes," "no," or "?" response format. The authors of the index have argued that a ? is closer to a no than a yes and consequently score the responses 3 for yes, 1 for ?, and 0 for no. In cases where responses are reverse coded, a ? response is given a score of 1. Hanisch (1992) has supported the appropriateness of this scoring scheme. In this study, arithmetic means are used as single indicators of satisfaction with work, pay, and co-workers.

Satisfaction with work is measured using an eighteen-item scale. Sample adjectives to describe satisfaction with one's work include "fascinating," "boring" (reverse coded), "pleasant," "respected," and "challenging." Coefficient alpha for this scale, 0.62, was below the normally reported 0.70, indicating that results involving this scale should be viewed with caution. Satisfaction with pay is measured using an eight-item scale. Sample adjectives include "income adequate," "bad" (reverse coded), and "highly paid." Coefficient alpha for this scale was 0.84. Satisfaction with co-workers is measured using an eighteen-item scale. Sample adjectives include "stimulating," "stupid" (reverse coded), "intelligent," and "responsible." Coefficient alpha for this scale was 0.86.

Intention to remain in profession. A single item explored the relationship between forms of commitment and one's attitude toward remaining in the insurance profession. The question asked, "Assuming things in your personal life and work remain about the same, what is the probability you will remain in your profession in the near future?" Response options ranged from 1, very low (I am 95-100 percent sure I will leave or try to leave), to 6, very high (I am 95-100 percent sure I will stay).

Dependent Variables: Income Performance Variables

The final set of dependent variables asked respondents to report on their performance, as measured by their self-reported income. Three single-item measures were used. The first asked, "What percent of your income is based on renewal business?" Response choices were 1, under 10 percent; 2, 10-15 percent; 3, 16-20 percent; 4, 21-25 percent; 5, 26-50 percent; and 6, over 50 percent. The second question asked respondents to report their total annual income. Response choices were 1, $0-$10,000; 2, $10,001-$25,000; 3, $25,001-$50,000; 4, $50,001-$75,000; 5, $75,001-$100,000; and 6, over $100,000. Finally, respondents were asked to indicate the percent of their total income that comes from their insurance activities. Response options were 1, 0-10 percent; 2, 11-25 percent; 3, 26-50 percent; 4, 51-75 percent; and 5, 76-100 percent.

Research Design

Interactive commitment models, such as that proposed by Blau and Boal (1987), are predicated on a dichotomization of forms of commitment. In essence, individuals said to be high on two different forms of commitment, high on one form but low on another, or low on both, are predicted to differ in their job attitudes and behaviors from individuals holding other combinations of commitment. Consistent with these models, the three measures of commitment--job involvement, professional commitment, and community commitment--were dichotomized into high and low levels by dividing the scores on each measure at the median. This resulted in a 2 X 2 X 2 factorial research design with two levels of three forms of commitment for eight possible combinations of commitment. Cell sizes within these eight possible combinations ranged from 7 to 34. Multiple analysis of covariance (MANCOVA) and analysis of covariance (ANCOVA) were then used to examine the main and interaction effects of these three forms of commitment on the above-mentioned 14 dependent variables, with age, education, and type of agent serving as covariates.



MANCOVA, controlling for the demographic covariates, was used to assess the overall impact of each form of commitment on the dependent variables.(1) Because of the variety of dependent variables of interest--some of which constituted perceptions, some attitudes, and others which were behavioral in nature--separate MANCOVAs were conducted on each set of dependent variables: job perceptions, job attitudes, and job performance. Results revealed significant MANCOVA main effects for job involvement on job perceptions (F = 2.41, p |is less than or equal to~ 0.05) and job attitudes (F = 2.88, p |is less than or equal to~ 0.05), while the MANCOVA associated with performance was not significant. Professional commitment exhibited strong main effects on all three sets of variables, significantly affecting job perceptions (F = 5.62, p |is less than or equal to~ 0.001), job attitudes (F = 6.41, p |is less than or equal to~ 0.001), and performance variables (F = 9.41, p |is less than or equal to~ 0.001). Community commitment had a statistically significant impact on job perceptions (F = 2.79, p |is less than or equal to~ 0.01) and job attitudes (F = 2.54, p |is less than or equal to~ 0.05) but no relationship with the performance variables. In order to further explore these findings, separate ANCOVAs were conducted on each of the dependent variables.

