White vs. blue: does the collar color affect job attitudes and behaviors?

Rozell, Elizabeth J. ; Pettijohn, Charles E. ; Parker, R. Stephen 等

INTRODUCTION

Researchers continue to be curious about the role of job attitudes and behaviors as they relate to a variety of workplace variables. Specifically, past empirical studies have investigated a variety of work attitudes related to dispositional affectivity, with most studies examining a few workplace attitudinal variables within a study. In the current study, we investigate the impact of dispositional affectivity on a wide spectrum of workplace attitudes and behaviors using a sample of white and blue collar workers. Much of the research in the area of dispositional affectivity has largely focused on negative affectivity (Fredrickson & Losada, 2005; Hochwarter et al, 2003). As such, researchers have criticized the exclusive focus on the negative affectivity construct (Fortunato & Stone-Romero, 1999; Stone-Romero, 2005). The positive psychology movement, first advocated by Seligman (2000), has shifted the focus to positive affectivity and its advantage for promoting a healthy organizational environment. Hence, there has been a shift in the literature on dispositional affectivity to a greater emphasis on evaluating positive affect and its statistical relationships with a variety of variables. This paper looks at both positive and negative affect and their impact on a range of job attitudes and behaviors.

In most research regarding workplace attitudes and behaviors, research is conducted with employee samples without regard to the 'type of job'. The assumption seems to be one that holds that workers are workers. Thus, factors that may affect one group will logically affect another group. However, what if this assumption fails to hold true? What if there are differences in attitudes and behaviors that exist independent of one's workplace environment? It seems logical to assume that individuals enter the workplace with certain predispositions that have been formed as a result of their experiences and perhaps genetics. Further, it seems logical to assume that an element of self-selection exists in the workplace, with certain individuals possessing specific predispositions selecting careers that match these predispositions. Thus, this research fills a void in the literature pertaining to differences in white and blue collar workers. Indeed, little, if any recent research has examined the attitudinal and behavioral differences between white and blue-collar workers. Therefore, another purpose of the current research study was to investigate these differences.

We begin by reviewing the pertinent literature for dispositional affectivity and each of the individual difference variables. Next, we consider the conceptual linkages between these variables, as well as their effects on work-related attitudes and behaviors such as job satisfaction, organizational commitment, turnover intentions, absenteeism, and tardiness. Drawing on this discussion, we use regression analysis comparing the white and blue-collar samples to test a set of hypotheses regarding the relationships between dispositional affectivity and certain work attitudes and behaviors. We examine these models using a powerful sample of 595 employees (an 85% response rate) of a Midwestern manufacturing company. We conclude with a discussion of the results and their implications for management research and practice.

LITERATURE REVIEW

Dispositional Affectivity and Work-Related Attitudes and Behaviors

Social scientists have long been intrigued by individual differences in people's interpretations of their own emotional experiences (Berry & Hansen, 1996). In particular, research shows that some individuals report experiencing increased amounts of positive emotions relative to others. The phenomenon is referred to as positive affect, and these persons are usually self-described as joyful, exhilarated, excited, and enthusiastic. Those low in PA have been described as listless, lethargic, drowsy, apathetic, and dull (Cropanzano et al, 1993; Watson & Tellegen, 1985). In contrast, other individuals describe themselves as experiencing greater amounts of negative feelings than others, and are often referred to as high-negative-affect individuals (Berry & Hansen, 1996; Cropanzano et al., 1993). Such individuals report being afraid, anxious, angry, and tend to be nervous and tense. Those low in NA tend to view conditions as less upsetting and stressful than high NA individuals (Chiu & Francesco, 2003). Interestingly, the research on dispositional affectivity has shown that there are two general dimensions of affective responding: trait-positive affect (PA) and trait-negative affect (NA). These dimensions do not appear to represent opposite ends of a continuum; but rather they are independent of one another (Berry & Hansen, 1996; Diener & Emmons, 1985). That is, it is possible for an individual to be high on both, low on both, or high on one but not the other (George, 1992; Watson & Tellegen, 1985). An individual who rates high on both dimensions would be characterized as quite emotional, and would experience fluctuating moods in response to environmental stimuli (Diener & Emmons, 1985). In sharp contrast is the individual that rates low on both who would likely display little affect; i.e. the person would likely be unemotional and unresponsive (Cropanzano et al., 1993).

