Entrepreneurial Expectancy,Task Effort,and Performance *

Abstract

Research to date has not adequately explained the role that expectancy of entrepreneurial performance based on perceived ability plays in motivating persons to persevere on an entrepreneurial task. This study investigated the entrepreneurial expectancy, effort–performance linkage via a World Wide Web–based experiment involving 179 undergraduate business students at a large midwestern university. Results indicated that the type of feedback (positive versus negative) that individuals received regarding their entrepreneurial ability (regardless of actual ability) changed expectancies regarding future business start–up, but did not alter task effort or quality of performance. Individuals receiving positive feedback about their entrepreneurial abilities had higher entrepreneurial expectancies than individuals receiving negative feedback. We also found that males had higher expectancies regardless of experimental condition than females.

Introduction

Much early entrepreneurship research focused on the search for an entrepreneurial personality (Smilor, 1997; Wortman, 1987). However, the quest to find a consistent set of traits that characterized successful entrepreneurs was troubled at best (Gartner, 1988; Carland, Hoy, & Carland, 1988; Shaver, 1995), largely mirroring earlier efforts in leadership research that sought to differentiate leaders from nonleaders (Geier, 1967; Yukl & Van Fleet, 1992). As a result, entrepreneurship research shifted to explore new venture creation from other perspectives. Shaver and Scott (1991) pointed out, however, that despite the failure of personological trait research, it was still the individual who created a new venture.

Recent research has demonstrated the impact that cognitive and social processes have on entrepreneurial behavior (see for example, Baron, 1998; Douglas & Shepherd, 2000; Gatewood, Shaver, & Gartner, 1995; Krueger, Reilly, & Carsrud, 2000; Nicholson, 1998). One important investigation has focused on how entrepreneurs think about themselves and their abilities, particularly as these thoughts relate to their willingness to persevere with a new venture, even in the face of failure (Baron, 1999; Simon, Houghton, & Aquino, 2000; Shaver & Scott, 1991).

Studies of existing entrepreneurs, however, are necessarily compromised by the fact that the entrepreneur's expectancies and success go hand–in–hand. Only in an experimental context is it possible to manipulate expectancies independently of past performance, and that was the primary objective of the present research. One past experiment has investigated the expectancy–performance link, but its dependent variables were limited in scope. This study, by Pieterman, Shaver, and Gatewood (1993), reported that executive MBAs who were provided negative feedback about their entrepreneurial abilities showed less effort critiquing a business plan than their counterparts who were provided positive feedback. The primary dependent variable in that study was a count of the number of words used in the critique. Arguably, this measure did not fully capture the persistence needed for entrepreneurial success. More important, a mere word count cannot speak to the quality of the advice offered. Consequently, the present research investigated whether feedback regarding entrepreneurial abilities would change business expectancies, and whether those expectancy changes, in turn, would affect subjects’ level of effort and performance on an entrepreneurial task. In the context of a Web–based experiment using undergraduate students, the questions we sought to answer were, “Can people be induced to believe in certain ways that affect their effort and performance on an entrepreneurial task? Furthermore, do these effects, if they exist, differ by sex?”

THEORETICAL CONTEXT

Entrepreneurship Motivational Research

Few topics in the applied social sciences have received as much attention as has the subject of motivation. In the field of management research, motivation has been conceptualized using a number of theoretical frameworks intended to explain outcomes as diverse as work performance (Miller & Monge, 1986), goal achievement (Locke, 1968), turnover (Vroom, 1964), and job satisfaction (Iaffaldano & Muchinsky, 1985). Defined as a psychological force inciting an individual to exert effort toward particular individual or organizational goals, motivation serves as a mechanism for satisfying an individual need (George & Jones, 1999; Robbins, 1998). Within the broad area of motivation research, the notion of expectancy has played a central role in explaining human motivation in the workplace (Ambrose & Kulik, 1999; Katzell & Thompson, 1990; Locke & Latham, 1990). It has been used to account for everything from occupational preferences and job satisfaction to volunteer attendance decisions (Harrison, 1995). Expectancy theory's fundamental premise is that behavior is a function of an individual's expectation that a response will bring reinforcement together with the perceived value of the reinforcement (Rotter, 1954).

Central to expectancy theory is an individual's cognitive recognition of three important relationships (Vroom, 1964). First, people must believe that exerting a given amount of effort can result in the achievement of a particular level of performance (the effort–performance relationship). If people perceive that their given skill and ability set is not adequate or that circumstances beyond their control will conspire against the needed level of performance, they will not be motivated to engage in the necessary behaviors. Second, people must believe that a particular performance level will in fact result in a specified outcome (instrumentality relationship). Assuming that the first relationship linkage holds, an individual would still not be motivated to perform if he or she perceives that the needed level of performance will not result in the outcome anyway (Shepperd & Taylor, 1999). Finally, the reward or outcome must be attractive in order for people to be motivated to attain it (the valence–personal goals relationship). Thus, even though the first two relationships might hold, an individual would remain unmotivated if the benefit or satisfaction associated with the reward or outcome was not high enough.

Many entrepreneurship researchers have offered expectancy and subjective expected utility “like” models to describe the factors that influence an individual's choice to pursue an entrepreneurial career (e.g., Campbell, 1992; Gatewood, 1993; Herron & Sapienza, 1992; Katz, 1992). Two recent studies provide good examples of how expectancy ideas have been incorporated (but not described explicitly) into models describing the cognitive mechanisms influencing the choice to start a business. Douglas and Shepherd (2000) offered a model of entrepreneurial intentions that is grounded in ideas from economics. In their model, the choice to pursue entrepreneurship is based on a person's utility function, which reflects perceptions about the income anticipated, the amount of work effort anticipated to achieve this income, the risk involved, plus other factors such as the person's desired attitudes for independence and perceptions of the anticipated work environment. In a mathematical model of these variables, they suggest that individuals will seek to maximize their utility from their work choices. When external conditions (such as funding) and the presence of opportunities are also favorable, individuals choosing entrepreneurship are more likely to have high abilities for entrepreneurship and, therefore, expend more effort toward success at entrepreneurship. The Douglas and Shepherd (2000) utility framework has obvious similarities to an expectancy approach. They implicitly suggest that perceived utility is a function of an individual's perception of the likelihood that personal abilities and efforts in entrepreneurial activity will be successful (expectancy) and that the outcomes will be of value (instrumentality and valence).

In a second study, Krueger, Reilly, and Carsrud (2000) compared predictions from two models of factors that influence entrepreneurial intentions, one based on Ajzen's theory of planned behavior (Ajzen, 1987; Kim & Hunter, 1993) and the other based on Shapero's model of the “entrepreneurial event” (Shapero, 1982). Both models suggest that an individual's expected values will influence the perceived desirability of the intention to pursue entrepreneurship. In an empirical comparison of both models, using a survey of 97 senior university business students, Krueger, Reilly, and Carsrud (2000) found that measures of perceived desirability and expected utility were significantly correlated with intentions for entrepreneurship. Their conclusion was that in the direct comparison, Shapero's model was preferable to the theory of planned behavior for explaining occupational intentions. For our purposes, however, it is worth noting that both models contain terms—feasibility, desirability, anticipated effort—that are also at home in expectancy theory.