Impact of Job Involvement

The ANCOVA results indicated that job involvement significantly affected five of the dependent variables. As shown in Table 3, job involvement significantly affected two job perceptions--perceptions of task variety and task significance--and three of the job attitude measures--satisfaction with work and with pay, and one's attitude toward remaining in the insurance profession. Specifically, agents classified as high in job involvement perceived their jobs as having greater task variety (|M.sub.low JI~ = 5.00, |M.sub.high JI~ = 5.58, p |is less than or equal to~ 0.01) and greater task significance (|M.sub.low JI~ = 5.50, |M.sub.high JI~ = 6.13, p |is less than or equal to~ 0.01) than did agents low in job involvement. Moreover, agents highly involved in their jobs reported significantly higher levels of satisfaction with work (|M.sub.low JI~ = 2.01, |M.sub.high JI~ = 2.24, p |is less than or equal to~ 0.01), higher levels of satisfaction with pay (|M.sub.low JI~ = 1.80, |M.sub.high JI~ = 2.18, p |is less than or equal to~ 0.05), and a higher probability of remaining in the insurance profession (|M.sub.low JI~ = 5.08, |M.sub.high JI~ = 5.55, p |is less than or equal to~ 0.05) than agents classified as low in job involvement. The amount of variance in these dependent variables explained by job involvement, as measured by ||Eta~.sup.2~, ranged from 3.4 percent in probability of remaining in insurance to 6.7 percent in satisfaction with work. In addition, the impact of job involvement on the job perception of feedback approached significance (p |is less than or equal to~ 0.10) as shown in Table 3. These results provide partial support for Proposition 1. That is, job involvement appears to affect some job perceptions and attitudes, but it has no impact on insurance agent performance, at least as measured in this study.

Impact of Professional Commitment

Professional commitment significantly affected nine of the fourteen dependent variables and exerted an especially strong influence on job perceptions. Agents highly committed to their profession perceived their jobs more favorably than those less committed to their profession. They perceived higher levels of task autonomy (|M.sub.low PC~ = 5.62, |M.sub.high PC~ = 6.25, p |is less than or equal to~ 0.01), task identity (|M.sub.low PC~ = 5.19, |M.sub.high PC~ = 5.85, p |is less than or equal to~ 0.01), task variety (|M.sub.low PC~ = 4.84, |M.sub.high PC~ = 5.77, p |is less than or equal to~ 0.001), task significance (|M.sub.low PC~ = 5.40, |M.sub.high PC~ = 6.27, p |is less than or equal to~ 0.001), and task feedback (|M.sub.low PC~ = 5.24, |M.sub.high PC~ = 5.90, p |is less than or equal to~ 0.001).

Professional commitment similarly affected most job attitudes. Agents classified as highly committed professionally reported higher levels of satisfaction with both work (|M.sub.low PC~ = 2.00, |M.sub.high PC~ = 2.27, p |is less than or equal to~ 0.01) and pay (|M.sub.low PC~ = 1.68, |M.sub.high PC~ = 2.34, p |is less than or equal to~ 0.001) than agents low in professional commitment. These same agents also reported a higher probability of remaining in the insurance profession (|M.sub.low PC~ = 5.03, |M.sub.high PC~ = 5.60, p |is less than or equal to~ 0.05) than the agents lower in professional commitment.

Finally, professional commitment significantly affected one of the performance measures and approached statistical significance in another. Highly professionally committed agents reported higher levels of annual income (|M.sub.low PC~ = 2.56, |M.sub.high PC~ = 3.43, p |is less than or equal to~ 0.001) than agents less committed to their profession. More professionally committed agents reported greater percentages of their income coming from insurance business sources than less committed agents, although the difference was not statistically significant.

These results provide relatively strong support for Proposition 2. Professional commitment appears to affect insurance agent job perceptions and attitudes. Professional commitment was also found to account for a substantial 19.2 percent of the variance in annual income of the agents represented in this sample. Overall, the amount of variance explained in these dependent variables by professional commitment, as measured by ||Eta~.sup.2~, ranged from 3.4 percent for probability of remaining in the insurance profession to 19.2 percent for annual income.