Several researchers have documented the significant relationship between dispositional affectivity and work attitudes. For example, an inverse relationship has been found to exist between NA and job satisfaction ENRfu(Levin & Stokes, 1989; Staw, Bell, & Clausen, 1986). A minority of researchers has criticized negative affectivity as a construct (Stone-Romero, 2005) citing construct validity problems, however, several others have shown success in using an established and validated scale (Watson et al., 1988; Watson, Clark, & Carey, 1988; Watson 1988a, 1988b). Researchers have documented that NA may be negatively correlated with not only job satisfaction, but also organizational commitment, and positively correlated with turnover intentions; the exact opposite pattern of correlations has been obtained for PA ENRfu(Cropanzano et al., 1993). One explanation for these relationships is that work attitudes are primarily a function of how an individual affectively responds to his or her work environment, and are therefore influenced by one's underlying affective disposition. Consequently, high PA individuals are likely to exhibit extremely positive responses to their work environment which are reflected in their work attitudes, while extreme negative responses are usually seen in high NA persons ENRfu(George, 1992).

Research notes the tendency of individuals to be dispositionaly inclined to form positive or negative attitudes about their work (Cropanzano et al., 1993). Interestingly, Arvey, Bouchard, Segal, and Abramson (1989) demonstrated that approximately 30% of the observed variance in general job satisfaction was attributable to genetic factors. Longitudinal studies indicate that scores on job satisfaction measures remain correlated over time, and that this relationship holds even when individuals change employers or occupations (Staw et al., 1986; Staw & Ross, 1985). These findings do not mean that work attitudes are entirely stable, or that the job context is unimportant; in actuality, work attitudes do indeed fluctuate over time. Instead, these longitudinal studies are consistent with the view that while work attitudes vary as a function of changes in the work setting (Cropanzano & James, 1990; Newton & Keenan, 1991), the rank ordering of individuals' attitudes remains relatively stable, and that such stability can be attributed to certain underlying personality dispositions (George, 1992) such as positive or negative affectivity (Cropanzano et al., 1993).

Research by Fredrickson (1998, 2001) has proposed a "broaden-and-build" theory of positive affect which contends that individuals who experience positive emotions and generally experience "chronic" positive affectivity are able to adapt and be flexible to workplace changes. Further, it has been proposed that positive affect individuals possess a wider range of thoughts than individuals who experience negative affectivity on a regular basis. Recent empirical support has shown how positive affect influences behavioral responses (Fredrickson & Branigan, 2005), and psychological growth (Fredrickson, Tugade, Waugh & Larkin, 2003). Indeed, Fredrickson and Losada (2005) contend that PA individuals experience a broader range of thoughts that are proactive in nature as opposed to thoughts that are single-mindedly stagnant, which in essence broadens their behavioral repertoire. Based on this reasoning, Fredrickson (2001) hypothesized that positive affectivity may lead to an increase in psychological resources over time.

In a recent study by Fisher (2002), it was found that positive affectivity was predictive of affective commitment and helping behaviors. Interestingly, in the same study, intention to leave was predicted by work attitudes rather than affective reactions. Further, research has indicated that positive affectivity is characteristic of employees that are successful at dealing with organizational stressors (Isen et al, 1987; Fredrickson et al 2003; Fredrickson & Branigan, 2005; Watson, Clark, & Tellegen, 1988). Moreover, in a study by Chiu and Francesco (2003) it was found that dispositional affectivity predicted turnover intentions. Based on the research outlined above, we hypothesized the following:

H1a: Higher positive affect levels will be significantly and positively related to organizational commitment levels for both white and blue collar workers.

H1b: Higher negative affect levels will be significantly and negatively related to organizational commitment levels for both white and blue collar workers.

H2a: Higher positive affect levels will be significantly and negatively related to turnover intention levels for both white and blue collar workers.

H2b: Higher negative affect levels will be significantly and positively related to turnover intention levels for both white and blue collar workers.

Most measures of job satisfaction include questions containing both positively and negatively worded items, for example, "my job makes me content", and "my job is disagreeable" from the Job in General scale by Ironson, Smith, Brannick, Gibson, and Paul (1989). Fisher (2002) contends that items such as these most likely trigger recall of both positive and negative emotions experienced in the workplace. Indeed, Price (2001) notes that PA and NA may impact job satisfaction through selective perception. That is, PA individuals may selectively perceive positive aspects of the job rather than the negative, resulting in greater job satisfaction. Other researchers have confirmed a similar relationship between dispositional affectivity and job satisfaction (Judge, 1993; Agho et al, 1992; Levin & Stokes, 1989; Cropanzano et al, 1993). Hence, we hypothesized the following:

H3a: Higher positive affect levels will be significantly and positively related to job satisfaction levels for both white and blue collar workers.