Entrepreneurial Ability, Effort, Performance, Expectancies

Expectancies represent the mechanism through which past experiences and knowledge are used to predict the future. As such, expectancies are beliefs about a future state of affairs. But expectancies are derived from beliefs about self, other people, and about the nonsocial world, for example, event expectancies. (Although expectancies are beliefs, not all beliefs are expectancies). Beliefs flow from three major sources: direct experience, indirect experience, and other beliefs (see Olson, Roese, & Zanna, 1996, for a detailed description). Entrepreneurial expectancies, therefore, can derive from personal experience, for example starting a business; communication, feedback, or information from other people (Gatewood, 1993); or from other beliefs, for example, attributions from past events (Gatewood & Shaver, 1991).

Previous research on expectancy theory suggested that perceptions of skills and abilities influence expectancy perceptions (Katzell & Thompson, 1990; Rasch & Tosi, 1992). Individuals choose tasks and put forth effort on the basis of their expectancies. People avoid activities that they believe exceed their abilities, while choosing those activities they feel capable of handling (Bandura, 1982). Finally, individuals persist longer and put more effort on tasks in which they expect to succeed (Olson, Roese, & Zanna, 1996). In general, individuals who expect to perform well do (Oettingen, 2000). However, feedback concerning capabilities and past performance may affect an individual's motivation and future performance levels (Earley, P. C., Northcraft, G. B., Lee, C. L., & Litushy, T. R. (1990)).

Entrepreneurship literature has also found that persons who believe that their skill and ability set is adequate for achieving success with a new venture are motivated to exert the necessary effort (Douglas & Shepherd, 2000; Shaver, Gatewood, & Gartner, 2001). Gatewood and Shaver (1991), however, argued that motivation might be undermined when people's confidence in their capabilities is reduced through previous failure or through negative feedback about their abilities. As noted earlier, Pieterman, Shaver, and Gatewood (1993) found that executive MBAs who were provided negative feedback about their entrepreneurial aptitudes showed less effort in critiquing a business plan than those who were provided positive feedback. Busenitz's (1999) research also suggested that entrepreneurs lacking in personal confidence are less likely to take on the considerable risks of a new venture. The perceived risk, however, is also a function of the time horizon involved (Das & Bing–Sheng, 1997). In other words, if the risk is perceived to be immediate, some entrepreneurs might choose to desist from pursuing a new venture.

Based on this research, then, an individual who questions his or her ability to succeed with an entrepreneurial task, particularly when the risks are immediately evident, will be less likely to expend the effort needed to be successful than a person with high confidence in his or her ability, particularly when it is externally reinforced (i.e., these people have received negative feedback from other persons about their capabilities). Second, doubting one's own capabilities is likely to result in the self–fulfilling prophecy of reduced performance on the task (Hilton & Darley, 1991; Neuberg, 1994). Furthermore, when expectancies are disconfirmed, the expectancies will become less certain for the future (Olson, Roese, & Zanna, 1996). Hence, the following hypotheses are offered:

Hypothesis 1a: Individuals receiving negative feedback about their entrepreneurial abilities will exert less effort on an entrepreneurial task than individuals receiving positive feedback.

Hypothesis 1b: Individuals receiving negative feedback about their entrepreneurial abilities will demonstrate lower levels of performance on an entrepreneurial task than individuals receiving positive feedback.

Sex Differences

Although the above relationship is presumed to hold in the broad case, previous research suggests that males and females in general, and male and female entrepreneurs in particular, may have different expectancies, specifically regarding effort and performance beliefs. Moreover, the effects of negative feedback may be more pronounced for women than for men. An early review of sex differences (Maccoby & Jacklin, 1974) showed very few differences between men and women. One of the few noted, however, was in achievement–related self–confidence. Self–confidence (performance expectancies and self–evaluations of performance) was lower for women than men. However, Lenney (1977) showed that the lower self–confidence of women was related to the nature of the task, performance feedback, and the presence of social comparison or evaluation. Hackett and Betz (1981) proposed that expectations of personal efficacy (the ability to perform a task or behavior) influence career decisions and motivations of men and women. In particular, women held lower expectancies of self–efficacy for nontraditional female occupations than for traditional female occupations (Clements, 1987; Hackett & Betz, 1981; Nevill & Schlecker, 1988).

Kourilsky and Walstad (1998), in their national study of entrepreneurial tendencies among youth, found that females were significantly less interested in starting businesses, less confident in their abilities, and less tolerant of market dynamics than their male counterparts, findings consistent with previous research (Crant, 1996; Matthews & Moser, 1996; Scherer, Brodzinski, & Wiebe, 1990). Shaver, Gatewood, and Gartner (2001), in a comparison of male and female nascent entrepreneurs and a control group on their expectancies for business startup, found that male and female nascents had higher expectancies for business startup than the male and female controls. They also found that female controls had significantly lower expectancies for business startup than male controls. In an analysis of individual items, females scored lower than males (regardless of nascency status) on their confidence in putting in the effort to start businesses, on presuming their skills and abilities would be very valuable in starting businesses, and considering starting businesses as more desirable than other career opportunities. Males and females, however, had similar scores on the beliefs that if they worked hard they could successfully start businesses, that starting businesses would lead to achieving other important goals, and the value they placed on their past experience for starting businesses.

These findings support Whitley, McHugh, and Frieze's (1986) contention that males and females differ in their reliance on ability and effort attributions. Males rely more on ability self–assessments and females on how much effort they intend to expend. Henry and Strickland (1994) found that self–assessed ability and effort was predictive of task performance among males, but not for females.

Finally, a number of theorists have proposed that women have lower achievement motivation in the face of failure because of their reliance on other's opinions of them (Roberts & Nolen–Hoeksema, 1994). Roberts and Nolen–Hoeksema found that women reported that their self–assessments were more affected by other's evaluations whether positive or negative, while men were influenced by positive assessments. They also found that women showed a greater responsiveness to feedback than men and saw these assessments as more accurate. Women may also see failures as more threatening to success. For example, Van Auken (1999) determined that women entrepreneurs were less likely to perceive that business obstacles could be overcome.

Considering this research, it is plausible that when confronted by negative feedback (e.g., having others question their abilities) women may internalize that feedback more readily than men and in turn reduce their efforts and commitment to an entrepreneurial task. Hence, the following hypotheses are offered:

Hypothesis 2a: Females receiving negative feedback about their entrepreneurial abilities will exert less effort on an entrepreneurial task than males receiving negative feedback.

Hypothesis 2b: Females receiving negative feedback about their entrepreneurial abilities will demonstrate lower levels of performance on an entrepreneurial task than males receiving negative feedback.

