Sign up

Person-environment fit and self-employment: opportunities and needs for achievement, affiliation, autonomy, and dominance.
Abstract:
Survey data from a sample of US Certified Public Accountants (N = 322) was used to test hypotheses that psychological needs would differ among those who worked in three types of work arrangements: self-employed who owned enterprises with employees (n = 129), self-employed without employees (n = 98), and traditional employees (n = 95). I also tested hypotheses derived from person-environment theory that there would be differences among these work arrangements with respect to opportunities to satisfy those psychological needs. I used multivariate analysis to test for differences across groups with respect to perceived opportunities and needs for achievement, affiliation, autonomy, and dominance. I found statistically significant differences with respect to all four categories of opportunities and with respect to needs for autonomy and dominance. Further, the levels of perceived opportunities and needs were congruent (e.g., the self-employed who employed others reported greater opportunities for dominance than did the self-employed with no employees while exhibiting greater needs for dominance). The research supports the extension of person-environment fit to types of employment and the need to treat the self-employed as heterogeneous.

Article Type:
Report
Subject:
Person-environment fit (Research)
Author:
Prottas, David
Pub Date:
11/01/2011
Publication:
Name: North American Journal of Psychology Publisher: North American Journal of Psychology Audience: Academic Format: Magazine/Journal Subject: Education; Psychology and mental health Copyright: COPYRIGHT 2011 North American Journal of Psychology ISSN: 1527-7143
Issue:
Date: Nov, 2011 Source Volume: 13 Source Issue: 3
Topic:
Event Code: 310 Science & research
Geographic:
Geographic Scope: United States Geographic Code: 1USA United States
Accession Number:
274955415
Full Text:
Theories of person-environment fit have long been prevalent in management literature (Kristof, 1996; Kristof-Brown, Zimmerman, & Johnson, 2005; Schneider, 2001) and are used to explain how individuals' personalities and traits influence them to join and remain in organizations (Bernard, 1938; Schneider, 1987) and vocations (Dawis & Lofquist, 1984; Holland, 1985; Schein, 1978, 1994) and engage in entrepreneurial activities (McClelland, 1961; Schumpeter, 1934). Entrepreneurial trait researchers commonly use a design that compares a subgroup of study participants labeled as entrepreneurs to a subgroup of participants labeled as non-entrepreneurs with respect to specified personality characteristics. However, Gartner (1988) criticized prior trait research for using definitions of entrepreneurs that were vague or nonexistent and which varied across studies. This lack of definitional precision often led, in his view, to samples of entrepreneurs that were heterogeneous (for example, combining small business owners with people who serially created new organizations with high growth aspirations. Finally, he argued that researchers studied an excessive number of different traits and characteristics. Researchers have more recently adopted meta-analytical techniques and the Five Factor personality model as an organizing framework to quantitatively review multiple studies (Collins, Hanges, & Locke, 2004; Rauch & Frese; 2007; Stewart & Roth, 2001, 2007, Zhao & Seibert, 2006).

Subsequently, primary researchers, while accepting that there is no universally accepted narrow definition of an entrepreneur and have developed alternative typologies to distinguish among types of entrepreneurs based on their characteristics and objectives (Hisrich, Langan-Fox, & Grant, 2007). Researchers have also provided more detailed information on the criteria for classifying individuals into different subgroups and recognized that entrepreneurs are not a homogeneous. Researchers have adopted meta-analytical techniques and the Five Factor personality model as an organizing framework to quantitatively review primary research studies that employed different definitions of participants and assessed different traits (Collins et al, 2004; Rauch & Frese; 2007; Stewart & Roth, 2001, 2007, Zhao & Seibert, 2006)

Person-environment theories have provided a framework to study the interactions between individuals and foci of fit such as vocation, organization, group, and supervisor, and job (see Kristof, 1996; Kristof-Brown et al. 2005 for review of the conceptualizations and the results of prior empirical research). The different conceptualizations of person-environment fit share core assumptions: (i) work environments differ, (ii) individuals differ, and (iii) individuals tend to move toward environments which are congruent with the individual's needs, values, or capabilities. Person-environment fit theories focus on the individual as an active agent who enters and leaves work environments to achieve personal objectives.

Henry Murray (1938) was one of the early pioneers of person-environment, proposing that individuals had distinct psychological traits ("needs") that they would attempt to satisfy by finding opportunities ("supplies"), at work and elsewhere, to engage in certain behaviors. The needs-supply conceptualization of person-environment fit undergirds much of the trait oriented entrepreneurial research. Under that conceptualization, fit is achieved when there is congruence between what the person needs, desires, or prefers (material or psychological) and what is provided by the work environment. For example, an individual with a high need for dominance would likely enter an organization or a profession where he or she would be able to direct and manage other people and act as a leader and avoid those environments where positions of power and authority were less available. Trait oriented entrepreneurial research is based on the assumption that entrepreneurship, however defined, provides opportunities for behaviors ("supplies") that are different from the behaviors in non-entrepreneurial work setting. Accordingly, individuals with certain psychological needs would be expected to become entrepreneurs as they would find the requirements of entrepreneurship as needs satisfying.

For example, Stewart and Roth (2007) recognized that entrepreneurial research on achievement motivation was grounded on Murray's (1938) needs theory as further developed by McClelland's (1961) need for achievement and Miner's (1993) task motivation theories. McClelland and Miner argued that entrepreneurial careers provided greater opportunities, relative to traditional employment, for individuals to achieve success through their individual efforts. Accordingly, individuals with higher needs for achievement would be more likely to embark upon entrepreneurial careers because entrepreneurship would offer greater opportunities to satisfy their needs for achievement. Similarly, Stewart and Roth (2001) indicated in their meta-analysis of risk propensity that researchers posited that entrepreneurial activity was, relative to managerial activity, less structured and more uncertain with the entrepreneur bearing more responsibility for the consequences of his or her decisions. Consequently, more risk-tolerant individuals would 'self-select' into entrepreneurial careers whereas more risk-averse individuals would opt for traditional employment.

However, if there are different types of entrepreneurs, and if entrepreneurship is not coterminous with self-employment, then the environments would differ between types, provide different supplies, and attract individuals with different needs. Researchers have recognized the need to distinguish between individuals who serially create organizations, markets and products and owners of small restaurant or stores. For example, Stewart and Roth (2001, 2007) distinguished between entrepreneurs who were growth oriented versus those that were income oriented with the assumption that the different assumptions lead them to confront different environments which would require different behaviors. For example, the income oriented entrepreneur would not need to engage in unusually risky activities and would not need to have an unusually high propensity for risk whereas someone with a growth orientation would likely need to taker higher levels of risk. Similarly, Schein (1978, 1994) distinguished between entrepreneurs who have creativity career anchors and need to create and build enterprises in order to be content and those who had autonomy/independence anchors and would intentionally restrict the growth of their businesses to retain the independency that satisfied their need. The individual with the creativity career anchor would need to hire and direct employees in order to meet his or her objective of growing the enterprise whereas the individual with the autonomy/independence anchor would not need to.