H3b: Higher negative affect levels will be significantly and negatively related to job satisfaction levels for both white and blue collar workers.

Other workplace behaviors have also been linked to dispositional affectivity. Interestingly, Iverson and Deery (2001) found that high PA individuals were associated with increased tardiness and early departure but decreased absenteeism. These same authors note the lack of empirical research exploring the causes of tardiness and absenteeism. Indeed, most research on these two workplace variables has focused on the Big Five personality traits (Iverson & Deery, 2001). For example, Cooper and Payne (1967) found that extraversion was significantly associated with both tardiness and absenteeism. In a more recent example, Furnham and Miller (1997) found that PA had a positive relationship to absenteeism. With regard to NA, Ferris, Youngblood, and Yates (1985) and Cooper and Payne (1967) both found that anxiety was associated with absenteeism. Based on the research noted above, we hypothesized the following:

H4a: Higher positive affect levels will be significantly and negatively related to levels of absenteeism for both white and blue collar workers.

H4b: Higher negative affect levels will be significantly and positively related to levels of absenteeism for both white and blue collar workers.

H5a: Higher positive affect levels will be significantly and negatively related to levels of tardiness for both white and blue collar workers.

H5b: Higher negative affect levels will be significantly and positively related to levels of tardiness for both white and blue collar workers.

METHODS

It was determined that the sample for this study should be drawn from a firm engaged in manufacturing operations employing both white and blue-collar workers. This firm had approximately 400 employees engaged in blue collar shift-work and 300 white-collar workers. Therefore, the population consisted of 700 hourly employees of a manufacturing firm located in the Midwestern United States. The final sample size resulted in 594 workers.

In the construction of the survey, a variety of standardized instruments were used to measure the variables included in the research model. Descriptions of these measures and the evidence of reliability and validity are provided below.

Positive and negative affect were measured using the Positive and Negative Affect Schedule (PANAS) developed by Watson, Clark, and Tellegen (1988). The PANAS includes a list of 20 mood-relevant adjectives, of which 10 indicate positive (e.g., active, enthusiastic) and 10 indicate negative (e.g., angry, afraid) mood states. Respondents are instructed to "indicate to what extent you generally feel this way, that is, how you feel on the average." Extensive validity evidence is provided by Watson et al. (1988), Watson, Clark, and Carey (1988), and Watson (1988a; 1988b). Alpha coefficients of .86 and .80 for the PA and NA scales, respectively, were obtained in the current study.

A measure of intent to leave developed by O'Reilly, Chatman, and Caldwell (1991) was employed in this study. This scale is composed of four 7-point Likert-type questions: (1) "To what extent would you prefer another more ideal job than the one you now work in?" (2) "To what extent have you thought seriously about changing organizations since beginning to work here?" (3) "How long do you intend to remain with this organization?" (4) "If you have your own way, will you be working for this organization three years from now?" Each employee was asked to respond to these questions. A coefficient alpha of .80 for this scale was obtained in this research.

Tardiness was measured by a single item which read "How frequently do you arrive at least 10 minutes late to work?" A 7 point Likert scale was used ranging from "never" (1) to "very often (7)." Absenteeism was also measured with a single item which read "Not counting holidays, vacation days, hospitalizations and surgeries, how many days of scheduled work did you miss over the past year?"

In a review of the organizational commitment literature, Meyer and Allen (1991) identified affective, continuance, and normative commitment as three distinctive components of commitment. Affective commitment refers to an affective attachment to the organization. Continuance commitment involves a perceived cost of leaving the organization. Normative commitment stems from a perceived obligation to remain with the organization. Based on the Organizational Commitment Questionnaire developed by Mowday et al. (1982), Allen and Meyer (1990) developed and validated separate measures for each component. Given the focus of the current study, we included Allen and Meyer's 8-item Affective Commitment Scale (ACS) as our measure of organizational commitment. Coefficient alphas for the ACS of .87 and .90 were obtained by Allen and Meyer, and in the present study, respectively.

Overall job satisfaction was measured using the 18-item "Job in General" (JIG) scale (Ironson, Smith, Brannick, Gibson, & Paul, 1989) from the revised version of the Job Descriptive Index (JDI) (Smith, Kendall, & Hulin, 1969). Validation evidence for the JIG is provided by Ironson et al. (1989); coefficient alphas for the JIG scale range from .91 to .95. In the present study, an alpha coefficient of .89 was obtained. Additionally, a single item was used to assess job satisfaction. Subjects were asked to respond to the following question using a 7-point Likert scale: "All in all, how satisfied are you with your current job?"

The administration of the instrument packets was conducted in cooperation with contact members of the targeted organization. Specifically, data collection was designed to reach all employees at the participating manufacturing firm. The method used was a "drop-off method whereby contact persons in the firm distributed the survey packets to all employees in their work units. Respondents completed the instruments during normal work hours, and returned them directly to the researchers using a pre-addressed and pre-paid postage packet.

Of the survey packets distributed, 594 were completed and returned for a response rate of 85 percent. Table 1 provides a summary of the demographic attributes of the subjects.

ANALYSIS

The research plan was designed to first determine whether there were significant relationships between the variables of interest and positive and negative affect levels exhibited by both white and blue collar workers. Second, the research was then focused on whether white and blue collar workers exhibited similar levels of organizational commitment, turnover intentions, job satisfaction, tardiness, absenteeism, positive affect, and negative affect. The third research question was whether the regression equations relating positive and negative affect levels to the dependent variables of interest were fundamentally equal with regard to the statistical relationships.

Since the research was designed to compare the mean levels of organizational commitment, job satisfaction, absenteeism, tardiness, and turnover intentions of workers, regression analysis were used to investigate the relationships between worker type and the outcome variables. A separate regression analysis was performed for each of the outcome variables. Further, we used the Chow test to compare the equality of a series of regression equations which had evaluated the statistical relationships between the variables. The results of all analyses are presented in the results section.

RESULTS

As the results in Table 2 indicate, H1 is supported by the results. Both groups of employees, white and blue collar, show significant univariate relationships between their levels of positive affect, negative affect, and organizational commitment. Positive affect scores are significantly and positively related to levels of organizational commitment. Thus, as positive affect levels rise, levels of organizational commitment also rise. Conversely, as levels of negative affect increase, both groups of workers' organizational commitment levels decline significantly. An examination of the results indicates that positive affect contributed over 20 percent of the explanation of the variation in organizational commitment levels for both blue and white collar workers (R2 > .20).

Table 2 also indicates that H2 was supported by the results, as turnover intentions are significantly related to the workers' levels of positive/negative affect. As levels of positive affect increase turnover intentions decline and as levels of negative affect increase turnover intentions increase. While these findings are significant for both blue and white collar workers, an examination of the results indicates that 22 percent of the variance (R2 = .22) in turnover intention levels was explained by positive affect scores for blue collar workers, but less than 10 percent of the variance (R2 = .09) in turnover intention levels was explained by positive affect scores for white collar workers.

The findings also lend support to H3. As indicated in Table 2, positive affect scores are significantly related to the levels of job satisfaction for both white and blue collar employees. Also, the findings show that higher levels of negative affect lead to significantly lower levels of job satisfaction (lower levels of negative affect lead to higher levels of job satisfaction). Positive affect scores explain over 20 percent of the variation (R2 > .20) in job satisfaction levels.

The fourth hypothesis is not supported by the findings as neither positive nor negative affect are significantly related to worker absenteeism. As shown in the table, for both white and blue collar workers, the results indicate the affect levels are not significantly related to levels of absenteeism.

H5 is largely supported by the findings as negative affect levels are positively related to worker tardiness for both white and blue collar employees. However, with regard to the levels of positive affect, the relationship is significant only for white collar employees. As indicated in Table 2, as white collar employee levels of positive affect increase, worker tardiness levels decline. However, the relationship between tardiness and positive affect levels is not significant for blue collar workers.

While the tests of the hypotheses provide some insight into the issues regarding whether blue and white collar workers are substantially equal with regard to the relationships existing between affect levels (positive/negative) and the dependent variables (organizational commitment, turnover intentions, job satisfaction, absenteeism, and tardiness), questions may still remain regarding the equality of the two groups and their relationships. To determine whether differences exist between the two groups, the Chow (1960) test was used. Chow (1960) developed an equation designed to determine the degree to which two sets of observations might be "regarded as belonging to the same regression model." The equation for assessing these differences is provided below:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Where:

RSS = residual sum of squares - pooled

RSS1 = residual sum of squares - group 1

RSS2 = residual sum of squares - group 2

n1 = number of observations - group 1

n2 = number of observations - group 2

k = number of parameters

Using this test, models were tested which evaluated the relationships existing between positive/negative affect and the relevant dependent variables (organizational commitment, turnover intentions, job satisfaction, absenteeism, and tardiness). The results of these regressions and the Chow Test are provided in Table 3 and are discussed below.