Impact of Community Commitment

Community commitment significantly affected one job perception measure (probability of getting another job in the community) and one job attitude (satisfaction with pay). Agents more committed to their communities reported enhanced perceived probabilities of obtaining another job in the community (|M.sub.low CC~ = 4.04, |M.sub.high CC~ = 4.81, p |is less than or equal to~ 0.01) and, contrary TABULAR DATA OMITTED to Proposition 3, lower levels of pay satisfaction (|M.sub.low CC~ = 2.04, |M.sub.high CC~ = 1.92, p |is less than or equal to~ 0.05). The amount of variance in these dependent variables explained by community commitment, as measured by ||Eta~.sup.2~, ranged from 5.3 percent of the variance in pay satisfaction to 6.4 percent of the variance in the probability of obtaining another job in the community. Although the MANCOVA results for community commitment were statistically significant for job perceptions and job attitudes, ANCOVA results reveal this to be the byproduct of a strong significant impact on only one job perception and one job attitude variable. Consequently, these results provide little support for Proposition 3.

Interaction Effects

Virtually no support was found for Proposition 4. When the three forms of commitment under investigation impacted on job perceptions, attitudes, and performance behaviors, they did so independently of one another. Evidence for the lack of significant interaction effects is reported in Table 3. Only six of a possible 56 interaction effects reached statistical significance, a number of significant findings that might be due to chance. Job involvement and professional commitment interacted to affect one's reported probability of obtaining another job in the community, while professional commitment and community commitment interacted to affect satisfaction with pay. The amount of variance explained by these interactions was 7.2 percent and 6.0 percent, respectively. Job involvement and community commitment were represented in three rather disparate interactions involving a job perception (variety, 4.2 percent), an attitude (satisfaction with pay, 5.1 percent), and a performance variable (annual income, 3.7 percent). Finally, a lone three-way interaction explaining 3.9 percent of the variation was observed in conjunction with satisfaction with pay.


This study empirically verifies a commonly held belief about and a current movement within the insurance industry: the need to enhance professionalism in insurance sales. Enhancing commitment to the insurance sales profession appears to have an impact on how agents view their jobs, their attitudes, and their performance, as reflected in annual income.

These results provide considerable evidence to suggest that the three forms of commitment examined herein operate independently. The impact of professional commitment was pervasive with respect to job perceptions and job attitudes, with nine of fourteen relationships observed reaching statistical significance. Although its association with performance was limited, it did manifest the single largest effect (19.2 percent) on any dependent variable, annual earned income. Job involvement demonstrated an intermediary capacity to explain variance with five significant findings, three entailing attitudes. Only two significant relationships involving community commitment were detected. Collectively, these findings complement previous research (e.g., Brooke, Russell, and Price, 1988; Morrow and McElroy, 1986) showing that various forms of commitment can be measured independently, that is, with little redundancy or overlap.

The findings associated with each dependent variable were mixed relative to their agreement with past research. Task variety and task significance, for example, were found to be positively related to job involvement, while other job design variables failed to reach statistical significance. The hypothesized inverse relationship between overqualification and job involvement was not borne out by the data, while perceived job mobility was unrelated to job involvement, as suggested by previous inquiry. There were some surprises in the job attitude category. While satisfaction with work and pay were positively related to job involvement, as anticipated, satisfaction with co-workers was unrelated, and the probability of remaining in the insurance profession was positively related to job involvement. The satisfaction with co-workers findings are perhaps attributable to the fact that some agents have peer co-workers while others do not and that respondents were confused as to whom the co-worker designation referred. Sales agents may view co-workers as competitors, and thus the conventional research on co-worker satisfaction does not apply to sales populations. Our finding with regard to the probability of remaining in the insurance profession is good news for those charged with managing agent turnover in that it implies that strategies for enhancing job involvement may increase intent to remain in the profession. Finally, the results connected with performance were counter to the majority of previous studies. Whether this is an aberrant finding or something peculiar to the insurance industry is a question for future investigation.