In his research I identified three different types of self- and organizational employment (which I refer to as "work arrangements") that I expected would require and permit individuals working in them to engage in different types of behaviors. I assessed the perceptions of individuals working in each of these categories as to the amount of opportunities for four types of behaviors ("supplies") and tested for hypothesized differences. I also assessed four corresponding personality characteristics of the individuals working in each category ("needs") and tested for hypothesized differences.

I distinguished among three types of work arrangements: the self-employed who had ownership interests in enterprises that employed others ("employers"), the self-employed who owned businesses with no employees ("non-employers"), and those who worked for others ("employees"). This bifurcation of the self-employed is consistent with the distinction made by others (Barbato, DeMartino, & Jacques, 2009; Bond, Thompson, Galinsky, & Prottas, 2003; Hipple, 2004; Prottas, 2007; Prottas & Thompson, 2006) and recognizes that non-employers represent the dominant form of self-employment. According to the Bureau of Labor standards, there were 15.9 million self-employed individuals in 2008, representing 10.9% of the employed (Hipple, personal communication, 5.28.09). Prior BLS studies reported that only a small minority (16.9% in 2003) of the self-employed owned businesses that employed others and three-quarters of that minority had only one to four employees (Hipple, 2004).

Moreover, the distinction between employers and non-employers appears crucial with respect to understanding how the behaviors required in each work arrangement could lead individuals with different personalities to be attracted to them. I believe a key distinction between the work environments of employers and non-employers is that employers and employees are members of organizations which provide opportunities for different types of social interactions. Indeed, Barnard (1938) suggested that the "inducements of a personal, non-materialistic character are of great importance to secure cooperative effort above the minimum material reward essential to subsistence. The opportunities for distinction, prestige, personal power, and attainment of dominating position are much more important in the development of all sorts of organizations, including commercial organizations" (p.145). The non-employer works outside of such an organization and would have fewer opportunities and requirements for such social interactions. I developed and tested hypotheses about differences among the three groups with respect to opportunities and needs for achievement, affiliation, autonomy, and dominance.

One distinction between the self-employed and traditional employees is that the latter have "maintenance free" employment with others responsible for maintaining the business, leaving the employee responsible for a narrow range of responsibilities (Eden, 1975; Katz, 1994). In contrast, the self-employed have a broad range organizational maintenance responsibilities such as maintaining payroll, paying employee-related taxes, hiring, disciplining, firing, negotiating remuneration with suppliers, customers, etc. (Eden, 1975; Katz, 1994). Nonemployers have a broader range of responsibilities and areas of decision than do employees, such bookkeeping, financial management and control, setting up and maintaining physical work space, legal, negotiating with suppliers and customers, and obtaining health and business insurance. Both categories of self-employed must also obtain and serve their customers. The characteristics of self-employment, for both employers and nonemployers, require them to engage in a wider variety of behaviors would enable them to satisfy needs for achievement. Employers and nonemployers are more likely than employees to achieve success primarily through their individual efforts and to more directly experience the consequences of their own decisions.

While there is little research on opportunities for achievement among entrepreneurs, there has been considerable empirical research on their need for achievement. Empirical support for the relationship between need for achievement and entrepreneurial activity has been found in meta-analytic reviews (Collins et al., 2004; Rausch & Frese, 2007; Stewart & Roth, 2007; Zhao & Seibert, 2006). In the one study comparing needs for achievement among employers, nonemployers and employees, Barbato et al. (2009) found employers had greater needs for achievement than both nonemployers and employees with nonemployers and employees having similar needs.

Therefore, I hypothesized that:

Hypothesis 1: There will be differences across groups with respect to perceived opportunities for achievement within their work settings. Specifically, employers and nonemployers will report greater opportunities for achievement than employees.

Hypothesis 2: There will be differences across groups with respect to needs for achievement. Specifically, employers and nonemployers will have greater needs for achievement than employees.

Differences in Opportunities and Needs for Affiliation

One difference between employers and nonemployers is that the former are members of the social systems constituted by their businesses whereas nonemployers are not. Barnard (1938) noted that joining a work-related organization necessitated membership in a social system and that some individuals would find such inclusion in a social system to be onerous and repellent rather than rewarding and attractive. Employers and employees should experience more opportunities--and requirements --to socially interact and affiliate than nonemployers. Of course, there is likely to be substantial variation within each work arrangement. An employee working at a location which is physically remote from coworkers might have little social interaction with fellow employees. On the other hand, a nonemployer may have long lasting relationships with a customer and work onsite, thusly becoming integrated with a social system.

There is limited empirical evidence that opportunities to engage in social interaction and a desire to do so may vary across work arrangements Bond et al. (2003) found that nonemployers reported less social and work related support from coworkers than did employers and employees (and employers reported more than employees). There is more research on a related personality trait, extraversion, that suggests it might be related to entrepreneurship. Extraversion is related to both leadership ability (Judge Bono, Ilies, & Gerhardt, 2002) and management and sales ability (Barrick & Mount, 1991) and it has been hypothesized that individuals who were more extraverted should develop the broad social networks that facilitate self-employment (Allen, 2000). However, in their meta-analysis, Zhao and Seibert (2006) found no relationship between extraversion and entrepreneurial activity. I hypothesized that:

Hypothesis 3: There will be differences across groups with respect to opportunities for affiliation with others within their working setting. Specifically, employers and employees will report greater opportunities for affiliation than nonemployers.

Hypothesis 4: There will be differences across groups with respect to needs for affiliation. Specifically, employers and employees will have greater needs for affiliation than non-employers.

Both employers and non-employers are responsible for directing and managing their own work while in general most employees are managed, to a degree, by others in their organization. Self-employment, whether as employer or non-employer, is likely to provide more opportunity for autonomy than is working as an employee. There is research with respect to both opportunities for autonomy and personality traits. Studies found that the self-employed report greater job autonomy (Eden, 1975; Hundley, 2001; Tetrick et al., 2000). Furthermore, a desire for autonomy is related to preference for self-employment (Buttner & Moore, 1997; Kolvereid, 1996; Peel & Inkson, 2001). Danziger and Valency (2006) found the incidence of autonomy and entrepreneurial creativity was greater among the self-employed in their sample of Israelis than it was among employees and Feldman and Bolino (2000) found that the autonomy and independence anchor was the most common among the self-employed. While they found no difference between self-employed and employees with respect to need for autonomy, Hornaday and Aboud (1971) reported the self-employed placed a higher value on independence. In their meta-analysis, Rausch and Frese (2007) found that need for autonomy was positively related to being an entrepreneur rather than non-entrepreneur and measures of organizational performance. Two studies with the self-employed classified as either employers or nonemployers found that employers reported greater job autonomy than nonemployers who in turn reported more than employees (Prottas, 2007; Prottas & Thompson, 2006).