With regard to organizational commitment, the relationship existing between worker affect levels and organizational commitment is not significantly different between the two worker categories. The regression equations indicate that only positive affect is significantly related to organizational commitment in the regression model, while negative affect is not significantly related. The findings in this case indicated that there are no differences based on worker type (white vs. blue).

Similar findings exist pertaining to the relationships between positive/negative affect and turnover intentions. As shown, the differences between the two regression equations are not significant, and one can thus assume that the two models are equal. However, for these two equations, both positive and negative affect levels are significantly related to turnover intentions.

The Chow Test indicates significant differences between the two regression equations computed for job satisfaction. As shown, the Chow Test reveals that the two equations are not equal (p = .0021). A review of the findings indicates that the differences may lie in the increased size of the standardized betas for the blue collar grouping. As may be noted, the blue collar betas are .56 and .51 for positive and negative affect levels while the white collar betas are .45 and .43 respectively.

The results also indicate that differences between the two regression equations computed for absenteeism exist. However, in this case, interpretation is limited because the regression models themselves are not significant for either the white or blue collar workers. However, a review of the results indicates that the differences may lie in the increased standardized beta coefficient pertaining to the relationship between negative affect and absenteeism for the blue collar sample.

Finally, the Chow Test indicates that the two regression equations are not significantly different as they relate to the relationship between the employees' affect (positive/negative) levels and their tardiness. Nevertheless, the findings show that tardiness is significantly affected by positive affect for the white collar grouping and significantly affected by negative affect for the blue collar grouping. Yet, these differences are not significant and thus one cannot interpret the two equations as being significantly different.

DISCUSSION AND IMPLICATIONS

The findings clearly indicate that the workers' relative PA/NA levels are significantly related to their job satisfaction, organizational commitment, turnover intentions, and tardiness. These findings suggest that firms could logically use PA/NA as a tool in their employee selection and training processes. By selecting employees with higher levels of positive affect and lower levels of negative affect, firms might discover that their employees are more satisfied, more organizationally committed, and have lower levels of turnover intentions.

Indeed, these findings suggest that managers might use positive and negative affect levels as a selection tool. It has generally been assumed that "positive people" make better employees. However, these findings indicate that being "positive" alone is not the "ideal" circumstance. Similarly, the results indicate that one's being negative alone is not the "worst" circumstance. Instead the findings show that one who has the following traits: positive, happy, perceiving the "best" in situations; combined with traits of being low in anger, negativity, etc. will obtain the optimal work attitudes. On the separate end of the continuum, the individual who has traits that don't allow him/her to experience joy, to see the good in situations, or to be positive; combined with the worker who possesses traits that make him/her angry, negative, etc. will possess the least desirable work attitudes. However, combinations of these traits, may allow a worker to experience less than optimal work attitudes.

Thus, it may be concluded that managers might use positive and negative affect levels of their employees discriminately. For example, the fact that a worker has a high positive affect score (or a high negative affect score) alone should not necessarily qualify (or disqualify) him/her for a job. Instead, the manager needs to assess the combinations of affect levels to use this as a tool in selection.

A manager interested in selecting and developing high performing workers may discover that the measurement of the individual's dispositional affect is an indicator of his/her likely work attitudes. However, the findings in this study indicate that the relationship is not a clear-cut as one might speculate. Instead, the findings indicate that combinations of positive and negative affect levels are related to work attitudes. Based on this finding, managers should evaluate the applicants' levels of both positive and negative affect to ensure that those with the lowest (i.e., worst) combination of scores are not selected and then encourage the development of higher levels of positive affect and lower levels of negative affect through selection decisions.

This study examines a topic which has not been studied in depth in nearly 30 years. Indeed, an important purpose of the current study was to assess differences in work attitudes in blue and white collar samples. Further, a strength of the current study was the high response rate (85%) which reduces non-response bias in the data. Within the sample, significant differences were found with regard to dispositional affectivity and job satisfaction and absenteeism. In both instances, it was found that the relationships were stronger for the blue collar sample. It is interesting to note that there were no significant differences found with regard to dispositional affectivity and organizational commitment, turnover, and tardiness. Hence, companies should be aware of the strong relationship that exists for blue collar workers in terms of dispositional affectivity and job satisfaction and absenteeism. That is, companies desiring high levels of both of these job attitudes should certainly pay attention to their blue collar workers. Selecting for certain levels of both positive and negative affectivity might be advantageous for companies employing large numbers of blue collar workers. Given the mounting evidence of the impact of dispositional affectivity as it relates to many work attitudes, firms should seriously consider selection issues with regard to both positive and negative affectivity.