METHOD

Sample

The results of this study are based on data collected in Spring 2000 from 179 undergraduate college students (79 females and 100 males) in an organizational behavior course in a school of business. This upper–level course for juniors and seniors is a required class for business majors as well as for some students in some other programs outside of the school of business. At this point it might be useful to note that the objective of the research was not to identify the preexisting prevalence of entrepreneurial attitudes among the general population (an objective that would argue against the use of members of a business undergraduate course as research participants). Rather, the research rests on the now–widely accepted view that entrepreneurial action is a joint function of both individual differences and situational influences. In this study, as in other experimental research, the objective was to determine whether the conditions could produce differences in the behavior of individuals who have not been preselected for their entrepreneurial tendencies. This more limited objective requires only that research participants be familiar with the “jargon” used in the research.

The research procedures required participants to complete an initial questionnaire, receive (bogus) feedback on their entrepreneurial propensity, read and comment on elements of a standard business case, and complete a postexperimental questionnaire that included—among other things—a check on the manipulation of expectancy. Only 156 participants (69 females and 87 males) actually finished the postexperimental questionnaire, so the sample analyzed was restricted to these individuals. (There was no significant relationship between feedback received and whether or not the final questionnaire was completed.)

Procedure

Study participants were asked to take part in an online experiment in which they were told that their entrepreneurial potential would be assessed. Participants were assigned unique user names and passwords to access the Web–based study and invited to complete it at their leisure during a five–day period. By initiating the study in an online environment, the students could then spend as much or as little time on the study as they chose, an important component of the research related to measuring effort.

The study itself consisted of three distinct tasks that students completed in sequence. First, they were to complete the Entrepreneurial Attitude Questionnaire (EAQ), a 35–item instrument described by Shaver, K. G., Garther, W. B., Gatewood, E. J., & Vos L. H. (1996) to measure beliefs about one's entrepreneurial attitudes and values. For this study, the EAQ was used for two distinct purposes. First, it served as a measure of individual differences in entrepreneurial potential (to be used as a covariate in the statistical analyses, as described below). More important, it served as the vehicle to provide research participants immediate (bogus) feedback about their entrepreneurial potential. Specifically, participants were randomly assigned “positive” or “negative” feedback about their potential as entrepreneurs compared to a normative group of “real” entrepreneurs. This feedback was said to have been derived from computerized scoring of their EAQ answers, with the results shown in a graphic profile and also described in words.

Participants were told that their scores on the EAQ were useful for understanding their own profile potential for success as an entrepreneur. Because students could have completed the study in a computer laboratory with multiple machines, we created eight different versions of the “positive” feedback and eight different versions of the “negative” feedback. Each of these written versions was accompanied by one of three different bar graphs purported to be the student's profile, so that there were 48 different feedback possibilities. These were randomly presented, so across the entire set of 179 participants, there were no more than four individuals who received identical feedback patterns. It is important to note, however, that all these variations were within one scale point (on a series of 7–point scales) of one another, whereas the differences across conditions were a minimum of 3 scale points. We instructed the students that they were to do independent work, and could not talk with classmates during the study. Together, these precautions made it highly unlikely that any adjacent people received feedback patterns that were identical. Positive and negative feedback samples are shown in Table 1.

Table 1

Sample EAQ Feedback

Written Example	Corresponding Bar Graph
(Positive Example)
Overall, your EAQ profile suggests that you have strong entrepreneurial potential. Your scores on the Actions and Original subscales are higher than those of the entrepreneur sample, and your score on the Social subscale is lower than those of the entrepreneur sample. Your scores on the other four subscales are within the normal limits of the entrepreneur sample average. Together, these scores suggest that you are determined, creative, and recognize that it is impossible to change the entire world. This is an excellent profile for entrepreneurial success.
Written Example	Corresponding Bar Graph
(Negative Example)
Overall, your EAQ profile suggests that you have lower than average entrepreneurial potential. Your scores on the Actions and Original subscale are lower that those of the entrepreneur sample, while your scores on the Social subscale are higher than those of the entrepreneur sample. Your scores on the other four subscales are within the normal limits of the entrepreneur sample average. Together, these scores indicate that you are less likely than the average entrepreneur to take determined action, to be creative, and to recognize that it is impossible to change the entire world. This profile suggests that you would not likely achieve entrepreneurial success.

Once students completed the EAQ and received their feedback, the second task was to read a business case involving two entrepreneurs seeking to start and grow a business (Vesper, 1996). From a central main page, students could navigate to various pages of the case in any order or depth that they chose by clicking and viewing text links to sections such as background on the case, firm financials, and the business plan. Students were instructed to use this information to answer five questions about the future potential of the new venture. These questions, and the directions on how to answer them, are shown in Table 2. Students were told that three winners of a monetary prize would be chosen based on their performance answering the case questions and the results from their EAQ. As some of the EAQ subscales are worded in the positive direction, whereas others are worded in the negative direction, it is unlikely that the prizes affected overall EAQ scores. In fact, six monetary prizes were subsequently awarded, three for the best–judged answers in the positive feedback condition and three more for the best–judged answers in the negative feedback condition.

Table 2

Case Questions And Instructions For Completing Them

At any time feel free to answer the case questions below. In the event that you do not finish answering the questions in one sitting, what you enter will be saved and you can return to enter more information or edit what you have already entered. However, once you click on the submit button on the bottom of this page, you will not be able to return to this page to enter additional information. Instead, you will be instructed to proceed to the Post–Case Questionnaire.

1) How profitable do you think this venture can be if things go well?

2) What are the most important assumptions that must hold in order for this venture to be a success?

3) How well suited to this venture are the entrepreneurs and their boards?

4) If this new venture is unsuccessful, what do you think will be the reasons for failure?

5) Would you invest your time and resources in this venture? Why?

As students maneuvered through the case, information on the pages that they visited and the time spent on each page was automatically downloaded into a central spreadsheet. Whenever participants submitted information, it was also automatically logged. Additionally, students could enter and exit the case as often as they wished, and this information was also captured. The third and final component of the study involved students answering a questionnaire that included items measuring their expectancies for starting a business, as well as some additional “filler” questions about the case.

Measures

The dependent measures in this study were 1) specific and global measures of page viewing, 2) overall counts of words provided in participants’ answers to the questions about how the plan might be improved or evaluated, and 3) a qualitative assessment of participants’ answers to the case questions. Thus, we tested both the sheer effort on task, and the quality of the effort (performance) put forward. More important, effort and performance were measured in ways that were much less apparent to participants than is the case in most laboratory studies. Total page hits and times were also analyzed.

The first dependent variable was the overall word count present in the participants’ answers to the case questions. Previous research by the authors (Pieterman, Shaver, & Gatewood, 1993) used similar word counts in an expectancy–based study among EMBA students. The number of words used to assess a business plan (the experimental material) distinguished EMBA students who had been led to believe that they had high entrepreneurial potential and those who had been led to believe that they did not have such potential.