There is limited research comparing needs for autonomy between the two types of self-employed. Schein's theory of career anchors (1978, 1994) holds that the self-employed are heterogeneous; some, such as independent contractors, have autonomy anchors while others, who create organizations, have creativity or entrepreneurial anchors. Studies show that constructs related to need for autonomy were in fact negatively related to the size of the businesses of self-employed (Lee & Tsang, 2001; Robichaud et al., 2001). Barbato et al. (2009) reported that both employers and nonemployers had greater needs for autonomy than employees and employers had greater needs than nonemployers.

With respect to autonomy, I expect that both categories of self-employed would have greater opportunities and needs for autonomy than will employees. However, I expect that although employers will have greater opportunities than nonemployers given their ability to delegate work to others, I expect nonemployers will have a greater needs for autonomy. I predicted that:

Hypotheses 5: There will be differences across groups with respect to opportunities for autonomy within their work setting. Specifically, employers and nonemployers will report greater opportunities for autonomy than employees, and employers will report greater opportunities than nonemployers.

Hypotheses 6: There will be differences across groups with respect to needs for autonomy. Specifically, employers and nonemployers will have greater needs for autonomy than employees and nonemployers will have greater needs than employers.

Employers and employees who are managers must direct and influence the behavior of other organizational members. As Katz (1994) stated, "self-employment by definition places the individual on the top of the firm's hierarchy" and he distinguished between the owners of firms that employ others and those who don't: "The owner of a one-person firm would be seen as having power over no one. Hence, the employer of a one person firm has less hierarchy than the employer of a 100-employee firm." (p. 26). The two types of self-employed differ with respect to whether the individual manages and directs the activities of others. Employers must perform such management functions while the typical nonemployer has no such responsibilities or opportunities. The amount of responsibility and ability of an employee to direct the work of others varies, of course, with his or her level in both formal and informal organizational structures. However, even the newest hire without any form of formal or informal responsibility or power is entering a system with the potential for gaining such organizational influence.

There is limited evidence that the self-employed and employees differ with respect to dominance. Kolvereid (1996) found students who preferred to be self-employed valued authority more highly than those who wanted to be employees. Hornaday and Aboud (1971) found entrepreneurs placed greater value on leadership than employees while DeCarlo and Lyons (1979) found them higher in dominance. Baum et al. (1993) found U.S. entrepreneurs were higher than U.S. managers in need for dominance.

Accordingly I predicted that:

Hypotheses 7: There will be differences across groups with respect to opportunities for dominance within their work setting. Specifically, employers and employees will report greater opportunities for dominance than nonemployers and employers will report greater opportunities than employees.

Hypotheses 8: There will be differences across groups with respect to needs for dominance. Specifically, employers and employees will have greater needs for dominance than employees and employers will have greater needs than employees.

METHOD

Sample

Participants were 322 Certified Public Accountants (CPAs) who were members of the New York State Society of CPAs. A sampling frame of 1,200 members was created from the publicly accessible membership directory on its website (http://www.nysscpa.org). Most entries included a definition of the member's position (such as sole practitioner, CFO, partner/principal/shareholder); these were used to provide preliminary classification into the categories of work arrangements for the purpose of recruiting appropriate numbers of participants in each group. As my intention was to produce groups of similar size for comparison purposes, I over-sampled nonemployers and employers relative to their percentages in the general membership. I systematically reviewed the membership lists by counties and used a quota system to assemble a mailing list of 1,200 members, broken down by preliminary classification as 500 sole practitioners, 500 as partners in public accountancy firms, and 200 as traditional employees. Based on first names, 78% were male, the same proportion of males in the general membership (W. A. Pape, personal communication, December 6, 2005). Email pre-advices were sent to potential participants followed by mailings containing cover letters, surveys, and stamped return envelopes. Email follow ups were sent several months later. Participation was anonymous (participants could enclose a business card to be advised of any subsequent publications; cards were immediately separated from the surveys). Fourteen of the mailings were returned as undeliverable; about 2% of the emails sent on each emailing were returned due to faulty email addresses. The 322 responses represented a response rate of 27% (comparable to similar survey studies of members of accounting societies (Fogarty et al., 2000).

Measures

Work Arrangement. Participants classified themselves by checking one of seven categories: "sole practitioner (with no full-time employees other than yourself)", "sole practitioner (with one or more non-CPA employees)", "partner (with at least one-third partnership interest) in a public accountancy firm"; "partner (with less than one-third partnership interest) in a public accountancy firm", "employee of a public accountancy firm (non-partner)", "employee of other than a public accountancy firm," "other." Participants were asked how many other people worked in their organization. Sole practitioners without employees were classified as nonemployers. Sole practitioners with employees and partners with at least one-third partnership interest were classified as employers. Seven individuals who chose "other" provided sufficient information to be classified as employers. Partners with less than one-third partnership interests and employees of public accountancy and other organizations were classified as employees. As some large public accountancy firms have thousands of "partners" for purposes of this study I felt it necessary to distinguish between partners whose ownership interests were sufficiently large to provide them with rights and responsibilities that are more generally associated with ownership. Participants were also asked to indicate "given a choice, under which arrangement would you most prefer to work?" using those same categories.

Demographics. Items assessed age, sex, marital status, and individual income.

Opportunities. Medcof and Hausdorf (1995) developed three scales intended to be facilitate substantive research on person-environment fit by assessing opportunities at work to satisfy three psychological needs: for achievement (five items, Cronbach alpha coefficient of internal reliability ([alpha]) = 79), for dominance (four items, [alpha] = .84), and for affiliation (three items, [alpha] = .72). Medcof and Hausdorf (1995) critiqued prior person-environment research for assessing the psychological traits of individuals in certain jobs or vocations and then making the assumption that opportunities varied accordingly without assessing the opportunities themselves. They developed the measures of opportunities using Jackson's (1989) definitions as a conceptual guide. I used all of the items from Medcof and Hausdorf (1995) in this study. As I was concerned with having only three affiliation items and were concerned about the reliability of the measure, I added one additional item from Sims et al (1976). Sample items are "On this job, I work towards clear challenging goals" (achievement); "To do my job properly, I have to spend quite a bit of time influencing others" (dominance); "On this job, I spend a great deal of time with other people" (affiliation). Job autonomy was measured using all four items ([alpha] = .74) that Beehr (1976) used to assess perceived autonomy. A sample item is "I have a lot to say over what happens on my job." Respondents were asked to indicate the extent to which each items accurately or inaccurately described their current job with using a Likert-type scale with options from 1 (=Very inaccurate) to 7 (=Very accurate). The [alpha]'s in the current study were: achievement, .78; affiliation, .67; autonomy, .83; and, dominance, .86. The [alpha] would have been somewhat lower had I not added a fourth item to the affiliation scale.