LIMITATIONS AND SUGGESTIONS FOR FUTURE RESEARCH

While the findings reported in this research provide strong indications that there exist significant differences in specific work attitudes and behaviors between blue and white-collar workers, limitations do exist. The first limitation is related to the fact that these results are based on a single company, a single group of workers, at a single point in time. Thus, the sampling frame limits the generalizability of these findings. Although a strength of the current study was the examination of many attitudinal and behavioral variables in a single sample, it also warrants replication. Second, the research is limited by the degree to which both the criterion variables and the independent variables are accurately measured.

These limitations provide potential avenues for future research. The first suggestion for subsequent research involves expanding the sample to include workers from other firms, industries and in other geographic regions. A related extension of the present research could entail a longitudinal study. This research would assess the stability of these relationships over time and could lead to a more concrete evaluation of the empirical relationships between these variables. A third area for future research might entail an evaluation of the measures used in the research. This research would then lead to an establishment of norms for the scales which could then be used in identifying employees with the most desirable work attitudes and behaviors.

REFERENCES

Agho, O.A., Mueller, C. W., & Price, J.L. (1992). Discriminant validity of measures of job satisfaction, positive affectivity, and negative affectivity. Journal of Occupational and Organizational Psychology, 65, 185-196.

Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents of affective, continuance, and normative commitment to the organization. Journal of Occupational Psychology, 63, 1-18.

Arvey, R. D., Bouchard, T. J., Jr., Segal, N. L., & Abramson, L. M. (1989). Job satisfaction: Environmental and genetic components. Journal of Applied Psychology, 74, 187-192.

Berry, D. & Hansen, J. (1996). Positive affect, negative affect, and social interaction. Journal of Personality & Social Psychology, 71(4), 796-809.

Chiu, R.K. & Francesco, A.M. (2003). Dispositional traits and turnover intention. International Journal of Manpower, 24(4), 284-298.

Chow, G. (1960). Tests of equality between sets of coefficients in two linear regressions. Econometrica, 28(3), 591- 605.

Cooper, R. & Payne, R. (1967). Extraversion and some aspects of work behavior. Personnel Psychology, 20, 45-57.

Cropanzano, R., & James, K. (1990). Some methodological considerations for the behavioral genetic analysis of work attitudes. Journal of Applied Psychology, 75, 433-439.

Cropanzano, R., James, K., & Konovsky, M. A. (1993). Dispositional affectivity as a predictor of work attitudes and job performance. Journal of Organizational Behavior, 14, 595-606.

Diener, E., & Emmons, R. A. (1985). The independence of positive and negative affect. Journal of Personality and Social Psychology, 47, 1105-1117.

Ferris, G. R., Youngblood, S. A. & Yates, V. L. (1985). Personality, training performance, and withdrawal: A test of the person-group fit hypothesis for organizational newcomers. Journal of Vocational Behavior, 27, 377-388.

Fisher, C.D. (2002). Antecedents and consequences of real-time affective reactions at work. Motivation and Emotion, 26(1), 3-30.

Fortunato, V. J., & Stone-Romero, E. F. (1999). Taking the strain out of negative affectivity: Development and initial validation of scores on a strain-free measure of negative affectivity. Educational and Psychological Measurement, 59, 77-97.

Fredrickson, B.L. (1998). What good are positive emotions? Review of General Psychology, 2, 300-319.

Fredrickson, B.L. (2001). The role of positive emotions in positive psychology: The broaden-and-build theory of positive emotions. American Psychologist, 56, 218-226.

Fredrickson, B.L., & Branigan, C.A. (2005). Positive emotions broaden the scope of attention and thought action repertoires. Cognition and Emotions, 19, 323-332.

Fredrickson, B.L. & Losada, M.F. (2005). Positive affect and the complex dynamics of human flourishing. American Psychologist, 60, 678-686.

Fredrickson, B., Tugade, M., Waugh, C. & Larkin, G. (2003). What good are positive emotions in crises? A prospective study of resilience and emotions following the terrorist attacks on the United States on the September 11th, 2001. Journal of Personality and Social Psychology, 84, 365-376.

Furnham, A. & Miller, T. (1997). Personality, absenteeism and productivity. Personality and Individual Differences, 23, 705-707.

George, J. M. (1992). The role of personality in organizational life: Issues and evidence. Journal of Management, 18, 185-213.

Hochwarter, W.A., Kiewitz, C., Castro, S.L., Perrewe, P.L. & Ferris, G.R. (2003). Positive affectivity and collective efficacy as moderators of the relationship between perceived politics and job satisfaction. Journal of Applied Social Psychology, 33(5), 1009-1035.

Hulin, C. & Blood, M. (1968). Job enlargement, individual differences, and worker responses. Psychological Bulletin, 69, 41-55.

Ironson, G. H., Smith, P., Brannick, M. T., Gibson, W. M., & Paul, K. B. (1989). Construction of a job in general scale: A comparison of global, composite, and specific measures. Journal of Applied Psychology, 74, 193-200.

Isen, A.M., Daubman, K.A., & Nowicki, G.P. (1987). Positive affect facilitates creative problem solving. Journal of Personality and Social Psychology, 52, 1122-1131.

Iverson, R.D., & Deery, S.J. (2001). Understanding the "personalogical" basis of employee withdrawal: The influence of affective disposition on employee tardiness, early departure, and absenteeism. Journal of Applied Psychology, 86 (5), 856-866.

Judge, T.A. (1993). Does affective disposition moderate the relationship between job satisfaction and voluntary turnover? Journal of Applied Psychology, 78, 395-401.

Levin, K., & Stokes, J. P. (1989). Dispositional approach to job satisfaction: Role of negative affectivity. Journal of Applied Psychology.

Meyer, J. P., & Allen, N. J. (1991). A three-component conceptualization of organizational commitment. Human Resource Management Review, 1, 61-98.

Mowday, R. T., Porter, L. W., & Steers, R. M. (1982). Employee-organization linkages: The psychology of commitment, absenteeism, and turnover. New York: Academic Press.

Newton, T., & Keenan, T. (1991). Further analyses of the dispositional argument in organizational behavior. Journal of Applied Psychology, 76, 781-787.

O'Reilly, C. A., III, Chatman, J., & Caldwell, D. F. (1991). People and organizational culture: A profile comparison approach to assessing person-organization fit. Academy of Management Journal^ 34, 487-516.

Price, J. (2001). Reflections on the determinants of voluntary turnover. International Journal of Manpower, 22(7), 600624.

Smith, P., Kendall, L. & Hulin, C. (1969). The measurement of satisfaction in work and retirement. Chicago: Rand.

Seligman, M. & Csikszentmihalyi, M. (Eds.) (2000). Positive psychology (Special Iss.). American Psychologist, 55, 1.

Staw, B. M., Bell, N. E., & Clausen, J. A. (1986). The dispositional approach to job attitudes: A lifetime longitudinal test. Administrative Science Quarterly, 31, 56-77.

Staw, B. M., & Ross, J. (1985). Stability in the midst of change: A dispositional approach to job attitudes. Journal of Applied Psychology, 70, 469-480.

Stone-Romero, E. F. (2005). Personality-based stigmas and unfair discrimination in work organizations. In R. L. Dipboye, & A. Colella (Eds.), Discrimination at work: The psychological and organizational bases (pp. 255280). Mahwah, NJ: Lawrence Erlbaum.

Watson, D. (1988a) Intraindividual and interindividual analyses of positive and negative affect: Their relation to health complaints, perceived stress, and daily activities. Journal of Personality and Social Psychology, 54, 1020-1030.

Watson, D. (1988b) The vicissitudes of mood measurement: Effects of varying descriptors, time frame, and response formats on measures of Positive and Negative Affect. Journal of Personality and Social Psychology, 55, 128141.

Watson, D., Clark, L. A., & Carey, G. (1988). Positive and negative affectivity and their relation to anxiety and depressive disorders. Journal of Abnormal Psychology, 97, 346-353.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063-1070.

Watson, D., & Tellegen, A. (1985). Toward a consensual structure of mood. Psychological Bulletin, 98, 219-235.