Findings associated with professional commitment were more consistent with prior, albeit limited, research. The nearly universal finding of a positive relationship between job design variables and professional commitment was surprising as this relationship has not been previously documented. Contrary to expectations, overqualification was not associated with professional commitment, and no relationship was postulated or found between perceived job mobility within the community and professional commitment. The pattern of findings connected with job attitudes was identical to that observed in conjunction with job involvement. With the exception of the co-worker satisfaction findings, these results were also consistent with previous investigations. It is also possible that the relationship between professional commitment and the probability of remaining in the profession was inflated because of the "intention to remain" items embedded in the professional commitment measure. Lastly, the performance findings indicated a sizable relationship between professional commitment and annual earned income. The implications of this finding are potentially far-reaching; that is, activities which promote professional loyalty among insurance agents are associated with higher incomes. This finding suggests that the option of developing a sense of professionalism may spill over into many aspects of one's life.

The inclusion of community commitment in this study was admittedly exploratory, and little previous research was available from which to derive hypotheses. Not surprisingly, community commitment was positively related to perceived job mobility. The finding of lower satisfaction with pay among agents high in community commitment is rather disquieting and not readily explainable. Taken together, however, the data suggest that community commitment effects are rather isolated and do not appear to be systematically related to job perceptions, attitudes, or behaviors.

Another issue that can be addressed by these findings is the utility of interactive models of commitment. Contrary to Blau and Boal's (1987) interactive model of commitment, these data add to the growing body of research suggesting that various forms of commitment do not interact in unique ways to affect job perceptions, attitudes, and behaviors (e.g., Huselid and Day, 1991). If future research upholds these independent commitment effects, then understanding job perceptions, attitudes, and behaviors may be less complicated than Blau and Boal proposed. Similarly, commitment need not be considered a zero-sum game in which the benefits of one form of commitment come at the expense of another form of commitment.


Several limitations of this study should be noted. The sample used was essentially one of convenience and may not be representative of all insurance agents. It included only rural agents and, thus, the results observed may not extend to agents working in more urbanized environments. Future research employing a cross-section of rural and urban agents may shed additional light on the role of community commitment. Also, regional differences in the insurance firms and agents in the state studied may affect the generalizability of these findings.

Several measurement limitations also merit comment. While the use of standardized measures avoided problems of low reliability in most cases, the satisfaction with work alpha coefficient (0.62) was slightly below the conventional standard of 0.70. In addition, two single-item measures were used: perceptions of mobility within the community and intent to remain in the profession. There was also some restriction in the range associated with the question about agents' percent of income from insurance (i.e., 51 percent reported the highest response option of 5, indicating that 76 to 100 percent of their income was from insurance activity). The attenuating effect of this restriction may explain the absence of significant findings associated with this variable.

Finally, given that the data were collected in a self-report, cross-sectional manner, the findings may have been affected by common method error variance, and any inferred causal connections are grounded in theory rather than in data. The causal ordering of the relationships observed may even be reversed; for example, positive job perceptions and attitudes as well as higher levels of annual income may cause agents to exhibit higher levels of professional commitment.

Implications for the Insurance Industry

To the extent that the insurance industry views agent turnover as excessive, the finding that the perceived probability of remaining in the profession is positively related to job involvement and professional commitment is important. Stated plainly, agents with higher levels of these forms of commitment were found to exhibit a stronger intent to remain in the profession. Consequently, insurance companies and agencies should place a high priority on allocating resources spent on agent training toward activities that enhance job involvement and, especially, professional commitment. Job involvement, for example, can be increased by elevating job related self-esteem, intra-salesforce competitions for acquiring product knowledge, and reduction of role ambiguity and conflict through training and socialization programs (Ingram, Lee, and Skinner, 1989). Examples of activities for enhancing professional commitment include establishing mentoring programs, college agents programs, intra-agency educational goal-setting, and reward structures that recognize continuing education units, certification programs, etc.

Future research should determine if insurance agents who quit and move to other insurance companies or leave the industry altogether hold differing commitment levels prior to termination. Such research would further establish the causal nature of the association found here between forms of work commitment and agent intentions to stay or quit. Unfortunately, little is known about this phenomenon since most studies evaluate only organizational and not professional turnover. For example, a recent study discusses why successful agents leave one independent agency for another but does not address the interindustry turnover issue (CPCU Journal, 1991).