Needs. Needs for achievement, affiliation, autonomy, and dominance were assessed by four scales consisting of sixteen items each from the Jackson Personality Research Form E (PRF) (Jackson, 1999). Jackson used Murray's (1938) work as a starting point to develop measures of 23 different psychological needs. Jackson's (1999) description of high scorers on each scale and sample items are: achievement: "aspires to accomplish difficult tasks, maintains high standards and is willing to work toward distant goals, responds positively to competition, willing to put forth effort to attain excellence" (p. 5) with a sample item being, "I often set goals that are very difficult to reach"; affiliation: "enjoys being with friends and people in general; accepts people readily; makes efforts to win friendships and maintain associations with people" (p. 5) and a sample item being "Often I would rather be alone than in a group of friends"; autonomy: "tries to break away from restraints, confinement, or restrictions of any kind; enjoys being unattached, free, not tied to people, places, or obligations; may be rebellious when faced with constraints" (p. 5) with a sample item being "My greatest desire is to be independent and free"); "attempts to control environment, and to influence or direct other people; expresses opinions forcefully; enjoys the role of leader and may assume it spontaneously" (p. 6) with a sample item being "In an argument, I can usually win others over to my side." Response options were True or False. The [alpha]'s for these scales in this study were: achievement .59; affiliation, .82; autonomy, .66; and dominance, .80. These compare to the internal reliabilities reported by Jackson (1999) in two samples: achievement, .66/.57, affiliation, .82/.86; autonomy, .61/.66, and dominance, .78/.67) and by Medcof and Hausdorf (1996: achievement, .57; affiliation, .86, and dominance, .67).

RESULTS

Table 1 provides basic descriptive statistics and correlations for the 322 participants. The majority were male (74.5%) and married or in a similar relationship (82.8 %). The average age was 52.9 years (SD = 11.69) and the median individual income was $135,000. Participants were classified as nonemployers (n = 98, or 30.4%), employers (n = 129, or 40.1%), or employees (n = 95, or 29.5%).There were significant relationships between age and opportunities for affiliation (-.12, p = .030) and autonomy (.22, p = .000) and needs for affiliation (-.14, p = .015) and dominance (-.15, p = .007). Gender was related to opportunity for dominance (r with female = -.11, p = .041. Consistent with other studies (e.g. Prottas & Thompson, 2006), the self-employed in this sample tended to be older and were more likely to be male than employees.

The correlations among the measures of needs and opportunities were similar in magnitude and direction to prior studies. The relationships among the four measures of needs were similar to those reported for male college students by Jackson (1999) and by Medcof and House (1995). The correlation between achievement and affiliation in my study was .13 versus Jackson's .01 and Medcof and Hausdorf's .15; achievement and dominance of .33 versus .26 and .44; between affiliation and dominance of .37 versus .30 and .18. Medcof and House did not assess needs or opportunities for autonomy. The correlation in my study between autonomy and achievement was -.04 versus Jackson's .03; autonomy and affiliation of -.34 versus -.29; and autonomy and dominance of -.14 versus -.01. My correlations between opportunities were similar to those reported by Medcof and Hausdorf (1995): achievement and. affiliation of .37 in this study versus .44; achievement and dominance of .60 versus .41, and affiliation and dominance of .35 versus .44.

The four opportunities for needs-related behaviors were conceptually linked and likely to be inter-correlated (as were the four needs). Therefore, I first conducted two sets of MANOVAs with work arrangement as a non-ordered categorical variable (with three values: employer, non-employer, and employee). I examined multivariate and univariate F statistics based on the Wilks lambda for statistical significance and used the partial eta-squared ([[eta].sub.p.sup.2]) as the effect size.

Given the differences across work arrangements with respect to age and sex, I also conducted two sets of MANCOVAs with work arrangement as a fixed factor and age and sex as covariates (Model 2 in Table 2). Finally, I reclassified participants according to their preferred work arrangement and conducted another set of MANOVAs (Model 3 in Table 2). For the hypotheses regarding between-group relationships, the standardized mean difference (d) between each group was calculated and the statistical significance of the differences were determined using Bonferroni corrections. Based on Cohen (1977) and Stevens (2002), effect sizes for [[eta].sub.p.sup.2] were categorized as follows: .01 is small, .06 is medium, and .14 is large. For the standardized mean differences: d, .20 is small, .50 is medium, and .80 is large (Cohen, 1977).

With respect to opportunities for achievement, affiliation, autonomy and dominance as dependent variables, the multivariate F statistic based on the Wilks lambda for work arrangement as the fixed factor was significant (F (8,628) = 22.96, [[eta].sub.p.sup.2] = .23, p = .000). As evidenced by the univariate F statistics, there were significant differences across groups as predicted in hypotheses 1, 3, 5, and 7: for achievement (F = 3.45, [[eta].sub.p.sup.2] = .02, p = .033); for affiliation (F = 14.69, np2 = .09, p = .000); for autonomy (F = 35.84, np2 = .18, p = .000); for dominance (F = 46.83, [[eta].sub.p.sup.2] = .23, p = .00.

With respect to needs for achievement, affiliation, autonomy, and dominance, the multivariate F statistic based on the Wilks lambda for work arrangement was also significant (F (8,622) = 3.82, [[eta].sub.p.sup.2] = .05, p = .000). Further, the univariate F statistics for work arrangement were significant for needs for autonomy (F = 10.00, [[eta].sub.p.sup.2] = .06, p = .000) and for dominance (F = 5.38, [[eta].sub.p.sup.2] = .01, p = .005) as predicted in hypotheses 3 and 4 but not for achievement (F = 1.48, [[eta].sub.p.sup.2] = 01, p = .23) or for affiliation (F = 1.53, [[eta].sub.p.sup.2] = 01, p = .22) as predicted in hypotheses 1 and 2. The multivariate and univariate effects sizes of the differences across groups with respect to needs were smaller than the differences with respect to opportunities.

The means and standard deviations for participants in each work arrangement, the standardized mean differences of each paired-comparison, and the statistical significance of each paired difference (with Bonferroni correction) are shown in Table 3. The results with respect to the opportunities for each type of behavior are shown on the left side of the table and with respect to each type of need on the right side.

With respect to opportunities for achievement, employers reported significantly greater opportunities than employees. Although the mean of opportunities of non-employers was greater than those of employees with an effect size that reached the threshold for small but the difference was not statistically significant (p = .40). Similarly with respect to needs, both employers and employees reported had higher means than non-employers with effect sizes that reached the threshold for small but the differences were not statistically significant (p =.44 and p = .37). The needs of employers and employees were virtually identical.