Elizabeth J. Rozell, Missouri State University

Charles E. Pettijohn, Nova Southeastern University

R. Stephen Parker, Missouri State University

Table 1: Demographic Attributes

Gender                              Frequency-Percentage

  Male                                   272-92.5%
  Female                                  22-7.5%

Skill Level

  High                                    94-32.9%
  Med-High                                69-24.1%
  Low                                     65-22.7%
  Shipping                                27-9.4%
  Maintenance                             31-10.8%

Education

  Less than High School                   46-16.0%
  High School                            132-46.0%
  Some College                            79-27.5%
  Associates                              11-3.8%
  Bachelor's                               5-1.7%
  Graduate                                 2-.7%
  Other                                   12-4.2%

Marital

  Single                                  50-17.2%
  Married                                197-67.9%
  Widowed                                  2-.7%
  Divorced                                41-14.1%

Average Years Worked in Company             11.3

Average Years Worked in Job                 7.2

Table 2: Relationship Between Positive/Negative Affect and the
Dependent Variables

                 Dependent        Independent
               Variable (R2)     Variable (B)    F-Value  Significance

White Collar    Commit (.21)    Positive (.46)    65.6      < .0001
Blue Collar     Commit (.25)    Positive (.51)    107.8     < .0001
White Collar   Turnover (.09)   Positive (-.31)   18.8      < .0001
Blue Collar    Turnover (.22)   Positive (-.47)   60.3      < .0001
White Collar  Job Satis. (.25)  Positive (.50)    80.0      < .0001
Blue Collar   Job Satis. (.23)  Positive (.48)    94.8      < .0001
White Collar   Absent (-.003)   Positive (.02)     .11        .74
Blue Collar    Absent (-.003)   Positive (.03)     .20        .65
White Collar    Tardy (.03)     Positive (-.17)    7.6        .007
Blue Collar    Tardy (-.003)    Positive (.02)     .1         .71
White Collar    Commit (.03)    Negative (-.17)    7.4        .007
Blue Collar     Commit (.01)    Negative (-.10)    3.7        .05
White Collar   Turnover (.09)   Negative (.31)    18.7      < .0001
Blue Collar    Turnover (.07)   Negative (.27)    15.9      < .0001
White Collar  Job Satis. (.17)  Negative (-.41)   49.7      < .0001
Blue Collar   Job Satis. (.15)  Negative (-.38)   55.3      < .0001
White Collar   Absent (-.004)   Negative (.008)    .01        .90
Blue Collar    Absent ( .010)   Negative (.10)    2.93        .09
White Collar    Tardy (.03)     Negative (.17)     6.9        .007
Blue Collar     Tardy (.02)     Negative (.15)     7.6        .006

Table 3: Comparisons of Models Using the Chow Test

                  Dependent         Positive        Negative
                Variable (R2)      Affect (p)      Affect (p)

Full Model       Commit (.26)      .63 (<.0001)   -.10 (.0843)
White Collar     Commit (.21)      .59 (<.0001)   -.07 (.4661)
Blue Collar      Commit (.26)      .60 (<.0001)   -.07 (.3492)
Chow Test
Full Model      Turnover (.23)    -.36 (<.0001)    .24 (<.0001)
White Collar    Turnover (.14)    -.22 (.0015)     .30 (.0010)
Blue Collar     Turnover (.26)    -.42 (<.0001)   .18 (.0120)
Chow Test
Full Model     Job Satis. (.37)    .55 (<.0001)   -.51 (<.0001)
White Collar   Job Satis. (.33)    .45 (<.0001)   -.43 (<.0001)
Blue Collar    Job Satis. (.34)    .56 (<.0001)   -.51 (<.0001)
Chow Test
Full Model       Absent (.01)      .01 (.7437)     .08 (.0222)
White Collar    Absent (-.01)      .01 (.7502)     .01 (.8206)
Blue Collar      Absent (.01)      .03 (.4568)     .10 (. 0728)
Chow Test
Full Model       Tardy (.02)       .00 (.8130)     .04 (.0012)
White Collar     Tardy (.04)      -.03 (.0343)     .03 (.0669)
Blue Collar      Tardy (.03)       .01 (.2873)     .04 (.0047)
Chow Test

               F-Value   Significance

Full Model      98.7       < .0001
White Collar    32.2       < .0001
Blue Collar     53.7       < .0001
Chow Test        1.8        .1461
Full Model      56.4       < .0001
White Collar    15.7       < .0001
Blue Collar     36.6       < .0001
Chow Test        2.4         .07
Full Model      162.2      < .0001
White Collar    59.2       < .0001
Blue Collar     80.5       < .0001
Chow Test        4.9        .0021
Full Model       2.7        .0711
White Collar     .06        .9419.
Blue Collar      1.7        .1791
Chow Test        3.1        .0282
Full Model       5.5        .0045
White Collar     5.7        .0040
Blue Collar      4.3        .0150
Chow Test        1.8        .1545

* regression coefficients are standardized