As noted earlier, however, simple word counts may not be sufficient to capture various sorts of effort, and they certainly cannot provide any estimate of the quality of the performance that is rendered. As it regards effort, we expected that the unique capabilities of a Web–based survey to capture time on tasks, as well as Web–page “hits,” would provide a unique means of measuring effort that was more robust than what could be accomplished with word counts alone. Thus, effort was also measured as the combination of the maximum number of hits received by each participant's most frequently visited page and the time (in seconds) spent viewing the longest–viewed single page. These two variables were treated as a repeated measure, in case they might have produced different patterns.

The third dependent variable was a qualitative assessment of the answers for the case questions. Specifically, three MBA students were recruited and trained as evaluators. Based on a set of model answers for each question produced by one of the authors and the MBA evaluators, the evaluators were instructed to read all of the study participants’ answers and rate each answer for each question on a 5–point scale with 1 being poor and 5 being excellent. For each study participant, the scores were added across the five questions, producing a total participant score.

The manipulation of expectancy was checked with six expectancy questions from the final questionnaire. These six questions had been used in the Panel Study of Entrepreneurial Dynamics (PSED) as a measure of entrepreneurial expectancies. Nascent entrepreneurs scored significantly higher on these expectancy items than a nonnascent control group (Shaver, Gatewood, & Gartner, 2001). In the present research the six items were subjected to a principal components factor analysis (described in detail below) that revealed only one factor (as had been the case in the Shaver, Gatewood, & Gartner (2001) paper. The six items were, therefore, combined into a single score to be used as a covariate in the statistical design.

Results

Preliminary Analyses

Given the nature of the dependent variables, preliminary analyses were conducted on the EAQ scores, the raters’ judgments of the quality of respondents’ answers, and the expectancy manipulation check. Although the EAQ subscales have shown essentially the same pattern in a number of prior studies (e.g., Calver & Shaver, 1998; Shaver et al., 1996), we confirmed the item placements in the present study. For present purposes, seven of the EAQ subscales are particularly important as possible covariates in the subsequent analyses of variance (these were the same seven subscales described as part of the feedback the participants received). Participants’ behavior might well be influenced by their preexisting individual differences in 1) achievement orientation (two separate EAQ subscales), 2) entrepreneurial creativity (three separate EAQ subscales), and 3) locus of control (two EAQ subscales).

Specifically, participants higher in achievement motivation might exert more effort in what they regarded as an achievement task, so achievement was used as a covariate for the subsequent analyses of page hits, page times, and words used in the answers, regardless of the manipulations. Creativity might also affect participant behavior; participants higher in entrepreneurial creativity might produce “better” answers to the case questions, again without regard to the experimental conditions, so individual differences in creativity were used as covariates for the analyses of answer quality. Finally, participants higher in internal locus of control might be more likely to believe that their efforts would be successful, and this might also affect effort, regardless of the experimental conditions. So locus of control was an additional covariate for the dependent measures.

The seven EAQ subscales included in the analysis are shown in Table 4, which includes the subscale content, number of items involved, and Cronbach reliability level obtained in this study. As it happens, only one of these subscales achieved the >.70 reliability normally recommended. We will still use all, as they are intended only as covariates rather than as dependent variables. Nevertheless, the moderate reliabilities might reduce the preexperimental variance for which they might account.

Table 3

Expectancy Items

1) If I work hard, I can successfully start a business

2) Starting a business is much more desirable than other career opportunities I have

3) If I start a business, it will help me achieve other important goals in my life

4) Overall, my skills and abilities will help me start a business

5) My past experience will be very valuable in starting a business

6) I am confident I can put in the effort needed to start a business

Table 4

EAQ Subscales, Content, Numbers of Items, And Cronbach Reliabilities

EAQ Subscale	Content	Number of Items	Cronbach Reliability
Actions	The extent to which you are ready to take achievement–related actions.	3	.69
Beliefs	Your core beliefs about the value of achievement.	4	.73
Personal	The extent to which you think that you can control outcomes of primary importance to you.	3	.57
Social	The extent to which you believe that people can control large social institutions such as aspects of government or world business.	4	.62
Efficient	Your level of personal organization and attention to detail.	5	.62
Original	The extent to which you have original ideas that might produce radical change in the way things are done.	3	.57
Rules	The importance to you of following rules and accepted procedures.	4	.64

Before the raters’ judgments of work quality could be used as a dependent variable, it was necessary to show that there was sufficient consistency among the three judges’ evaluations of the participants’ responses to the open–ended case questions. The raters were very familiar with the experimental materials because they had been members of an MBA course that had used the case for class discussion. The raters were instructed to reread all of the case materials, answer the five case questions, and develop a scoring guide. The raters met with one of the authors to develop a joint scoring guide. The raters then graded the answers to all five questions from the first 20 respondents. They then met with one of the authors to compare their assessments of the answers to check for inconsistencies in scoring, and then graded all the respondents’ answers. The five questions dealt with the anticipated profit from the venture, the success factors for the venture, the suitability of the entrepreneurs and their board for the venture, the potential reasons for failure of the business, and the presumed value of investing in the business. The interjudge reliability was assessed separately for each of the five questions, using an intraclass correlation coefficient as the estimator. For each of the five questions, the reliability among judges was impressive: alpha levels of .90 for profit, .93 for success, .88 for board, .94 for failure, and .89 for investment. Thus, the ratings for all judges were combined into a single score.

The six expectancy items included in the postexperimental questionnaire were developed to assess the links in the expectancy–value approach to motivation. If the purpose of the present study had been to explore these links in detail, it would make sense to treat the six items separately. Our purpose here, however, was to separate the expectations of the negative feedback group from those of the positive feedback group, regardless of what particular form that change might take. Consequently, we hoped to be able to treat changes in expectations as a single consequence of the manipulations. To do so, we need to show that in this study the six expectancy items could be regarded as a single measure.

We subjected the six expectancy items to a principal components factor analysis (varimax rotation). This factor analysis produced a single factor, with an initial eigenvalue of 3.50, which accounted for 58 percent of the variance (Cronbach reliability of the resulting scale was .84). As there was only a single factor, no rotation was possible. To test the effectiveness of the feedback manipulation, this single expectancy score was subjected to a 2 × 2 (Feedback × Participant Sex) analysis of variance. This analysis revealed two main effects. One main effect confirmed the effectiveness of the feedback manipulation, with participants in the positive condition having higher expectancies (M=5.53) than participants in the negative feedback condition (M=5.17), F (1, 152)=4.58, p < .05. The other main effect showed that male participants had higher overall expectancies (M=5.67) than did female participants (M=5.00), F (1, 152) =14.46, p < .001. This unanticipated sex difference in expectancy—regardless of experimental condition—suggested that sex differences that were found in any of the dependent variables would need to be interpreted with great care.