As hypothesized, both employers and employees reported greater opportunities for affiliation than did non-employers (d = .66, p = .000 and d = .57, p = .000). However, my hypotheses that employers and employees would have greater needs for affiliation were not supported; employers and employees reported higher mean needs than higher needs than non-employers but the differences were not statistically significant (p = .25 and p = .85, respectively). Employers and employees were virtually identical.

With respect to opportunities for autonomy, employers and nonemployers reported significantly greater opportunities than employees (d = .93, p = .000 and d = .95, p = .000) with little difference between employers and nonemployers. With respect to needs, nonemployers had greater needs than employees (d = .62, p = .000) and employers (d = .41, p = .013). Employers had higher means than employees with a difference that exceeded the threshold for small but which did not achieve statistical significance (p = .30).

With respect to opportunities for dominance, both employers and employees reported significantly greater opportunities than did nonemployers (d = 1.18, p = .000 and d = 1.13, p = .000). However, my hypothesis that employers would report greater opportunities for dominance than employees was not supported as they reported virtually identical opportunities. The relationships were similar with respect to needs as both employers and employees had greater needs than nonemployers (d = .36, p = .02 and d = .44, p = .003). As with opportunities, my hypothesis that employers would have greater needs for dominance than employees was not supported as they had very similar needs.

Given the significant age and sex differences across groups and the bivariate relationships between gender and age and some of the measures of opportunities and needs, I conducted two sets of MANCOVAs with work arrangement as the fixed factor and age and sex as covariates. The multivariate statistics for opportunities (F = 18.75, p = .00, [[eta].sub.p.sup.2] = .20) and needs (F = 3.69, p = .00, [[eta].sub.p.sup.2] = .05) were similar to those of the MANOVAs (the univariate relations are shown in Model 2 of Table 2). That is, the multivariate [[eta].sub.p.sup.2] exceeded the threshold of large for opportunities and was slightly below the threshold for medium for needs. The patterns were similar with respect to the univariate relationships with significant differences with respect to opportunities for achievement, affiliation, autonomy, and dominance and needs for autonomy and dominance. The differences between groups were also similar to the MANOVA results.

I examined the possibility that the magnitude of the differences of personality variables across groups was attenuated by constraints on individuals that prevented them from moving into their desired work arrangement. The vast great majority of employers (92%) preferred that arrangement with only 4% wanting to be nonemployers or employees. The majority (59%) of nonemployers also indicated they preferred that arrangement; with 35% expressing a preference to be employers, and only 6% wanting to be employees The majority of employees (73%) preferred their arrangement with 25% wanting to be employers and only 1% wanting to be non-employers. This is consistent with the 2002 NSCW (Bond et al, 2003) which reported that 12% of employers and 32% of non-employers said they would prefer to be working for someone else and with 26% of employees saying they planned to be self-employed at some point in their future. I conducted a second set of MANOVAs with individuals classified according to their preferred work arrangement. As discussed above, this resulted primarily in actual non-employers and employees being reclassified as employers. The multivariate statistics for opportunities were F = 9.89, p = .000, [[eta].sub.p.sup.2] = .11 and for needs, F = 5.26, p = .000, [[eta].sub.p.sup.2] = .06 (the univariate relationships are shown in Model 3 of Table 2). As would be expected, the effect sizes of work arrangement with respect to opportunities were smaller than those found in Model 1. However, with respect to needs, the multivariate and univariate effect sizes were slightly greater (multivariate [[eta].sub.p.sup.2] = 06 vs. .05 and univariate [[eta].sub.p.sup.2]s for achievement of .02 vs. .01, affiliation of .01 vs. .01, autonomy, .09 vs. 06, and dominance, .04 vs. .03). The between group relationships were similar to those from the MANOVA on individuals classified according to their actual status.

DISCUSSION

My research provided support for extending the theory of person-environment fit to choice of work arrangement. Individuals who worked as members of organizations (as employers or employees) reported greater opportunities for affiliation and dominance and had personalities with higher needs for affiliation and dominance. Those who worked for and by themselves as nonemployers reported relatively lower opportunities and needs. Individuals who were self-employed reported greater opportunities for autonomy than those who worked for others as employees. Consistent with Schein's (1978, 1994) distinction between autonomy/independence and creativity anchors, it was the nonemployers who had the highest needs for autonomy. Differences were less pronounced and consistent with respect to achievement. Employers and nonemployers reported relatively high levels of both opportunities for achievement, consistent with the prior research on relationships between entrepreneurial activity and entrepreneurship

I did not find that employers and employees differed as I had expected with respect to opportunities and needs for dominance. This might be idiosyncratic to the profession as there is evidence that the CPA credential enables some individuals to obtain relatively high levels of authority --and remuneration-- in large organizations. The mean individual income of employees was highest (M = $222,095, SD = $256,963) than employers (M = $193,829, SD = $152,174) and independents (M = $117,812, SD = $78,808). However, a few outliers (CFOs of very large organizations) skewed the employee mean. The median incomes of employees ($147,500) and owners ($150,000) were similar and greater than the median for nonemployers ($100,000). Additionally, the majority (65%) of employees had titles that suggested they would have managerial responsibilities (e.g., partners in public accounting firms and directors, vice presidents, senior vice presidents, CFOs of other types of firms) and worked in large firms (the mean number of fellow employees was 7,000, SD = 2,917.9, Mdn = 50) whereas the employers managed relatively small firms (with the mean number of employees of 7.3, SD = 13.3, Mdn = 3.0). Consistent with the BLS findings (Hipple, 2004), the majority of the employers had relatively few employees: 56% had four or fewer and the median was 3.0. That employees in this study earned more than employers contrasts with the 2002 National Study of the Changing Workforce (Bond et al, 2003). In that cross-industry national sample, only 35% of employees classified themselves as professionals or managers (with the latter category including fast food restaurant managers as well as CEOs). Employers in that sample reported higher income than nonemployers and employees. In a sample that is more representative of the general population, the average employee would have less hierarchical and professional authority than did the CPAs in this sample and correspondingly report lower opportunities and needs for dominance.

This research contributes to the person-environment fit literature in several ways. First, it supports the extension of the conceptualization of person-environment fit to work arrangement in addition to the traditional foci of organizations and vocations. Second, it simultaneously assesses both work characteristics and personality traits rather than only one or the other as has more common in person-environment fit research (Medcof & Hausdorf, 1995). The magnitude of differences in needs across work arrangements was similar to those found in other studies. For example, meta-analyses that compare entrepreneurs to managers report small-to-medium differences for neuroticism, openness to experience, and conscientiousness (Zhao & Seibert, 2006) and need for achievement (Collin et al., 2003; Stewart & Roth, 2007).

Furthermore, consistent with a causal model of person-environment fit, in which situational and personal constraints moderate the degree to which individuals can migrate into the work arrangement which best fits their personality needs, I found that there were greater differences across work arrangements with respect to perceived opportunities to fulfill needs than there were with respect to the personality traits. This is consistent with other research. For example, Kristof-Brown et al. (1996) meta-analysis found the relationships between degree of fit and attitudes (such as job satisfaction) were stronger than those between fit and behaviors (such as turnover). Medcof and Hausdorf (1995) also found that the differences across occupations were larger for opportunities for behaviors than they were for the corresponding needs. Individuals do not make their career decisions purely to satisfy their psychological needs and the work place is not the only setting where needs may be satisfied. For instance, individuals with needs for achievement that are not being met through their work may engage in leisure activities that allow greater, compensatory opportunities (Adams & Stone, 1977).

While my results are consistent with a model showing that individuals with certain needs migrate into needs-satisfying environments, the cross-sectional design of my study does not allow us to discount alternative explanations regarding causality. It is possible that the behavioral demands of the different work arrangements affect, over time, the personalities of incumbents (Davis-Blake & Pfeffer, 1989). Future research that is longitudinal and identifies processes by which individuals form preferences and attain different work arrangements is necessary.

As with much person-environment fit research (Kristof-Brown et al., 2005), this research is limited by reliance of self-reported data for both job characteristics and personality needs. Kristof-Brown et al. (2005) suggested that their finding that the effect sizes of studies with all common sources were larger than those that used partially common raters might be explained by dissonance and self-perception effects that lead to cognitive distortions. However, the greatest cognitive distortion would be most likely to occur when asking individuals to report on their perception of fit and their attitudes towards their environment which I did not ask my participants to do. Additionally, my measures of perceived opportunities and needs were quite different. The wording of the opportunity items was tied to the workplace whereas the personality items were related to a broad variety of life settings. They differed as well in response options. Therefore, it would seem less likely that individuals' responses to the items would be distorted by a desire to be consistent.

The relatively low internal reliabilities of several of my measures of key constructs should also be viewed as a limitation on my research. The Cronbach alpha coefficients for opportunity for affiliation ([alpha] = .67) and need for autonomy ([alpha] = .66) fell somewhat below Nunnally (1967's) oft cited threshold of .70 for acceptability while that for need for achievement ([alpha] = .59) was well below. It is possible that my failure to find some significant differences among groups with respect to certain variables was attributable to the relatively low reliability of some of my measures. However, if the common source bias augmented the strength of the relationships, the within profession design may have attenuated them. Person-vocation fit suggests that individuals who become CPAs differ with respect to their personalities from others and that the characteristics of their work environments would also differ. Range restriction with respect to both opportunities for behaviors and needs may have resulted in differences across work arrangements being less pronounced than they would be in an across-industry sample.

Future research should study individuals in professions that are dissimilar (such as employer/operators of long-haul trucks) to determine whether the relationships found here hold more broadly. The research is further limited by the relatively low internal consistencies reported for several of measures, most, notably the need for achievement. Low reliabilities may attenuate relations but also cast doubt whether the underlying construct is being assessed with sufficient accuracy for any reliance to be placed on the findings. Certainly, any further research should be undertaken with instruments that may reliably assess the construct in the sample population.

This research has practical as well as theoretical implications. Self-management of one's own career is enhanced by individuals gaining awareness of their own values and needs in order to initiate actions to fulfill them (Schein, 1978). After becoming aware of their own needs, individuals may then seek out information as to how different career choices might satisfy them. Given the increasing availability of alternative work arrangements and alternatives for self-employment, it would be important to understand that they may provide very distinct work environments. These findings may then be relevant to individuals making career choices as well as to those who provide vocational counseling. For example, it appears that a person may have ample opportunities for autonomy as either an employer or a nonemployer, but a person with high needs for dominance be would be dissatisfied being an nonemployer whereas an individual with low needs may be unhappy being an employer.

My research may also have implications for organizations. As proposed by Hackman and Oldham (1975) and confirmed by numerous studies (Kinicki, McKee-Ryan, Schriesheim, & Carson, 2002), job autonomy is positively related to job satisfaction and other outcomes. Self-employment, as either employer or nonemployer appears to be an effective tactic to obtain more autonomy. As alternative employment arrangements become more accessible to more workers, a given employer may need to be less concerned with its best and brightest employees defecting to a competing organization and more worried about them departing to work for themselves. Employers may need to redesign jobs to increase autonomy in order to compete with the allure of alternative arrangements.

Additionally, organizations that rely on the services of non-employers rather than employees may need to adjust their practices and policies to attract and motivate them. Management practices that are suitable for typical employees may not be for non-employers. For example, to the extent that non-employers have lower needs for dominance, offering them managerial responsibilities on either an on-going or project basis might be non-motivating or even de-motivating to non-employers though motivating to employees. Similarly, supervisory practices that are acceptable to traditional employees may be alienating to non-employers who are sensitive to violations of their autonomy. Organizations may need to design less intrusive forms of work supervision or more carefully select non-employers who have the ability to work effectively with less supervision. Additionally, as non-employers have needs for affiliation that are similar to employees yet typically operate in an environment with fewer opportunities to satisfy those needs, organizations which depend on non-employers might provide the option of working on-site and otherwise attempt to integrate them into the social fabric of the organization.

REFERENCES

Adams, A., & Stone, T. H. (1977). Satisfaction of need for achievement in work and leisure time activities. Journal of Vocational Behavior, 11, 174-181.

Allen, W. D. (2000). Social networks and self-employment. Journal of Socio-Economics, 29, 487-501.

Barbato, R., DeMartino, R., & Jacques, P. H. (2009). The entrepreneurial motivations of nonemployer entrepreneurs. New England Journal of Entrepreneurship, 12, 33-43.

Barnard, C. I. (1938). The functions of the executive. Cambridge, MA: Harvard University Press.

Barrick, M.R., & Mount, M. K. (1991). The big five personality dimensions and job performance: A meta-analysis. Personnel Psychology, 44, 1-26.

Baum, J. R., Olian, J. D., Erez, M., Schnell, E. R., Smith, K. G., Sims, H. P., Schully, J. S., & Smith, K. A. (1993). Nationality and work role interactions: A cultural contrast of Israeli and U.S. entrepreneurs' versus managers' needs. Journal of Business Venturing, 8, 499-512.

Beehr, T. A. (1976). Perceived situational moderators of the relationship between subjective role ambiguity and role strain. Journal of Applied Psychology, 61, 35-40.

Bond, J. T., Thompson, C., Galinsky, E., & Prottas, D. (2003). Highlights of the National Study of the Changing Workforce, No. 3, 2002. New York: Families and Work Institute.

Bureau of Labor Statistics. (7/27/2005). Contingent and alternative employment arrangements, February 2005. Department of Labor, Washington, D.C. USDL 05-1433.

Buttner, E. H., & Moore, D. P. (1997). Women's organizational exodus to entrepreneurship: Self-reported motivations and correlates with success. Journal of Small Business Management, 35, 34-46.

Cohen, J. (1977). Statistical power analysis for the behavioral sciences, NY: Academic Press.

Collins, C. J., Hanges, P. J., & Locke, E. A. (2004). The relationship of

achievement motivation to entrepreneurial behavior: A meta-analysis. Human Performance, 17, 95-117.

Danziger, N., & Valency, R (2006). Career anchors: distribution and impact on job satisfaction, the Israeli case. Career Development International, 11, 293-303.

Davis-Blake, A., & Pfeffer, J. (1989). Just a mirage: The search for dispositional effects in organizational research. Academy of Management Review, 14, 385-400.

Dawis, R. V., & Lofquist, L. H. (1984). A psychological theory of work adjustment. Minneapolis: University of Minnesota Press.

DeCarlo, J. F., & Lyons, P. R. (1979). A comparison of selected personal characteristics of minority and non-minority female entrepreneurs. Journal of Small Business Management, 17, 22-29.

Eden, D. (1975). Organizational membership versus self-employment: Another blow to the American dream. Organizational Behavior and Human Performance, 13, 79-94.

Feldman, D.C., & Bolino, M.C. (2000). Career patterns of the self-employed: career motivations and career outcomes. Journal of Small Business Management, 38, 53-67.

Fogarty, T. J., Singh, J., Rhoads, G. K., & Moore, R. K. (2000). Antecedents and consequences of burnout in accounting: Beyond the role stress model. Behavioral Research in Accounting, 12, 31-67.

Gartner, W. B. (1989). "Who is an entrepreneur?" is the wrong question. Entrepreneurship: Theory and Practice, 13, 47-68.

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

Hisrich, R., Langan-Fox, J., & Grant, S. (2007). Entrepreneurship research and practice: A call to action for psychology. American Psychologist, 2, 575-589.

Holland, J. L. (1985). Making vocational choices: A theory of vocational personalities and work environment. Englewood Cliffs: NJ: Prentice-Hall, Inc.

Hornaday, J. A., & Aboud, J. (1971). Characteristics of successful entrepreneurs. Personnel Psychology, 24, 141-153.

Hundley, G. (2001). What and when are the self-employed more satisfied with their work? Industrial Relations, 40, 293-316.

Jackson, D. N. (1999). PRF Test Manual, Port Huron, MI: Sigma Assessment.

Judge, T. A., Bono, J. E., Ilies, R., & Gerhardt, M. W. (2002). Personality and leadership: A qualitative and quantitative review. Journal of Applied Psychology, 87, 765-780.

Katz, J. A. (1994). Modelling entrepreneurial career progressions: Concepts and considerations. Entrepreneurship: Theory and Practice, 19, 23-41.

Kinicki, A. J., McKee-Ryan, F. M., Schriesheim, C. A., & Carson, K. P. (2002). Assessing the construct validity of the Job Descriptive Index: A review and meta-analysis. Journal of Applied Psychology, 87, 14-32.

Kolvereid, L. (1996). Organizational employment versus self-employment. Entrepreneurship: Theory and Practice, 21, 47-57.

Kristof, A. L. (1996). Person-organization fit: An integrative review of its conceptualizations, measurement, and implications. Personnel Psychology, 49, 1-49.

Kristof-Brown, A. L, Zimmerman, R. D., & Johnson, E. C. (2005). Consequences of individuals' fit at work: A meta-analysis of person-job, person-organization, person-group, and person-supervisor fit. Personnel Psychology, 58, 281-342.

Lee, D. Y., & Tsang, E. W. K. (2001). The effects of entrepreneurial personality, background and network activities on venture growth. Journal of Management Studies, 38, 583-602.

McClelland, D. C. (1961). The achieving society. Princeton, NJ: Van Nostrand,

Medcof, J. W., & Hausdorf, P. A. (1995). Instruments to measure opportunities to satisfy needs, and degree of satisfaction of needs, in the workplace. Journal of Occupational and Organizational Psychology, 68, 193-208.

Murray, H. A. (1938). Explorations in personality: A clinical and experimental study of fifty men of college age. NY: Oxford University Press.

Peel, S., & Inkson, K. (2004). Contracting and careers: choosing between self and organizational employment. Career Development International, 96, 542-558.

Prottas, D. J. (2007). Attitudes towards occupation, job, life, and family: Translators as independent contractors, employers, and employees. Journal of Business and Entrepreneurship, 19, 16-30.

Prottas, D. J., & Thompson, C. A. (2006). Stress, satisfaction, and the work family interface: A comparison of self-employed business employers, nonemployers, and organizational employees. Journal of Occupational Health Psychology, 11, 366-378.

Rauch, A., & Frese, M. (2007). Let's put the person back into entrepreneurship research: A meta-analysis on the relationship between business employers' personality traits, business creation, and success. European Journal of Work & Organizational Psychology, 16, 353-385.

Robichaud, Y., McGraw, E., & Roger, A. (2001). Toward the development of a measuring instrument for entrepreneurial motivation. Journal of Developmental Entrepreneurship, 6, 189-201

Schein, E. H. (1978). Career dynamics: Matching individual and organizational needs. Reading, MA: Addison-Wesley.

Schein, E. H. (1994). Commentary: What is an entrepreneur? Entrepreneurship: Theory and Practice, 19, 87-88.

Schneider, B. (1987). The people make the place. Personnel Psychology, 40, 437-453.

Schneider, B. (2001). Fits about fit. Applied Psychology: An International review, 59, 141-152.

Schumpeter, J. A. (1934). The theory of economic development; an inquiry into profits, capital, credit, interest, and the business cycle. Cambridge, MA: Harvard University Press.

Shane, S., Locke, E. A., & Collins, C. J. (2003). Entrepreneurial motivation. Human Resource Management Review, 13, 257-279.

Sims, H. P., Jr., Szilagyi, A. D., & Keller, R. T. (1976). The measurement of job characteristics. Academy of Management Journal, 19, 195-212.

Stevens, J. P. (2002). Applied Multivariate Statistics for the Social Sciences. Mahwah, NJ: Lawrence Erlbaum.

Stewart, W. H. Jr., & Roth, P. L. (2001). Risk propensity differences between entrepreneurs and managers: A meta-analytic review. Journal of Applied Psychology, 86, 145-153.

Stewart, W. H., Jr., & Roth, P. L. (2007). A meta-analysis of achievement motivation differences between entrepreneurs and managers. Journal of Small Business Management, 45, 401-421.

Tetrick, L. E., Slack, K. J., Da Silva, N., & Sinclair, R. R. (2000). A comparison of stress-strain process for business employers and nonemployers: Differences in job demands, emotional exhaustion, satisfaction, and social support. Journal of Occupational Health Psychology, 5, 464-476.

U.S. Bureau of Labor Statistics. (2005). Contingent and alternative employment arrangements, February 2005. USDL 05-1433.

U.S. Census Bureau. Nonemployer Statistics, 2005 Total for all sectors United States. Retrieved September 20, 2007 from http://www.census. gov/epcd/nonemployer/2005/us/US000.HTM.

Zhao, H., & Seibert, S. E. (2006). The Big Five personality dimensions and entrepreneurial status: A meta-analytical review. Journal of Applied Psychology, 92, 259-271 .

David Prottas

Adelphi University

Author info: Correspondence should be sent to: Dr. David Prottas, School of Business, Adelphi University, 1 South Ave., Garden City, NY 11530. prottas@adelphi.edu.
TABLE 1 Basic Statistics and Correlations (All Participants)

      Variables     M        SD      1      2      3      4       5

1.    Gender         .25      .44   -
2.    Age          52.93    11.69   -.26   -
3.    Marital        .83      .38   -.25    .13   -
4.    Income      179.4    180.4    -.32    .14    .18   -
5.    Opp. Ach.     6.10      .78    .01   -.00   -.05    .10   (.78)
6.    Opp. Aff.     5.46      .98   -.04   -.12   -.00    .27    .37
7.    Opp. Aut      6.33      .80   -.04    .22    .03    .11    .48
8.    Opp. Dom.     4.55     1.52   -.11   -.10    .05    .35    .34
9.    Need Ach.    11.29     2.49    .00   -.03   -.03    .09    .35
10.   Need Aff.     9.68     3.76    .07   -.14   -.04    .00    .18
11.   Need Aut.     7.00     2.88    .06    .03   -.18   -.07   -.11
12.   Need Dom.    10.79     3.54   -.11   -.15    .04    .19    .24

      Variables     6       7       8       9      10      11      12

1.    Gender
2.    Age
3.    Marital
4.    Income
5.    Opp. Ach.
6.    Opp. Aff.   (.67)
7.    Opp. Aut     .22    (.83)
8.    Opp. Dom.    .60     .05    (.86)
9.    Need Ach.    .21     .06     .15    (.59)
10.   Need Aff.    .25     .20     .16     .13    (.82)
11.   Need Aut.   -.18     .03    -.21    -.04    -.34    (.66)
12.   Need Dom.    .29     .13     .32     .33    .37     -.14    (.80)

Note. Ns: 317 to 322, except income, 280. Categorical variables:
Gender (0 = Male, 1 = Female); Marital (0 = Single, 1 = Married or in
a similar relationship). Income in $000's. Opp: = Opportunity. Ach. +
Achievement. Aff. = Affiliation; Aut. = Autonomy; Dom. = Dominance.
Cronbach alphas are shown along diagonals. Correlations > .20
significant at p < .001, two-tailed; > .15 significant at p < .01,
two-tailed; > .12 significant at p < 05, two-tailed.

TABLE 2 Multivariate Analysis of Differences: Work Arrangement as
Fixed Factor

                            Univariate Relationships

                      Achievement               Affiliation

Dependent                [[eta].sub.p.             [[eta].sub.p.
Variables        F          sup.2]         F          sup.2]

Model 1:

Actual Work Arrangement as Fixed Factor

Opportunities    3.45         .02         14.69         .09
Needs            1.48         .01          1.53         .01

Model 2:

Actual Work Arrangement as Fixed Factor--Age, Sex as covariates

Opportunities    3.39         .02         12.92         .08
Needs            1.60         .00          2.02         .01

Model 3:

Preferred Work Arrangement as Fixed Factor

Opportunities     .45         .00          5.26         .03
Needs            2.79         .02           .91         .01

                      Univariate Relationships

                    Autonomy                Dominance

Dependent               [[eta].sub.p.           [[eta].sub.p.
Variables         F        sup.2]         F        sup.2]

Model 1:

Actual Work Arrangement as Fixed Factor

Opportunities   35.84        .18        46.83        .23
Needs           10.00        .06         5.38        .03

Model 2:

Actual Work Arrangement as Fixed Factor--Age, Sex as covariates

Opportunities   27.07        .15        39.58        .20
Needs            9.30        .06         3.88        .03

Model 3:

Preferred Work Arrangement as Fixed Factor

Opportunities   13.24        .08        22.43        .12
Needs           14.59        .09         5.98        .04

Note. Work arrangement is non-ordered categorical variable with three
values: employer, nonemployer, or employee. Models 1 and 2 classified
each participant according to their actual work arrangement. Model 3
classifies each participant according to their preferred work
arrangement. F statistics based on the Wilks lambda. Univariate F
statistics [greater than or equal to] 9.30 significant at p < .001,
two-tailed; [greater than or equal to] 5.26 at p < .01, two-tailed;
[greater than or equal to] 3.45 at p < .05, two-tailed.

TABLE 3 Group Means and Differences in Opportunities and Needs
for Achievement, Affiliation, Autonomy and Dominance

                   Opportunities            Needs

Group            M    SD     d        M      SD     d

                    Achievement

Employers      6.21  .80             11.42  2.37
                  d =.12                 d =.20
Nonemployers   6.11  .81    .33      10.93  2.54   .03
                  d =.20                 d =.22
Employees      5.94  .86             11.49  2.58

                    Affiliation

Employers      5.69  .90             10.04  3.55
                  d =.66                 d =.24
Nonemployers   5.03  1.11   .13       9.15  3.90   .08
                  d =.57                 d =.15
Employees      5.58  .78              9.74  3.86

                    Autonomy

Employers      6.54  .56              6.83  2.65
                  d =.09                 d =.41
Nonemployers   6.59  .57    .93       7.98  2.99   .23
                  d =.95                 d =.62
Employees      5.81  1.01             6.20  2.79

                    Dominance

Employers      5.03  1.28            11.08  3.45
                  d =1.18          d = .36
Nonemployers   3.45  1.41   .02       9.84  3.49   .08
                  d =1.13          d = .44
Employees      5.01  1.34            11.39  3.55

Note. Opportunities n's: Employers, 128; nonemployers, 97; employees,
95. Needs n's: Employers, 127; nonemployers, 97, employees, 93.
Standardized mean difference (d) between employers and nonemployers
and between nonemployers and employees appear in the rows between.
The ds for the difference between employers and employees appear in
the columns to the right. The statistical significances were
calculated with Bonferroni correction. d's [greater than or equal to]
.57 are significant at p < .001, two-tailed; [greater than or equal
to] .44 at p < .01, two-tailed; [greater than or equal to] .33 at p <
.05, two-tailed.
Gale Copyright:
Copyright 2011 Gale, Cengage Learning. All rights reserved.