Self-Monitoring in Leadership Emergence Among Student-Athletes and Greek Members

The study examined links between self-monitoring behavior and leadership in two main settings: Greek organizations and athletic teams. In the study, 141 undergraduate examinees were given Snyder’s 25-Item Self-Monitoring Scale and a questionnaire regarding their leadership experiences in either the Greek organization or athletic team. Independent sample t-tests showed no significant differences. However, leaders in Greek organizations tended to have higher self-monitoring scores than did nonleaders in the same organizations. Athletic leaders tended to have lower scores than did nonleaders on their teams. Also, a gender difference, with males exhibiting higher levels of selfmonitoring than females, approached significance. INTRODUCTION The Self-Monitoring Construct Mark Snyder originated the social-psychological construct of self-monitoring as part of a doctoral dissertation (Snyder, 1972). Today, self-monitoring is one of the most widely studied personality constructs in psychology. Self-monitoring refers to the extent to which people change, modify or adapt their behavior in different social settings. High self-monitors are very responsive to perceptual cues in their environment. Consequently, high self-monitors have the ability to adapt their behavior to best fit their current situation. Those who exhibit high self-monitoring behavior tend to strive to fit into every social setting by regulating their behavior according to the particular situation (Snyder, Berscheid, & Matwychuk, 1988). Low self-monitors are not as sensitive to social cues as Self-Monitoring and Leadership 3 high self-monitors (Hazer & Jacobson, 2003). As a result, low self-monitors rely more on their internal feelings and appraisals for guidance in social situations. The 25 items on the scale were selected from 41 pilot items originally written (Snyder, 1974). These items assess five main areas of self-monitoring, as defined by Snyder: “concern with the social appropriateness of one’s self-presentation,” “attention to social comparison information as cues to appropriate self-expression,” “the ability to control and modify one’s self-presentation and expressive behavior,” “the use of this ability in particular situations,” and “the extent to which the respondent’s expressive behavior is cross-situationally consistent or variable (Snyder, 1974, pg. 529).” Later, Gangestad and Snyder (1985) removed seven items, increased the reliability of the measure, and introduced the 18-Item Self-Monitoring Scale. In the first sample (n=533) taken in 1972, the distribution had a mean of 12.51. In addition, the median was 12, standard deviation of 4.11, and a range of 2 to 24 (Snyder, 1974). A further sample 10 years later corroborated the finding of the first study. Self-Monitoring: Construct Validity In a series of articles, Snyder (1987; Gangestad & Snyder, 1985) and others (Ickes, Reid & Patterson, 1986) sought to show the validity of the construct by proving self-monitoring is distinctly different from need for approval, Machiavellianism, and extraversion. Seeking approval by giving a socially desirable response is quite different than self-monitoring behavior. High self-monitors use their ability to detect social cues to form their behavior to a particular situation. A person with a high need for approval may or may not have this ability. Snyder (1987) reports scores on the Marlowe-Crowne Self-Monitoring and Leadership 4 measure (an instrument used to measure need for approval) are not correlated with scores on the 25-Item Self-Monitoring Scale. Only one of the five dimensions of selfmonitoring as defined by Snyder (1974), “concern for the social appropriateness of one’s actions,” is related to need for approval. The other four dimensions measure and take into account aspects of self-monitoring behavior not related to need for approval (Snyder, 1974). Another concept that must be differentiated from self-monitoring is Machiavellianism. These two concepts do differ. High Machiavellians rely completely on manipulation, while the high self-monitor relies on impression management (Ickes, Reidhead & Patterson, 1986). Other evidence, including the tendency for high Machiavellians to use first-person singular pronouns more than high self-monitors lends credence to this argument. Extraversion is the hardest concept to differentiate from self-monitoing. Gangestad and Snyder (2000) acknowledge the tendency to see a relationship between self-monitoring and