Hypothesis Tests

Missing Data

When one's objective is to examine the effects of negative versus positive feedback, one needs to be careful about how to “score” the answers of people who elect to provide no answers at all. Specifically, in this research, there were 15 individuals who did not view any case pages. For these people, there were no page hits, no viewing time, and consequently no case answers. If these 15 nonresponders (who nevertheless answered the final questionnaire) had all been in the negative feedback condition, it would be plausible to argue that they were “showing effort effects.” In such a case their data should best be regarded as representing a zero, rather than as representing an absence of data (a missing value). As it happened, however, only 5 of the 15 people with no viewing time were in the negative feedback condition, and 10 were in the positive feedback condition. The comparison of some viewing/no viewing by feedback condition was not significant χ² (df =1)=.71, n.s. Because failure to answer was unrelated to feedback condition, we elected simply to drop the people who had no viewing time from the analyses of total time and page hits.

We predicted that, compared to negative feedback, positive feedback about entrepreneurial propensity would enhance both effort expended and quality of answers. The first prediction was tested on the overall total number of page hits and the total number of minutes spent viewing pages (measures of effort). The measures were treated as repeated measures in a 2 × 2 × 2 (Feedback Condition × Participant Sex × Hits/Time) analysis of variance. The analysis of variance (performed on the 141 individuals who had nonzero viewing time and page counts) revealed no significant differences. Several of the research participants had extreme numbers of minutes spent on the task (as much as 200 minutes or 63 total page views), so we were concerned that these outliers might be distorting the data. To check on this possibility, we converted both the total amount of time spent and the total number of pages viewed to z–scores, then eliminated the six participants whose standardized scores on either variable were greater than 3 standard deviations from the mean. When the repeated measures anova was performed on this slightly reduced data set (N=135), there were still no significant differences attributable to either feedback or participant sex. The same was true when the achievement–related individual difference variables (actions, beliefs, and personal locus of control) were added to the design as covariates or when most–viewed single page was the dependent variable.

The second way to test for effort was to examine the word counts present in respondents’ answers to the open–ended questions. There were five such questions, and because people might have produced different patterns across the five, we treated them as a repeated measure, in a 2 × 2 × 5 (Feedback Condition × Participant Sex × Question) analysis of variance. This analysis also revealed no significant differences based either on feedback condition or on participant sex. As with the other effort measures, we used the achievement–related individual difference measures as covariates. Again, there were no significant differences for either feedback or participant sex. The strongest effect in the analysis was a nearly–significant effect of achievement beliefs on overall number of words produced, with higher achievement beliefs associated with more words produced, F (1, 140)=3.49, p = .064.

Finally, turning to quality of the answers provided, we conducted analyses to determine whether there were differences in judged performance. Because people might have given answers of quite different quality, depending on the question content, we treated this qualitative assessment as a five–element repeated measure, as had been done with the word counts. The 2 × 2 × 5 anova (Feedback Condition × Participant Sex × Answer Quality) revealed no significant differences. To correct for possible individual differences on inherent creativity, we then used the three creativity–related EAQ subscales (Efficiency, Rules, and Originality) as covariates in the anova. The resulting anova also produced no significant differences attributable to feedback condition or participant sex. The only effect that was close to significance was an effect based on one covariate, Rules, with less rule–oriented participants producing answers that tended to be judged of higher quality, F (1, 139) = 3.68, p=.057.

We had included in the final questionnaire an item designed to discover potential disbelief of the feedback manipulation. This item asked respondents to indicate the extent that they believed the results of personality tests they had encountered previously. On the off chance that the results had been compromised by extensive disbelief, we filtered the data set by people's responses to this question, retaining only the 86 people who had scores of 5, 6, or 7 on this 7–point item (with higher scores indicating greater confidence in personality tests). Even among this selected sample, however, there were no statistically significant effects for feedback on 1) times/hits, 2) word counts, or 3) judged quality of answers. We do note that the feedback condition clearly changed people's expectations about their potential for entrepreneurial performance. However, that cognitive change was not reflected in any of the behavioral measures of effort or performance included in the present design.

Discussion AND CONCLUSIONS

The primary objective of this study had been to examine the potential influence of expectancies on performance in a context where those expectancies could be manipulated experimentally, independent of the behavior to which they might apply. Only in an experimental design such as this is it possible to eliminate the possibility that among entrepreneurs, it is their prior success that influences their expectancies, which, in turn, affect their subsequent success. An additional objective of the study had been to include dependent variables that had not been possible to gather in the prior study by Pieterman, Shaver, and Gatewood (1993). Not only did we include counts of page hits and time on task—relatively subtle measures that are available in a Web–based design—we also included a detailed set of ratings of the quality of the answers provided by our research participants.

Unfortunately, the overall findings were disappointing. Analyses of the hypothesized relationships indicated that the type of feedback did not have a significant effect on effort, or on quality of performance, for either male or female participants. Specifically, effort (operationalized as number of pages viewed, amount of time spent viewing, and word counts of subject answers to the case questions) was not affected by either positive or negative feedback received from the EAQ. Nor were there any significant differences for feedback condition on the quality of the answers produced (as rated by the MBA judges).

There was one important bright spot in the otherwise gray picture: Expectancies of success in future business start–ups were changed by the feedback manipulation. Specifically, participants who received negative feedback had significantly lower posttest expectancy for starting a business than those who received positive feedback. So the feedback manipulation did change expectancies, but those changes were not followed by corresponding changes in the dependent variables. Even without differences in performance on our task, this relationship between feedback and expectancies may have implications for entrepreneurial behavior. In our laboratory setting, students were doing something completely familiar to them—Web surfing. True, the surfing had an experimental purpose, and the students were asked to answer questions on material they had read. (Of course, answering questions on material read is also a task that is part of the everyday behavior of students.) It is possible that in the real–world entrepreneurial context, where successful action requires the self–directed seeking and recognition of novel opportunities, expectancy–lowering negative feedback could have serious deleterious effects.

Finally, our results showed that males had higher expectancies than females regardless of feedback type. This was not unexpected, as a number of studies have found significant differences between men and women on entrepreneurial goals, beliefs, and behaviors. For example, Chen, Greene, and Crick (1998) found that female students had lower self–efficacy (effort–performance expectancy) for entrepreneurship than male students. As noted earlier, Kourilsky and Walstad (1998) found that female youths were significantly less interested in starting businesses, less confident in their abilities, and less tolerant of market dynamics than males, findings consistent with earlier research (Crant, 1996; Matthews & Moser, 1996; Scherer, Brodzinski, & Wiebe, 1990). However, Shaver, Gatewood, and Gartner (2001), in their study of nascent and non–nascent entrepreneurs, found that only non–nascent females had less confidence in their ability to start a business. Females who were working on a startup (female nascents) had similar confidence in their abilities as nascent males. One possible explanation for this difference is that nascent females may be increasing their self–confidence through task activity and experience (Lenney, 1977). We know that as experience increases, individuals rely more on attributions to past performance and current motivations as determinants of self–efficacy, i.e., effort–performance expectancies (Gist & Mitchell, 1992).

In a number of ways, the study of entrepreneurial expectancy is uniquely enhanced in a Web–based environment. The traditional constraints of classroom lab–studies are avoided, such as the limited time frame to conduct the study, the requirement of students to report at a particular time and place to participate, and the paper–and–pencil nature of much lab research in the social sciences. The Web–based feature of our study allows for greater flexibility of participation over an extended time period and a multifaceted way of capturing effort not possible in our previous study. So in principle, the present research design should have provided a more comprehensive test of the theoretical relationship between feedback and entrepreneurial performance.

On the other hand, one of the potential explanations for the lack of such feedback effects on effort or quality of performance has to do with the Web–based nature of the research. The high variances in page hits and viewing times suggest the possibility of measurement problems. For example, we learned after the research was concluded that it had been possible for participants to print the pages they were viewing. So a short viewing time might either mean that the participant was inattentive to a particular page, or it might mean that the participant had saved the page in hard copy to read and digest later. These two alternatives would obviously represent great differences in “effort,” but would be indistinguishable in our data. On the other extreme, some of the very long viewing times might have occurred either because participants were concentrating on the material involved, or because they became involved in conversations with their lab neighbors while the page was being displayed. In the future, running the participants in a much more highly controlled setting—without the possibility of printing or distractions—could eliminate this sort of methodological concern.

Apart from the computer–related methodological issues, there are two other possible limitations of our study. First, it is possible that our incentives for participation (class credit and cash prizes for the top three case performers) were not sufficient to stimulate some persons toward anything beyond basic effort, irrespective of their EAQ feedback. In this vein it is worth noting that the total amount of the prizes offered to students was $450, so individual prizes should have been large enough to capture the attention of college undergraduates, even those in business (psychological research with undergraduates routinely produces motivation with amounts that are substantially less). Second, the findings could be a function of the undergraduates’ limited experience with business plans. If one doesn't have sufficient background in finance or operations, no amount of positive feedback can lead one to make detailed and sophisticated suggestions about how to improve the performance of a target company. Some support for this possibility might be found in our earlier study with executive MBA students (Pieterman, Shaver, & Gatewood, 1993). Recall that in that study positive and negative feedback did change word counts, the only effort measure available. Perhaps if participants in the present study had the advantage of years of business experience, the feedback might have had the “headroom” to make a difference.

In addition to these methodological possibilities, there are two theoretically interesting accounts that might explain the lack of performance effects. The first is the notion that entrepreneurs might be particularly immune to negative feedback. In the “real world,” entrepreneurs are typically bombarded with considerable amounts of negative feedback, yet often successfully persevere in venture organizing behaviors. This explanation is supported by Van Eerde and Thierry's (1996) findings. Their meta–analysis of 77 studies on Vroom's (1964) original expectancy models found that expectancy is more related to cognitions (intention and preference) rather than directly to action (performance, effort, and choice) (For other research on this topic, see Gollwitzer, 1993; Kanfer, 1990).

A second theoretically–based explanation for a lack of significance for negative feedback on effort and performance but a significant effect on expectancy might be the result of the depth of processing of the feedback by students. According to Gist and Mitchell (1992), when tasks are novel or particularly salient to the individual, a more in–depth analysis of feedback may occur. However, if tasks are more routine or less important to the individual, a more superficial judgment or automatic process may occur. Under this more automatic analysis, the individual may simply refer to previous performance levels on similar tasks for making expectancy judgments. In this experiment, students may have seen the case assignment as similar to their class work assignments and relied on previous performance levels rather than our feedback for their judgments. The prospect of an entrepreneurial career, however, may have been novel or more salient to students and produced a more in–depth analysis of our feedback. In which case, negative feedback about their entrepreneurial abilities produced a decrease in their entrepreneurial expectancies.

Opportunities for Future Research

Despite the limitations, this study suggests substantial promise for the use of Web–based technology for entrepreneurship research. Because of the tremendous possibility offered by the computer interface for initiating a study synchronously or asynchronously and to virtually anyone in the world with access to the Web, it is our intention to also conduct the study—under more highly controlled experimental conditions—using a sample of real entrepreneurs. If such a study were successful, then it would make sense to attempt to enlist collaborators who would supervise the Web–based participation of entrepreneurs from different countries or cultural backgrounds. In sum, the methodological advantages of our study design enhance our ability at reaching larger and more diverse populations for empirical investigation of entrepreneurial phenomena.

Second, it is clear that replicating the study using actual entrepreneurs, or at least a sizeable group of students with previous experience as entrepreneurs would be a valuable new avenue of inquiry. This study represented an initial attempt at exploring the expectancy—effort—performance linkages, but with a population of persons who were not actual entrepreneurs. Furthermore, given the lack of significance with the hypothesized relationships despite earlier research showing an expectancy–behavior link (Shaver, Gatewood, & Gartner, 2001), future studies of entrepreneurial cognitions along this line of inquiry would be valuable additions to the literature. At this point, our failure to find behavior change, even though there was cognitive change, ought to serve as a caution to those whose only dependent variables might be cognitive (with the assumption that such cognitive change would, almost necessarily, be represented in behavior change had the relevant behavior been assessed).

Third, considering the finding that females, regardless of feedback, had lower expectancies for entrepreneurship than males, it would be interesting to gain a better understanding of the reasons for those beliefs. As noted previously, Shaver, Gatewood, and Gartner (2001) found that women nascents did not hold significantly different entrepreneurial expectancies than male nascents. What direct or indirect experiences or feedback from others during or after college might change females’ perceptions about entrepreneurial expectancies?

Finally, future research using refinements of this methodological approach to the study of entrepreneurship would be helpful. With the advent of Web–based technologies, laboratory research has the potential for considerable enhancement and/or augmentation. With the ability to back–load a wide variety of data based on time and program usage of a Web–based experiment, new and enhanced forms of data collection and analysis can be obtained. The potential of the Web format for studying entrepreneurial phenomena strongly suggests the importance of future work in this medium.

Implications for Practice

There are also implications for practice. Although negative feedback did not affect the study participants’ effort or quality of performance on the immediate task, it did affect their expectancies for starting a business. This result suggests that individuals can be influenced to feel differently about the future, even though it might not affect their immediate actions. Since we know that personal self–confidence and motivation can be undermined by previous failures and self–doubt (Busenitz, 1999; Gatewood & Shaver, 1991), and that entrepreneurs face a series of barriers to venture success, lowered expectancies may, over the long term, result in premature abandonment of a venture, despite the effort or quality of the performance of the entrepreneur. Ilgen and Davis (2000) found that reactions to negative feedback did not necessarily produce improved performance, even when the person was capable of better performance. They suggested that when delivering negative feedback, it was important to communicate in such a way that performers accept responsibility for substandard performance, while not lowering self–concept. Feedback may be more appropriately aimed at the specifics of the venture than at the ability of the entrepreneur, and this may be particularly important when advising female nascent entrepreneurs.