Finally, since the postulated implications of community commitment were not borne out by this research, the findings serve to diminish the idea that psychological attachment to the community is important to insurance agents. From a practical standpoint, agents seeking to enhance their satisfaction and/or income would be better served by spending less time in community activities (e.g., officer positions in civic organizations) and more time in activities reflective of professional commitment (e.g., improving product knowledge).

These results suggest that a focus on community commitment should not be vigorously pursued. Perhaps a narrower conceptualization of community commitment emphasizing only an agent's willingness to serve members of the community would be more useful. Alternatively, other more precisely defined commitments, such as those to clients, may prove worthwhile. It may be that, rather than commitment to the community per se, commitment to a subset of the community--clients or potential clients, for example--may be the more important focal point. An analogous tack has already been proposed in the organizational commitment literature (Reichers, 1986; Zaccaro and Dobbins, 1989), where commitment to the work group has been offered as an alternative to conventional commitment to the organization. Such an alternative may be more viable because the client is a closer entity to the agent, one with whom the agent interacts on a frequent basis. Switching from a community-based to a client-based commitment focus would also put researchers in closer touch with industry trends. The current Total Quality Management movement with its customer orientation would provide a fertile backdrop in which to test the merits of client commitment. For now, however, this study suggests that efforts to increase professionalism by insurers are likely to provide the best return to companies and agents alike.

1 The results presented here focus on F values, calculated as a means of determining whether the relationship specified is statistically significant. The p values indicate the level of statistical significance associated with the F value (i.e., a p |is less than or equal to~ 0.05 means that we are confident that the observed relationship could not have occurred by chance more than five times out of 100). The |M.sub.low~ and |M.sub.high~ designations stand for the mean values of the dependent variable for respondents scoring less (|M.sub.low~) or more (|M.sub.high~) than the median on an independent variable.


Aranya, N., T. Kushnir, and A. Valency, 1986, Organizational Commitment in a Male Dominated Profession, Human Relations, 39: 433-448.

Blau, G. J., 1985, The Measurement and Prediction of Career Commitment, Journal of Occupational Psychology, 58: 277-288.

Blau, G. J., 1987, Using a Person-Environment Fit Model to Predict Job Involvement and Organizational Commitment, Journal of Vocational Behavior, 30: 240-257.

Blau, G. J., 1988, Further Exploring the Meaning and Measurement of Organizational Commitment, Journal of Vocational Behavior, 32: 284-297.

Blau, G. J., 1989, Testing the Generalizability of a Career Commitment Measure and Its Impact on Employee Turnover, Journal of Vocational Behavior, 35: 88-103.

Blau, G. J. and K. B. Boal, 1987, Conceptualizing How Job Involvement and Organizational Commitment Affect Turnover and Absenteeism, Academy of Management Review, 12: 288-300.

Brooke, P. P., Jr., D. W. Russell, and J. L. Price, 1988, Discriminant Validation of Measures of Job Satisfaction, Job Involvement, and Organizational Commitment, Journal of Applied Psychology, 73: 139-144.

Cherniss, C. and J. S. Kane, 1987, Public Sector Professionals, Job Characteristics, Satisfaction, and Aspirations for Intrinsic Fulfillment Through Work, Human Relations, 40: 125-136.

Chonko, L. B., 2986, Organizational Commitment in the Sales Force, Journal of Personal Selling and Sales Management, 6: 19-27.

Colarelli, S. M. and R. C. Bishop, 1990, Career Commitment: Functions, Correlates, and Management, Group and Organizational Studies, 15: 158-176.

CPCU Journal, 1991, Why Insurance Producers Leave Independent Agencies: A Study of Producer Turnover, 44(1): 12-19.

Cron, W. and J. W. Slocum, 1986, The Influence of Career Stages on Salespeople's Job Attitudes, Work Perceptions, and Performance, Journal of Marketing Research, 23: 119-129.

Dubinsky, A. J., T. L. Childers, S. J. Skinner, and E. Gencturk, 1988, Impact of Sales Supervisor Leadership Behavior on Insurance Agent Attitudes and Performance, Journal of Risk and Insurance, 55: 132-144.