extraversion. Extraversion is a widely studied and believed to apply to most social situations. However, some important differences between high selfmonitors and extraverts do exist. While extraverts may always be friendly, nice, and outgoing in all social situations, high self-monitors may not behave in such a way if the situation calls for a more reserved demeanor (Snyder, 1987). The stability of social behavior is one of the dimensions of self-monitoring identified by Snyder. The behaviors of a high self-monitor are not stable across all situations, which is in contrast to the extravert. Self-Monitoring and Leadership 5 Self-Monitoring and Emergent Leadership Bass (1981) differentiates between emergent leadership from those who are appointed or elected to leadership positions. Early leadership theorists believed leaders emerged out of interaction with a group. This kind of leadership is most easily seen in a group of random people who start out with equal status. Eventually, one or two people come to the forefront, and are recognized by other group members as those best suited to lead the group in attaining its desired goal (Bass, 1981). The traits possessed by these emergent leaders are of interest to researchers. Following the introduction of factor analysis by Cattell, clusters of personality traits were determined. Currently, most theorists agree five main traits are responsible for most individual leadership differences (Kassin, 2004). Neuroticism, extraversion, openness, agreeableness, and conscientiousness are included in the NEO Five Factor Inventory (Costa & McCrae, 1992). Extraversion, openness, and agreeableness appear to be the traits in the NEO-FFI most closely related in definition to self-monitoring. The way in which self-monitoring affects emergent leadership in small group settings has been extensively studied. Day, Unckless, Hiller and Schleicher (2002) found high self-monitors were more likely to emerge as leaders in work settings than were low self-monitors. This finding substantiates the results of other research (Cronshow & Ellis, 1991; Ellis, 1988; Ellis & Cronshaw, 1992; Kent & Moss, 1990) which found similar results. Cronshaw and Ellis (1991) found both high and low self-monitors are capable of becoming leaders. Once in leadership positions, high self-monitors rely on their ability to interpret social cues while low self-monitors rely more on personal attitudes. Ellis and Cronshaw (1992) also wanted to see what part of the high self-monitoring personality led Self-Monitoring and Leadership 6 to leadership emergence. They found flexibility and adaptation skills, and not social sensitivity, had an effect on leadership emergence. Ellis (1988) found high self-monitors were rated higher as leaders by group members than were low self-monitors. Finally, Kent and Moss (1990) reported high self-monitors were more likely to self-report leadership emergence than were low self-monitors. Zaccaro, Foti, and Kenny (1991) took a more trait-based approach to leadership than the one suggested by Stogdill (Bass, 1981). While Stogdill advocated the strength of situation and the interaction between people as predictors of emergent leadership, Zaccaro et al. (1991) believed leadership resulted more from flexibility. Of course, high self-monitors would be especially well suited for leadership should this be the case. Zaccaro and his colleagues found “trait-based variance in leadership may be due to social perceptiveness and response flexibility,” tasks where high self-monitors could be expected to thrive (Zaccaro et al., 1991, pg. 308). In the same line of research, Dobbins, Long, Dedrick, and Clemons (1990) concluded self-monitoring appears to be the trait that holds steady in leadership across different situations. Gender Differences in Self-Monitoring Results of several studies studies suggest sex differences in self-monitoring exist. In his first analysis of the construct, Snyder (1972) found males, on average, have a slightly higher score (12.78) than females (12.08). This difference, however, is not statistically significant, even with a sample size of over 500. Kumru and Thompson (2003) studied 476 adolescents in Turkey and found males had significantly higher selfmonitoring scores than did females. Gender role differences between Turkey and other nations may limit one’s ability to generalize these results. Frazier and Fatis (1980) had Self-Monitoring and Leadership 7 results similar to those of Snyder (1972). Of 252 undergraduate students, males were found to have higher self-monitoring