Entrepreneurs experience a myriad of challenges to their beliefs for success, and given the enormous time and effort generally required to achieve their performance goals, opportunities for positive reinforcement is important. For example, communities could establish networks of entrepreneurs so that individuals will not feel as isolated, but instead empowered by persons of like mind to whom they can turn to for advice and encouragement.

Finally, this study suggests that entrepreneurial cognitions are complex processes, and that actions to stimulate entrepreneurial activity need to consider the unique needs of the individual. Thus, although expectancy influences and sex of the entrepreneur may be important considerations when working with entrepreneurs, it is likely that many other factors, internal and external to the entrepreneur, need to be considered as well.

Conclusion

The purpose of our study was to investigate whether feedback concerning entrepreneurial ability influences a person's level of effort and quality of performance for an entrepreneurial task by changing expectancies for the future. Utilizing the unique features of a Web–based lab study design, the research permitted collection of several different indicators of possible effort. Results showed that type of feedback (positive versus negative) received did not affect effort or quality of performance on the entrepreneurial task nor did that outcome differ by sex. The type of feedback did, however, affect expectancies for future entrepreneurial performance. Thus, this article both opens the door to a new methodological approach to entrepreneurial research and is suggestive of ways in which individuals may or may not be stimulated or discouraged from engaging in current or future entrepreneurial activity.

References

Ajzen

(1987).Attitudes, traits, and actions: Dispositional prediction of behavior in social psychology. Advances in Experimental Psychology, 20, 1–63.

Ambrose

M. L.

& Kulik

C. T.

(1999).Old friends, new faces: Motivation research in the 1990s. Journal of Management, 25(3), 231–292.

Bandura

(1982) Self–efficacy mechanism in human agency. American Psychologist, 37(2), 122–147.

Baron

R. A.

(1998).Cognitive mechanisms in entrepreneurship: Why and when entrepreneurs think differently than other people. Journal of Business Venturing, 13, 275–294.

Baron

R. A.

(1999).Counterfactual thinking and venture formation: The potential effects of thinking about “What Might Have Been.”. Journal of Business Venturing, 15, 79–91.

Busenitz

L. W.

(1999).Entrepreneurial risk and strategic decision–making: It's a matter of perspective. The Journal of Applied Behavioral Science, 35, 325–340.

Calver

J. P. O.

& Shaver

K. G.

(1998).Who is on the Fast Trac? Person variables and venture growth following entrepreneurship education Reynolds, P. D., Bygrave, W. D., Carter, N. M., Manigart, S., Mason, C. M., Meyer, G. D., & Shauer, K. G. In Frontiers of entrepreneurship research. 675–676 Wellesley, MA: Babson College.

Campbell

C. A.

(1992).A decision theory model for entrepreneurial acts. Entrepreneurship Theory and Practice, 17(1), 21–28.

Carland

J. W.

, Hoy

& Carland

J. A. C.

(1988).“Who is an entrepreneur?” is a question worth asking. American Journal of Small Business, 12(4), 33–39.

10.

Chen

C. C.

, Greene

P. G.

& Crick

(1998).Does entrepreneurial self–efficacy distinguish entrepreneurs from managers? Journal of Business Venturing, 13, 295–316.

11.

Clements

(1987).The self–efficacy expectations and occupational preferences of females and males. Journal of Occupational Psychology, 23, 1455–1470.

12.

Crant

J. M.

(1996).The proactive personality scale as a predictor of entrepreneurial intentions. Journal of Small Business Management, 34, 42–49.

13.

Das

T. K.

& Bing–Sheng

(1997).Time and entrepreneurial risk behavior. Entrepreneurship Theory and Practice, 22(2), 69–88.

14.

Douglas

E. J.

& Shepherd

D. A.

(2000).Entrepreneurship as a utility maximizing response. Journal of Business Venturing, 15, 231–251.

15.

Earley

P. C.

, Northcraft

G. B.

, Lee

C. L.

& Lituchy

T. R.

(1990).Impact of process and outcome feedback on the relation of goal setting to task performance. Academy of Management Journal, 33(1), 87–105.

16.

Gartner

W. B.

(1988).Who is an entrepreneur? is the wrong question. American Journal of Small Business, 12(4), 11–32.

17.

Gatewood

E. J.

(1993).The expectancies in public sector venture assistance. Entrepreneurship Theory and Practice, 17(2), 91–96.

18.

Gatewood

E. J.

& Shaver

K. G.

(1991). Expectancies for success and attributions for failure: Toward a theory of entrepreneurial persistence. Paper presented at the Academy of Management, Miami, FL.

19.

Gatewood

E. J.

, Shaver

K. G.

& Gartner

W. B.

(1995).A longitudinal study of cognitive factors influencing start–up behaviors and success at venture creation. Journal of Business Venturing, 10, 371–391.

20.

Geier

J. G.

(1967).A trait approach to the study of leadership in small groups. Journal of Communication, 17, 316–323.

21.

George

J. M.

& Jones

G. R.

(1999).Understanding and Managing Organizational Behavior2nd ed.New York: Addison–Wesley.

22.

Gist

M. E.

& Mitchell

T. R.

(1992).Self–efficacy: A theoretical analysis of its determinants and malleability. The Academy of Management Review, 17(2), 183–211.

23.

Gollwitzer

P. M.

(1993).Goal achievement: The role of intentions. European Review of Social Psychology, 4, 141–185.

24.

Hackett

& Betz

N. E.

(1981).A self–efficacy approach to the career development of women. Journal of Vocational Behavior, 18, 326–339.

25.

Harrison

D. A.

(1995).Volunteer motivation and attendance decisions: Competitive theory testing in multiple samples from a homeless shelter. Journal of Applied Psychology, 80(3), 371–385.

26.

Henry

R. A.

& Strickland

O. J.

(1994).Performance self–predictions: The impact of expectancy strength and incentives. Journal of Applied Social Psychology, 24(12), 1056–1069.

27.

Herron

& Sapienza

H. J.

(1992).The entrepreneur and the initiation of new venture launch activities. Entrepreneurship Theory and Practice, 17(1), 49–56.

28.

Hilton

J. L.

& Darley

J. M.

(1991).The effects of interaction goals on person perception. In M.P. Zanna (ed.), Advances in Experimental Social Psychology, 24, 235–267. San Diego: Academic Press.

29.

Iaffaldano

M. T.

& Muchinsky

P. M.

(1985).Job satisfaction and job performance: A meta–analysis. Psychological Bulletin, 97, 251–273.

30.

Ilgen

D. R.

& Davis

C. A.

(2000).Bearing bad news: Reactions to negative feedback. Applied Psychology: An International Review, 49(3), 550–565.

31.

Kanfer

(1990).Motivation theory and industrial and organizational psychology. In M. D. Dunnette & L. M. Hough (eds.), Handbook of industrial and organizational psychology, 2nd ed.,1: 75–170. Palo Alto: Consulting Psychologists Press.

32.

Katz

J. A.

(1992).A psychosocial cognitive model of employment status choice. Entrepreneurship Theory and Practice, 17(1), 29–38.

33.

Katzell

R. A.

& Thompson

D. E.

(1990).Work motivation: Theory and practice. American Psychologist, 45, 144–153.

34.

Kim

& Hunter

(1993).Relationships among attitudes, intentions and behavior. Communication Research, 20, 331–364.

35.

Klein

H. J.

(1989).An integrated control theory of work motivation. Academy of Management Review, 14, 150–172.

36.

Kourilsky

M. L.

& Walstad

W. B.

(1998).Entrepreneurship and female youth: Knowledge, attitudes, gender differences, and educational practices. Journal of Business Venturing, 13, 77–88.

37.

Krueger

N. F.

, Reilly

M. D.

& Carsrud

A. L.

(2000).Competing models of entrepreneurial intentions. Journal of Business Venturing, 15, 411–432.

38.

Lenney

(1977).Women's self–confidence in achievement settings. Psychological Bulletin, 84, 1–14.

39.

Locke

E. A.

(1968).Toward a theory of task motivation and incentives. Organizational Behavior and Human Performance, 3, 157–189.

40.

Locke

E. A.

& Latham

G. P.

(1990).Work motivation and satisfaction: Light at the end of the tunnel. Psychological Science, 1, 240–246.

41.

Maccoby

E. E.

& Jacklin

C. N.

(1974).The psychology of sex differences. Stanford, CA: Stanford Press.

42.

Matthews

C. H.

& Moser

S. B.

(1996).A longitudinal investigation on the impact of family background and gender on interest in small firm ownership. Journal of Small Business Management, 34(2), 29–43.

43.

Miller

K. L.

& Monge

P. R.

(1986).Participation, satisfaction, and productivity: A meta–analytic review. Academy of Management Journal, 29, 727–753.

44.

Neuberg

S. L.

(1994).Expectancy–confirmation processes in stereotype–tinged social encounters: The moderating role of social goals. In M. P. Zanna & J. M. Olson (eds.), The psychology of prejudice: The Ontario symposium, Hillsdale, NJ: Lawrence Erlbaum Associates. vol. 7, pp. 103–130.

45.

Nevill

D. D.

& Schlecker

D. I.

(1988).The relation of self–efficacy and assertiveness to willingness to engage in traditional/nontraditional career activities. Psychology of Women Quarterly, 12(1), 91–98.

46.

Nicholson

(1998).Personality and entrepreneurial leadership: A study of the heads of the UK's most successful independent companies. European Management Journal, 16, 529–535.

47.

Oettingen

(2000).Expectancy effects on behavior depend on self–regulatory thought. Social Cognition, 18(2), 101–129.

48.

Olson

J. M.

, Roese

N. J.

& Zanna

M. P.

(1996).Expectancies. In E. T. Higgins & A. W. Kuglanski (eds.)Social psychology: Handbook of basic principles, 211–238. New York: Guilford Press.

49.

Pieterman

M. A.

, Shaver

K. G.

& Gatewood

E. J.

(1993).Expectations and entrepreneurial persistence. In Frontiers of entrepreneurship research. Wellesley, MA: Babson College.

50.

Rasch

R. H.

& Tosi

H. L.

(1992).Factors affecting software developers’ performance: An integrated approach. MIS Quarterly, 16, 395–413.

51.

Robbins

S. P.

(1998).Organizational behavior. 8th ed.Upper Saddle River, NJ: Prentice Hall.

52.

Roberts

& Nolen–Hoeksema

(1994). Gender comparisons in responsiveness to others’ evaluations in achievement settings. Psychology of Women Quarterly, 18, 221–240.

53.

Rotter

J. B.

(1954). Social learning and clinical psychology. Eaglewood Cliffs, NJ: Prentice Hall.

54.

Scherer

R. F.

, Brodzinski

J. D.

& Wiebe

F. A.

(1990).Entrepreneurial career selection and gender: A socialization approach. Journal of Small Business Management, 28(2), 37–44.

55.

Shapero

(1982).Social dimensions of entrepreneurship. In C. Kent, D. Sexton, & K. Vesper (eds.), The encyclopedia of entrepreneurship, 72–90. Englewood Cliffs, NJ: Prentice–Hall.

56.

Shaver

K. G.

(1995).The entrepreneurial personality myth. Business and Economic Review, 41(3), 20–23.

57.

Shaver

K. G.

, Gartner

W. B.

, Gatewood

E. J.

& Vos

L. H.

(1996).Psychological factors in success at getting into business Reynolds, P. D., Birley S., Butler, J. E., Bygrave, W. D., Davidson, P., Gartner, W. B., & McDougall, P. P. In Frontiers of entrepreneurship research. 77–90 Wellesley, MA: Babson College.

58.

Shaver

K. G.

, Gatewood

E. J.

& Gartner

W. B.

(2001). Differing expectations: Comparing nascent entrepreneurs to non–entrepreneurs. Paper presented at the Academy of Management, Washington, D.C.

59.

Shaver

K. G.

& Scott

L. R.

(1991).Person, process, choice: The psychology of new venture creation. Entrepreneurship Theory and Practice, 16(2), 23–45.

60.

Shepperd

J. A.

& Taylor

K. M.

(1999).Social loafing and expectancy–value theory. Personality and Social Psychology Bulletin, 25(9), 1147–1158.

61.

Simon

, Houghton

S. M.

& Aquino

(2000).Cognitive biases, risk perception, and venture formation: How individuals decide to start companies. Journal of Business Venturing, 15, 113–134.

62.

Smilor

R. W.

(1997).Entrepreneurship reflections on a subversive activity. Journal of Business Venturing, 12, 341–346.

63.

Van Auken

H. E.

(1999).Obstacles to business launch. Journal of Developmental Entrepreneurship, 4, 175–187.

64.

Van Eerde

& Thierry

(1996).Vroom's expectancy models and work–related criteria: A meta analysis. Journal of Applied Psychology, 75, 68–76.

65.

Vesper

K. H.

(1996).New venture experience. Seattle: Vector Books.

66.

Vroom

V. H.

(1964).Work and motivation. New York: John Wiley.

67.

Whitley

B. E.

, Jr., McHugh

M. C.

& Frieze

I. H.

(1986).Assessing the theoretical models for sex differences in causal attributions of success and failure. In J. S. Hyde & C. C. Linn (eds.), The psychology of gender: Advances through meta–analyses. 102–135. Baltimore: The John Hopkins University Press.

68.

Wortman

M. S.

(1987).Entrepreneurship: An integrating typology and evaluation of the empirical research in the field. Journal of Management, 13, 259–279.

69.

Yukl

& Van Fleet

D. D.

(1992).Theory and research on leadership in organizations. In M. D. Dunnette & L. M. Hough (eds.), Handbook of industrial & organizational psychology. 150. Palo Alto, CA: Consulting Psychologists.