Farrell, D. and C. E. Rusbult, 1981, Exchange Variables as Predictors of Job Satisfaction, Job Commitment, and Turnover: The Impact of Rewards, Costs, Alternatives, and Investments, Organizational Behavior and Human Performance, 27: 78-95.

Griffin, R. W., 1982, Task Design (Glenview, Ill.: Scott, Foresman).

Hackman, J. R. and G. R. Oldham, 1975, Development of the Job Diagnostic Survey, Journal of Applied Psychology, 60: 159-170.

Hafer, J. and B. A. McCuen, 1985, Antecedents of Performance and Satisfaction in a Service Sales Force as Compared to an Industrial Sales Force, Journal of Personal Selling and Sates Management, 5: 7-17.

Hanisch, K. A., 1992, The Job Descriptive Index Revisited: Questions About the Question Mark, Journal of Applied Psychology, 77: 377-382.

Hollenbeck, J. R. and C. R. Williams, 1986, Turnover Functionality Versus Turnover Frequency: A Note on Work Attitudes and Organizational Effectiveness, Journal of Applied Psychology, 71: 606-611.

Huselid, M. A. and N. E. Day, 1991, Organizational Commitment, Job Involvement, and Turnover: A Substantive and Methodological Analysis, Journal of Applied Psychology, 76: 380-391.

Ingram, T. N. and K. S. Lee, 1990, Sales Force Commitment and Turnover, Industrial Marketing Management, 19: 149-154.

Ingram, T. N., K. S. Lee, and S. J. Skinner, 1989, An Empirical Assessment of Salesperson Motivation, Commitment, and Job Outcomes, Journal of Personal Selling and Sales Management, 9: 25-33.

Johnston, M. W., A. Parasuraman, C. M. Futrell, and W. C. Black, 1990, A Longitudinal Assessment of the Impact of Selected Organizational Influences on Salespeople's Organizational Commitment During Early Employment, Journal of Marketing Research, 27: 333-344.

Kanungo, R. N., 1982, Measurement of Job and Work Involvement, Journal of Applied Psychology, 67: 341-349.

Khan, L. J. and P. C. Morrow, 1991, Objective and Subjective Underemployment Relationships to Job Satisfaction, Journal of Business Research, 22: 211-218.

King, C., 1992, Recruiting Drops: Is This a Sign of a Broader Malaise? National Underwriter (Life & Health, March 16), 15, 20.

Knowles, R. G., 1990, Agent Professionalism Touted, National Underwriter (Property & Casualty, July 30), 23, 27.

Koslowsky, M., T. Caspy, and M. Lazar, 1990, An Empirical Comparison of Commitment Scales, Journal of Applied Social Psychology, 20: 1063-1075.

MacKenzie, S. B., P. M. Podsakoff, and R. Fetter, 1991, Organizational Citizenship Behavior and Objective Productivity as Determinants of Managerial Evaluations of Salespersons' Performance, Organizational Behavior and Human Decision Processes, 50: 123-150.

Mathieu, J. E. and J. L. Farr, 1991, Further Evidence for the Discriminant Validity of Measures of Organizational Commitment, Job Involvement, and Job Satisfaction, Journal of Applied Psychology, 76: 127-133.

Mathieu, J. E. and S. S. Kohler, 1990, A Cross-level Examination of Group Absence Influences on Individual Absence, Journal of Applied Psychology, 75: 217-220.

Mathieu, J. E. and D. M. Zajac, 1990, A Review and Meta-analysis of the Antecedents, Correlates, and Consequences of Organizational Commitment, Psychological Bulletin, 108: 171-194.

McCloskey, J. C. and B. E. McCain, 1987, Satisfaction, Commitment and Professionalism of Newly Employed Nurses, Images, 19: 20-24.

Morrow, P. C., 1983, Concept Redundancy in Organizational Research: The Case of Work Commitment, Academy of Management Review, 8: 486-500.

Morrow, P. C., 1993, The Theory and Measurement of Work Commitment (Greenwich, Conn.: JAI Press).

Morrow, P. C. and J. C. McElroy, 1986, On Assessing Measures of Work Commitment, Journal of Occupational Behavior, 7: 139-145.