scores than did the examined females. In a metaanalysis of 136 independent samples with a collective sample size of 23,191, Day et al. (2002) found a .12 correlation between sex and self-monitoring, with males having higher scores than females. Broderick and Beltz (1996) reported results inconsistent with the rest of the literature. They found females had higher self-monitoring scores than their male counterparts. Again, such research is limited by the age (pre-adolescent) of the subjects. However, the majority of the literature does suggest males have higher scores on average than do females. Self-Monitoring and Leadership Independent of Sex The sex differences that may exist between males and females manifest themselves in leadership emergence as well. A great deal of research has shown that males almost always emerge as leaders when in mixed sex dyads (Megargee, 1969; Nyquist & Spence, 1986). Because men emerge as leaders more than women, and men tend to have higher self-monitoring scores, is leadership emergence more closely correlated with sex or self-monitoring? The confound between sex and leadership was examined by Garland and Beard (1979). Garland and Beard invited participants into the lab, and randomly generated same sex (both male and female) and mixed sex dyads. In same sex pairings, high selfmonitoring females emerged as leaders significantly more than did low self-monitoring females. This same relationship was not substantiated for either all male pairings or mixed sex pairings. These finding suggest that self-monitoring by itself does play a role in leadership emergence, at least for females among other females. A subsequent study Self-Monitoring and Leadership 8 (Ellis & Cronshaw, 1992) examined other variables that may influence leadership. Selfmonitoring was a significant factor in leader emergence in males only. Self-Monitoring, Advertising and Personnel Selection The high level of interest may be due to the numerous areas where selfmonitoring seems to have influence. Among areas where researchers have found selfmonitoring to have an effect are advertising (DeBono & Snyder, 1989; Snyder & DeBono, 1985) and personnel selection (Hazer & Jacobson, 2003; Snyder, Berscheid, & Matwychuk, 1988). In one study, DeBono and Snyder (1989) showed how high selfmonitors respond more to advertisements that stress appearance over function. Subjects were asked to choose between a sportylooking car and a rather regular-looking car. As a group, the high self-monitors were more likely to pick the car with a nicer appearance than were low self-monitors, who responded better to claims of quality and functionality. In another study, Snyder and DeBono (1985) presented subjects with claims from an advertiser about a new shampoo going on the market. One version of the ad emphasized how well the shampoo made hair look, while the other emphasized the cleaning ability of the shampoo. High self-monitors reported they were more likely to try the shampoo when the advertisement emphasized appearance, and low self-monitors were more likely to use it when the advertisement focused on cleanliness (Snyder & DeBono, 1985). Self-monitoring also appears to play a role in personnel selection. In a series of studies, Snyder, Berscheid, and Matwychuk (1988) showed high self-monitors, playing the role of a human resource manager, placed more weight on the physical attractiveness and compatability of that physical appearance with the job than did low self-monitors. That is, high self-monitors were more likely to recommend hiring a person Self-Monitoring and Leadership 9 with the physical qualifications alone than were low self-monitors. The low selfmonitoring HR managers recommended hiring based more on qualifications than on appearance. THE CURRENT STUDY Self-monitoring as a predictor of leadership in a small group setting has been extensively studied. As previously mentioned, the bulk of the research examines emergent leadership. The current study examines a different kind of leadership. The leaders in both the Greek organizations and athletic teams were either elected or appointed to their positions; they did not emerge at random from a group of people. That is, this study examines leadership traits in a reallife setting rather than an experimental one. Greek organizations and athletic teams are different from small, randomly assigned groups and may present a different dynamic and opportunity for self-monitoring to affect leadership. In addition to studying leadership, the study examines self-monitoring as a predictor of advertising preferences. METHODS Design Leadership and sex were examined as predictors of self-monitoring. The subjects were involved in either a Greek organization or an intercollegiate athletic team, and reported themselves as either leaders or nonleaders. For the study, Greek leaders were defined as those people who indicated on the questionnaire they had held a leadership position in their organization. For athletes, leadership was defined as having been Self-Monitoring and Leadership 10 selected as a captain for their team. Such a designation takes place for both setting conditions. Both males and females were included in the study. Participants Subjects were 141 undergraduate students from a small liberal arts university in the Midwest. 105 of the subjects were affiliated with a Greek organization on the university campus. Of those 105, 54 reported holding a leadership position in the organization. 56 of the subjects were on an intercollegiate athletic team at the time of the study. 14 reported having been elected captain during their careers. The other 42 reported they had not yet been elected as a captain. Stimulus Materials Self Monitoring Scale. Snyder first developed the 25-Item Self-Monitoring Scale (see Appendix 1) in 1972. Other scales have since been developed, most notably the LennoxWolfe scale (Lennox & Wolfe, 1984). Subsequently, an 18-Item Scale was developed (Gangestad & Snyder, 1985). However, due the ready availability of psychometric data for the 25-Item Scale, it was used for the analysis. Snyder (1987) reported a .66 internal consistency estimate for the 25-Item Self-Monitoring Scale. Personal Questionnaire. The personal information questionnaire asks subjects about their demographics and leadership experience (see Appendix 2). Before completing the two questionnaires, individuals in the study read and signed an Informed Consent Form (see Appendix 3). The main goal of the personal questionnaire is to ascertain into which condition the subjects should be placed. Selfreported leadership and athletic/social information is the most important. The personal questionnaire does not require any responses that may reveal the identity of the subject. For example, the examinee is asked Self-Monitoring and Leadership 11 if they are involved in a Greek organization, but are not asked to identify with which organization they are affiliated. Attitudes Survey. This questionna ire asked participants to rate two different vehicles on quality and appeal (Appendix 4). The Honda Element ad emphasized the functionality of the vehicle, while the Jaguar XK ad appealed to the form and looks of the vehicle. Both ads were obtained from the manufacturer’s websites (Honda, 2004; Jaguar, 2004). Each of the vehicles was rated on quality and appeal on a 1-10 Likert scale. To assess order effects, two forms of the attitude survey were produced: One form had the Element ad first, the other had the Jaguar ad first. Procedure The three surveys were stapled together in the following order; Self-Monitoring Scale, personal questionnaire, attitude survey. This packet was presented to the subject in a plain manilla file folder. Included in the folder was an unattached Informed Consent Form, which was filled out first, handed in, and placed in a separate collection folder to ensure anonymity. The subjects then completed the measures and returned them. At this point, no identifying marks were present aside from a three digit coding number in the upperleft hand corner on the front page of the packet. The 25-Item Self-Monitoring Scales were returned, along the personal questionnaire, to the examiner. Using an answer key, the self-monitoring scales were hand scored and entered into an SPSS data field according to code number. Data regarding gender and leadership experiences (acquired from the personal questionnaire) were also entered. Self-Monitoring and Leadership 12 The data were collected in a variety of settings. For those examinees involved in Greek organizations, the surveys were distributed and completed during chapter meetings during which all members were present. For all other subjects, the surveys were completed in the campus houses or dormitories of the participants. HYPOTHESES Effects on Leadership I believe the leaders examined in this study will have significantly higher selfmonitoring scores than nonleaders in their respective groups. Such a finding would be consistent with research showing high self-monitors tend to emerge as leaders in ambiguous situations. If one group (either Greek organizations or athletic teams) were to have an effect and not the other, I believe the difference between self-monitoring between Greek leaders and non-leaders would be greater than the difference between athletic leaders and nonleaders due to the high degree of social interaction present in a Greek house setting. Effects on Gender I predict the results of my research will support the majority of previous research showing male examinees tend to have slightly higher self-monitoring scores than do female examinees (Day et al., 2003; Frazier & Fatis, 1980; Kumru & Thompson, 2003; Snyder, 1972). However, it is not predicted that these differences will be statistically significant. Rather, a trend towards the relationship is expected. Effects of Self-Monitoring on Leadership Independent of Gender. Like Garland and Beard (1979), I believe I will find self-monitoring is a predictor of leadership Self-Monitoring and Leadership 13 independent of gender. For females, a significant effect will be found, with leaders having higher self-monitoring scores than non-leaders. For males, a significant effect is not expected. Self-Monitoring and Advertisements I predict the results will resemble those of Snyder and DeBono (1985). That is, high self-monitors will give higher ratings to the ad emphasizing form and low selfmonitors will give higher ratings to the ad emphasizing function. RESULTS For the sample collected in this study (n=141), a mean self-monitoring score of 11.465 and a standard deviation of 3.59 was found. Both of these figures are below the results found by Snyder, who found a mean of 12.51 and a standard deviation of 4.11 (Snyder, 1972). Table 1 shows number of subjects, means, and standard deviations for all groups. Independent samples t-tests were performed between various groups to examine differences between the groups. T-tests were done for the following comparisons: Greek leaders v. Greek nonleaders, athletic leaders v. athletic non-leaders, and males v. females. For Greek leaders vs. Greek nonleaders, t(104)=0.736, p=0.463. This result was not significant, though it was in the predicted direction. For athletic leaders vs. athletic nonleaders, t(55)=-1.434, p=0.157. Again, this result was not significant. Also, it was not predicted that nonleaders would have higher self-monitoring scores than leaders. For males (mean=12.135) vs. females (mean=11.073), t(139)=1.706, p=0.090. This test was not significant, although it did approach significance. Because of the Self-Monitoring and Leadership 14 potential confounds between gender and leader emergence, independent sample t-tests were conducted on single-sex samples. For females, t(89)=1.132, p=.261. This result was not significant, but was in the hypothesized direction. For males only, t(49)= -0.351, p=.526. This result also was not significant. An analysis of variance was conducted on the data obtained from the attitude survey (Table 2). All major results were insignificant. However, a main effect for rating was found. Participants, regardless of self-monitoring level, gave higher average “quality” ratings (mean=7.883) than they did “appeal” ratings (mean=7.593). Figure 3 shows the mean differences between the ratings. DISCUSSION I proposed self-monitoring behavior is a significant factor in leadership emergence and election. The results, though not significant, suggest self-monitoring behavior is related to emergent leadership. However, because athletic leadership did not have the expected relationship to self-monitoring, that part of the hypothesis was not confirmed. The hypotheses for genderindependent leader emergence and selfmonitoring were not supported by the results, as they were not significant. However, the trends were in the predicted direction. Leadership Differences Though no significant effects were found in self-monitoring between leaders and nonleaders, the results are interesting nonetheless. The direction of the relationship between Greek leaders and nonleaders (leaders having higher self-monitoring scores than non-leaders) was consistent with the hypothesis and previous research (Cronshow & Ellis, 1991; Day et al., 2002; Ellis, 1988; Ellis & Cronshaw, 1992; Kent & Moss, 1990). Self-Monitoring and Leadership 15 However, the relationship between athletic leaders and non-leaders was not consistent with the hypothesis, and was actually in the opposite direction. For athletes, though not significant, the results suggest leaders tend to have lower self-monitoring scores than nonleaders. Gender Differences While no significant effects were found for gender, the results do help confirm the trend for men to have slightly higher scores than women found in previous studies (Day et al., 2002; Frazier & Fatis, 1980; Kumru & Thompson, 2003; Snyder, 1972). However, some studies have found contrasting results (Broderick & Beltz, 1996). The reason for the apparent gender difference in self-monitoring is not well researched. A large scale study or meta-analysis looking into this gender difference may help to better explain the social-psychological construct of self-monitoring. In addition, showing self-monitoring affects leadership, independent of sex, may help enhance construct validity. Studies examining gender differences in leader emergence that do not take selfmonitoring into account have been conducted (Margargee, 1969). Because these studies find that men tend to emerge as leaders more than women, and men tend to have higher self-monitoring scores than women, the question arises: Is gender or self-monitoring more closely related to leader emergence? Judging from previous research, it appears leadership and sex are truly confounded. Research suggests men have consistently higher self-monitoring scores and consistently emerge as leaders more frequently than women. Thus, it is impossible to determine what amount of the variance in leadership emergence is caused by self-monitoring behavior and what amount is caused by gender differences in self-monitoring behavior. One method for getting around this confound is to conduct a Self-Monitoring and Leadership 16 study that looks at self-monitoring and leadership independent of gender. Indeed, studies using a single-sex methodology, such as that of Garland and Beard (1979) should be conducted. Self-Monitoring and Leadership Independent of Sex The results of independent samples t-tests conducted on same sex samples were not significant. However, the trend was in the expected direction for females. That is, female leaders had higher self-monitoring scores than did female nonleaders. For males, leaders had slightly lower self-monitoring scores than did non-leaders. If these results were significant, they would confirm the results reported in other studies (Garland & Beard, 1979). A difference between this study and the Garland and Beard study is way leadership is detected. The current study uses a self-report questionnaire, while the previous study uses an experimental random pairing of individuals. CONCLUSIONS Differences Between Athletic and Greek Leaders Why do Greek leaders and emergent leaders tend to exhibit higher selfmonitoring behavior than nonleaders? One explanation could be the flexibility and adaptive skills inherent in the high self-monitoring personality are important traits for success as a leader. This would dovetail with research done by Ellis and Cronshaw (1992). The literature does not suggest that high self-monitors seek out leadership positions, but when placed in ambiguous situations they often emerge as leaders. The disparity between leadership in Greek organizations and leadership on athletic teams may suggest leadership in each are tied to different personality traits or are situation-specific. The results, if they were significant, would suggest the leadership Self-Monitoring and Leadership 17 required in Greek organizations may be more closely related to the “emergent leadership” discussed in the literature than is athletic leadership. Athletic leadership may be something completely different than Greek or emergent leadership. While athletic leaders do “emerge”, it seems they do so over a much longer period of time. Most captains are not elected until their senior year. During this time, perhaps the initial advantage possessed by high self-monitors (due to their increased flexibility and adaptive ability) is lessened. Common sense would say most athletic leaders are chosen on a number of dimensions, including athletic skill. It is possible these factors have a limiting effect on self-monitoring. Comfort in their role, and self-confidence gained through athletic achievement may be also be desirable traits in athletic leaders. Reliability Issues One major criticism of Snyder’s Self-Monitoring Scales (Snyder, 1974, 1985) is an apparent lack of reliability. Day et al. (2002) found both Snyder’s 18-Item Scale (.73) and the Lennox and Wolfe 13-Item Scale (.81) had higher reliabilities than the 25-Item Scale (.71) used in the current study. This difference is significant given that these scales are both shorter in length than is the 25-Item Scale. Thus, the results of the study may have been weakened by the use of a relatively unreliable scale. In addition, Day et al. (2002) examined differences in reliabilities between dichotomous scoring (true-false) techniques and continuous scoring techniques. They found the continuous scoring scales had a higher average reliability (.77) than dichotomous scoring scales (.71). Self-Monitoring and Leadership 18 Future Research Future studies in this area would be very helpful in locating and establishing potential leaders. One area to examine would be the correlations and similarities between athletic leaders and items on the NEO-FFI Scale (Costa & McCrae, 1992). As previously discussed, the relationship between athletic leadership and self-monitoring may not correspond to previous research because the traits necessary for athletic leadership may not be the same as the emergent leadership emphasized in the literature. In addition, future research in the area should use an established leadership scale, such as the Campbell Leadership Index or Leadership Appraisal Survey. With the data obtained through these measures, correlations could be drawn between traits associated with leadership and self-monitoring. Such a technique would be especially useful in measuring leadership in a non-experimental setting. Another avenue for further study in the area of self-monitoring and leadership could be an adjective survey given to nonleaders in both Greek organizations and on athletic teams. Nonleaders would indicate the adjectives that describe their leaders, and statistical tests could be conducted to determine if non-leaders from both groups indicate separate sets of traits or any overlap that may occur. Limitations and Methodological Issues A relatively homogeneous sample would reduce one’s ability to generalize the results across other groups if the results had been significant. For example, the participants in this study had proportionately more Greek women than any other subgroup. Had the results of this study been significant, this over-emphasis on this Self-Monitoring and Leadership 19 demographic would have limited the ability to generalize this study across other demographics. Other methods to analyze the collected data could also be used. For example, the personal questionnaires contained a fillin space for Greek leaders. Any subjects who indicated they held leadership position were defined as leaders. This broad definition of leadership in Greek organizations led to 51.43% (54 out of 105) of Greeks being defined as leaders. If only those at the very top of Greek organizations (e.g. Presidents and VicePresidents) were defined as leaders, perhaps the results of the study would be different. Finally, with questions about reliability surrounding the 25-Item Scale, the seven items eliminated by Snyder could also be eliminated and the scores readjusted. In light of the results obtained in the meta-analysis by Day et al. (2002), perhaps the Lennox-Wolfe scale or a continuous scoring scale should be utilized. Self-Monitoring and Leadership 20 ReferencesBass, B. (1981). Stogdill’s Handbook of Leadership. New York: Free Press.Broderick, P., & Beltz, C. (1996). The contributions of self-monitoring and gender topre-adolescents’ friendship expectations. Social Behavior and Personality, 24,34-45.Costa, P. & McCrae, R. (1992). The Revised NEO Personality Inventory professionalmanual. Odessa, FL: Psychological Assessment Resources.Cronshaw, S. & Ellis, R. (1991). A process investigation of self-monitoring andleadership emergence. Small Group Research, 22, 403-420.Day, D., Schleicher, D., Unckless, A., & Hiller, N. (2002). Self-monitoring personalityat work: A meta-analytic investigation of construct validity. Journal of AppliedPsychology, 87, 390-401.DeBono, K. & Snyder, M. (1989). Understanding consumer decision-making processes:The role of form and function in product evaluation. Journal of AppliedPsychology, 19(5), 416-421.Dobbins, G., Long, W., Dedrick, E. & Clemons, T. (1990). The role of self-monitoringand gender on leader emergence: A laboratory and field study. Journal ofManagement, 16, 609-618.Ellis, R. (1988). Self-monitoring and leadership emergence in groups. Personality andSocial Psychology Bulletin, 14(4), 681-693.Ellis, R. & Cronshaw, S. (1992). Self-monitoring and leader emergence: a test ofmoderator effects. Small Group Research, 23, 113-129. Self-Monitoring and Leadership 21 Gangestad, S. & Snyder, M. (1985). “To carve nature at its joints”: On the existence ofdiscrete classes in personality. Psychological Review, 92, 317-349.Gangestad, S. & Snyder, M. (2000). Self-monitoring: Appraisal and reappraisal.Psychological Bulletin,126(4), 530-555.Garland, H. & Beard, J. (1979). Relationship between self-monitoring and leaderemergence across two task situations. Journal of Applied Psychology, 64(1), 77-81.Hazer, J. & Jacobson, J. (2003). Effects of screener self-monitoring on the relationshipsamong applicant positive self-presentation, objective credentials, andemployability ratings. Journal of Management, 29(1), 119-138.Honda Corporation. (2004, January 8). Where to now? Retreived fromhttp://www.hondacars.com/models/model_overview.asp?ModelName=Element.Ickes, W., Reidhead, S. & Patterson, M. (1986). Machiavellianism and self-monitoring:As different as “me” and “you.” Social Cognition, 4(1), 58-74.Jaguar Corporation. (2004, January 8). The 2004 XK. Retrieved fromhttp://www.jaguarusa.com/us/en/xk/overview/XK%20Overview.htm.Kassin, S. (2004). Psychology: 4 Edition. Upper Saddle River, New Jersey. PrenticeHall Publishing.Kent, R. & Moss, S. Self-monitoring as a predictor of leadership emergence.Psychological Reports, 66, 875-881.Kumru, A. & Thompson, R. (2003). Ego identity status and self-monitoring behavior inadolescents. Journal of Adolescent Research, 18, 481-495. Self-Monitoring and Leadership 22 Lennox, R. & Wolfe, R. (1984). Revision of the self-monitoring scale. Journal ofPersonality and Social Psychology, 46(6), 1349-1364.Margargee, E. (1969). Influence of sex roles on the manifestation of leadership. Journalof Applied Psychology, 53, 377-382.Nyquist, L. & Spence, T. (1986). Effects of dispositional dominance and sex roleexpectations on leadership behaviors. Journal of Applied Psychology, 50, 87-93.Snyder, M. (1972). Individual differences and the self-control of expressive behavior(Doctoral dissertation, Stanford University, 1972). Dissertation AbstractsInternational, 33, 4533-4534.Snyder, M. (1974). Self-monitoring of expressive behavior. Journal of Personality andSocial Psychology, 30(4), 525-37.Snyder, M. & DeBono, K. (1985). Appeals to image and claims about quality:Understanding the psychology of adve rtising. Journal of Personality and SocialPsychology, 49(3), 586-597.Snyder, M. (1987). Public Perceptions, Private Realities: The psychology of self-monitoring. New York: W.H. Freeman and Company.Snyder, M., Berscheid, E., & Matwychuk, A. (1988). Orientations towards personnelselection: Differential reliance on appearance and personality. Journal ofPersonality and Social Psychology, 54(6), 972-979.Zaccaro, S., Foti, R., & Kenny, D. (1991). Self-monitoring and trait-based variance inleadership: An investigation of leader flexibility across multiple group settings.Journal of Applied Psychology, 76(2), 308-315. Self-Monitoring and Leadership 23 Table 1Table of Means for Groups and Subgroups GroupSubgroup N Mean SM Score SDGreektotal10511.329 3.68leaders5411.583 3.62non-leaders 5111.059 3.76Athletes total5711.573.57leaders1510.367 3.78non-leaders 4212.048 3.34Gender males5212.135 3.24females 8911.073 3.73Males only leaders1611.938 3.11non-leaders 3612.286 3.37Females leaders4311.535 3.55non-leaders 4810.655 3.62TOTAL141* 11.465 3.59* Total N of columns do not sum 141 because somesubjects were in more than one group. Table 2Analysis of Variance Source Sum ofSquaredfMean Square F Sig. CAR 1.807 11.807 .515 .474CAR * SEX .424 1.424 .121 .729CAR * HIGHLOW 1.851 11.851 .527 .469CAR * SEX * HIGHLOW 4.219 14.219 1.202 .275Error(CAR) 449.415 1283.511RATING 10.363 1 10.363 6.567 *.012RATING * SEX .418 1.418 .265 .608RATING * HIGHLOW 4.685E-03 1 4.685E-03 .003 .957RATING * SEX * HIGHLOW .684 1.684 .433 .512Error(RATING) 201.985 1281.578CAR * RATING 2.387 12.387 1.221 .271CAR * RATING * SEX 5.758E-02 1 5.758E-02 .029 .864CAR * RATING * HIGHLOW 4.702E-02 1 4.702E-02 .024 .877CAR * RATING * SEX *HIGHLOW1.222 11.222 .625 .431 Error(CAR*RATING) 250.262 1281.955*Indicates a significant result Self-Monitoring and Leadership 24 Figure 1. Self-Monitoring Scores of Greeks and Athletes, Leaders and Non-Leaders 9.51010.51111.51212.5 leaders non-leaders leaders non-leadersGreekAthletesSelf-MonitoringScore Greek leadersGreek non-leadersAthletes leadersAthletes non-leaders Figure 2. Self-Monitoring Scores and Leadership Emergence by Gender 9.51010.51111.51212.5 malesfemalesleadersnon-leadersleadersnon-leadersGenderMales onlyFemalesSelf-MonitoringScore Gender malesGender femalesMales only leadersMales only non-leadersFemales leadersFemales non-leaders Self-Monitoring and Leadership 25 Figure 3. Mean Ratings for Quality and Appeal