Morrow, P. C., E. J. Mullen, and J. C. McElroy, 1990, Vocational Behavior 1989: The Year in Review, Journal of Vocational Behavior, 37: 121-195.

Morrow, P. C. and R. E. Wirth, 1989, Work Commitment among Salaried Professionals, Journal of Vocational Behavior, 34: 40-56.

Mowday, R. T., L. W. Porter, and R. M. Steers, 1982, Employee-Organization Linkages (New York: Academic Press).

Mowday, R. T., R. M. Steers, and L. W. Porter, 1979, The Measurement of Organizational Commitment, Journal of Vocational Behavior, 14: 224-247.

Norris, D. R. and R. E. Niebuhr, 1983, Professionalism, Organizational Commitment and Job Satisfaction in an Accounting Organization, Accounting, Organizations and Society, 9: 49-59.

Oliver, R. L., 1974, Expectancy Theory Predictions of Salesmen's Performance, Journal of Marketing Research, 11: 243-253.

Parasuraman, S. and J. A. Alutto, 1984, Sources and Outcomes of Stress in Organizational Settings: Toward the Development of a Structured Model, Academy of Management Journal, 27: 330-350.

Pinder, C. C., 1984, Work Motivation: Theory, Issues and Applications (Glenview, Ill.: Scott, Foresman).

Rabinowitz, S. and D. T. Hall, 1977, Organizational Research on Job Involvement, Psychological Bulletin, 84: 265-288.

Reichers, A. E., 1986, Conflict and Organizational Commitments, Journal of Applied Psychology, 71: 508-514.

Rusbult, C. E. and D. Farrell, 1983, A Longitudinal Test of the Investment Model: The Impact of Job Satisfaction, Job Commitment, and Turnover on Variations in Rewards, Costs, Alternatives, and Investments, Journal of Applied Psychology, 68: 429-438.

Salancik, G. R. and J. Pfeffer, 1978, A Social Information Processing Approach to Job Attitudes and Task Design, Administrative Science Quarterly, 23: 224-253.

Sekaran, U., 1989, Paths to Job Satisfaction of Bank Employees, Journal of Organizational Behavior, 10: 347-359.

Smith, P. C., L. M. Kendall, and C. L. Hulin, 1969, The Measurement of Satisfaction in Work and Retirement: A Strategy for the Study of Attitudes (Chicago: Rand McNally).

Staw, B. M., 1986, Organizational Psychology and the Pursuit of the Happy/Productive Worker, California Management Review, 28: 40-53.

Steffy, B. D. and J. W. Jones, 1988, The Impact of Family and Career Planning Variables on the Organizational, Career, and Community Commitment of Professional Women, Journal of Vocational Behavior, 32: 196-212.

Tyagi, P. K., 1985, Relative Importance of Key Job Dimensions and Leadership Behaviors in Motivating Salesperson Work Performance, Journal of Marketing, 3: 76-86.

Vandenberg, R. J. and C. E. Lance, 1992, Examining the Causal Order of Job Satisfaction and Organizational Commitment, Journal of Management, 18: 153-167.

Wiener, Y. and Y. Vardi, 1980, Relationships Between Job, Organization, and Career Commitments and Work Outcomes--An Integrative Approach, Organizational Behavior and Human Performance, 26: 81-96.

Zaccaro, S. J. and G. H. Dobbins, 1989, Contrasting Group and Organizational Commitment: Evidence for Differences Among Multi-Level Attachments, Journal of Organizational Behavior 10: 267-273.

James C. McElroy is Chair of the Departments of Management, Marketing, and Transportation and Logistics at Iowa State University. Paula C. Morrow is Director of the Industrial Relations Center, Iowa State University. Mark L. Power is Associate Professor of Finance at Iowa State University. Zafar Iqbal is currently a Ph.D. student at the University of Pittsburgh.

The authors would like to acknowledge the editor and two anonymous reviewers for their contributions to this article.
COPYRIGHT 1993 American Risk and Insurance Association, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:McElroy, James C.; Morrow, Paula C.; Power, Mark L.; Iqbal, Zafar
Publication:Journal of Risk and Insurance
Date:Sep 1, 1993
Previous Article:From the library shelf.
Next Article:The inefficiency of private constant annuities.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters