Connecting Gender and Mental Health to Imposter Phenomenon Feelings Claire E

Connecting Gender and Mental Health to Imposter Phenomenon Feelings Claire E. Cusack, Jennifer L. Hughes*, and Nadi Nuhu Agnes Scott College

ABSTRACT. The impostor phenomenon (IP) occurs when high-achieving individuals attribute their successes to external factors and are unable to internalize success (Clance & Imes, 1978). Previous data show correlations between this phenomenon with gender and psychological disorders (Oriel, Plane, & Mundt, 2004). This study expanded on prior research in determining whether gender, mental health, perfectionism, test anxiety, and low self-esteem are significantly related to IP feelings. The sample was composed of 506 participants. Women were significantly more likely to report impostor beliefs than men, t(504) = 3.44, p < .001. Mental health, r(500) = -.48, p < .001, perfectionism, r(506) = .48, p < .001, and test anxiety, r(504) = .50, p < .001, were significantly related to impostor beliefs, whereas low self-esteem was not related to the IP, r(500) = -.47, p = .09. Research on this topic is important because the impostor phenomenon is pervasive in academic and numerous professional environments.

he imposter phenomenon (IP) is a cognitive patterns related to IP have clinical psychological phenomenon of conflicting implications because of the potentially elevated Tresearch. The term was first coined by risk for other issues such as depression, anxiety, and Clance and Imes (1978) to describe a high- low self-esteem. We designed this study to expand achieving individual that feels like a fraud and thus upon existing research to determine if gender is attributes success to external factors, such as hard a significant factor in determining the IP and to work, luck, and charm. Research was ambiguous in determine other psychological variables that are regard to gender differences in IP beliefs. Although related to IP beliefs. some authors reported that women experience IP feelings more than men (King & Cooley, 1995; Gender Kumar & Jagacinski, 2006; McGregor, Gee, & There have been inconsistent findings about the Posey, 2008), other data showed no significant association of gender and IP beliefs. Some research- difference (Caselman, Self, & Self, 2006; Cromwell, ers argued that there were gender differences and Brown, Sanchez-Huceles, & Adair, 1990; Sonnak & that women were more likely to experience the IP Towell, 2001). Additionally, other authors found than men (Henning et al., 1998; King & Cooley, relationships between cognitive distortions, such 1995; Kumar & Jagacinski, 2006). When Clance as perfectionism, and IP feelings (Henning, Ey, & and Imes (1978) initially studied the IP, they only Shaw, 1998; Thompson, Davis, & Davidson, 1998). analyzed data for female participants because they SUMMER 2013 In previous research, correlations existed between believed IP predominantly occurred in women depression, anxiety, low self-esteem, and IP feelings based on the attribution theory. The attribution PSI CHI JOURNAL OF (Cozzarrelli & Major, 1990; Cromwell et al., 1990; theory, in relation to IP, suggests that women have PSYCHOLOGICAL Sonnak & Towell, 2001). These maladaptive lower expectations than men, so women attribute RESEARCH

74 COPYRIGHT 2013 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 18, NO. 2/ISSN 2164-8204) *Faculty mentor Cusack, Hughes, and Nuhu | Gender, Mental Health, and Imposter their success to a temporary cause (Clance & Imes, finding is not surprising, because depression and 1978; Deaux, 1976). Additionally, Clance and Imes low self-esteem are facets of neuroticism. Fur- (1978) might have initially thought the IP affected ther studies connected neuroticism with the IP, women more than men because it became a new indicating that of the Big Five personality traits, for psychological phenomenon during the second neuroticism, depression was specifically a signifi- wave of feminism (Jarrett, 2010). We believe this is cant predictor of the IP (Ross & Krukowski, 2003; important to note because the feminist wave may Ross et al., 2001). Additionally, there are links have brought women’s standings to their attention between the IP and negative affectivity, which is and challenged their roles in society. Later, Clance another characteristic of depression (Oleson, and colleagues hypothesized that IP feelings are Poehlmann, Yost, Lynch, & Arkin, 2000). Thus, the more present in women because the IP may stem IP and depression may often coexist in a person from society’s gender stereotypes with assertiveness because of similarities between the two psycho- and accomplishments more readily associated with logical constructs, such as pressure to live up to a men than women (Clance, Dingman, Reviere, successful image, negative thoughts, and self-doubt. & Strober, 1995). Perfectionism and anxiety. People with IP feel- Some researchers compared men and women ings are often notorious for having high standards, and found women report greater impostor fears and thus they may overwork to try to meet their than men (Henning et al., 1998; King & Cooley, idea of perfection (Clance, 1985a). However, ulti- 1995; Kumar & Jagacinski, 2006). However, the mately, they credit their success to hard work, which finding that women suffer more than men from they feel they have to put do, because they do not IP beliefs was not found universally. Although feel as smart or skilled in comparison to their peers originally the IP was thought to be more present (Parkman & Beard, 2008). Clance (1985b) stated in women, other researchers found that IP feelings that men and women who have multiple roles in are experienced by men at the same rate (Langford their lives and hold perfectionist standards may be & Clance, 1993). Still, data showed no significant likely to hold IP beliefs. This could be because the difference between gender and IP beliefs (Bernard, tendency to overwork and the need to feel special Dollinger, & Ramaniah, 2002; Thompson et al., is characteristic of the IP, which is also related to 1998). For example, when investigating the rela- perfectionism. Perfectionism was the strongest tionship between the IP and self-esteem, mental predictor of the IP in a study of medical, dental, health, and parental rearing style, Sonnak and nursing, and pharmacy students (Henning et al., Towell (2000) found no significant sex differences. 1998). An individual who balances numerous roles Furthermore, because most IP research was based could have the IP because the person may feel the on college, graduate, and professional samples, need to be perfect in all aspects of life (Clance, it is important to note that a few studies focusing 1985a). Authors of one study examined how people on adolescent populations found no significant with the IP reacted to both failure and success gender differences in IP scores (Caselman et al., (Thompson et al., 1998). The authors discovered 2006; Cromwell et al., 1990). Because there were that regardless of the outcome, be it failure or contradictory findings, further research is neces- success, people who endorse IP reported higher sary to clarify whether or not there is a relationship levels of anxiety than people who do not endorse between gender and the IP. IP (Thompson et al., 1998). The higher anxiety levels that people with the IP reported could be Mental Health a product of self-criticism and high standards, as In addition to gender, another aspect to inves- these are elements of perfectionism that are more tigate with the IP is mental health. When the IP pronounced in individuals with the IP than in is identified in a client, it is generally accompanied people who do not endorse IP (Thompson et al., by other psychological problems, such as depres- 1998). Several data indicated that the IP is related sion (McGregor et al., 2008; Oriel et al., 2004; to higher levels of anxiety (Clance & O’Toole, 1987; Ross & Krukowski, 2003; Ross, Stewart, Mugge, Cromwell et al., 1990; Oriel et al., 2004; Sonnak & Fultz, 2001) and anxiety (Clance & O’Toole, & Towell, 2001). Because anxiety is also a facet of 1987; Thompson et al., 1998). When Bernard, neuroticism, the relationships between anxiety SUMMER 2013 et al., (2002) studied the IP and the Big Five and neuroticism and neuroticism and the IP could PSI CHI personality factors, they also found that neu- explain why the IP and anxiety are correlated (Ross JOURNAL OF roticism was positively related to the IP. This & Krukowski, 2003). PSYCHOLOGICAL RESEARCH

Self-Esteem. As previously mentioned, the endorse more IP beliefs. Fourth, we predicted that IP is often presented with other psychological participants who had high levels of test anxiety ailments, such as low self-esteem. Low self-esteem would have higher levels of the IP than participants has been strongly, negatively correlated with who had low levels of test anxiety. Fifth, we believed IP feelings and positively related to depression participants with low levels of self-esteem would (Kolligan & Sternberg, 1991). Some researchers have higher levels of the IP than participants who argued that self-esteem and the IP are too closely had high self-esteem. Sixth, we hypothesized that related (Cozzarrelli & Major, 1990; Fried-Buchalter, gender, mental health, perfectionism, test anxiety, 1992). Fried-Buchalter (1992) argued that the IP and self-esteem would predict IP scores. may be overly similar to fear of success and fear of failure, which evokes low self-esteem and lack of Method confidence; therefore, she believed that caution Participants is necessary when examining data and developing The sample consisted of 506 participants, 105 men new psychological constructs. Cozzarrelli and Major (20.8%) and 401 women (79.2%). The average (1990) conducted a study to determine if the IP was age of participants was 21.02 years. Participants a separate measure of low self-esteem. As predicted, reported their racial backgrounds as White the IP was significantly related to low self-esteem, American (64.6%), African American (17.2%), but the authors believed the relationship between Hispanic (3.4%), Asian American (6.9%), multi- the IP and low self-esteem may have been a result racial (7.7%), and American Indian (0.2%). All of initial differences in self-esteem rather than students were enrolled in a U.S. institution of self-esteem being a uniquely contributing predic- higher education with 13.3% in their first year, tive factor; however, Langford and Clance (1993) 19.5% in their sophomore year, 26.8% in their argued that self-esteem is a separate psychological junior year, 25.4% in their senior year, and 15% construct than the IP because self-esteem measures were graduate students. See Table 1 for more a broader area of attitudes and feelings about demographic data. the self than the IP. Statistical analyses yielded a negative correlation between the IP and self-worth Procedure in adolescents, but that self-concept was more After obtaining Institutional Review Board predictive of IP beliefs in women than in men approval, participants were recruited by 11 under- (Caselman et al., 2006). The relationship between graduate research assistants. Convenience sampling low self-esteem and IP scores was found in adult and snowball sampling technique were used for samples as well (Ross & Krukowski, 2003; Son- this study. The research assistants used e-mail and nak & Towell, 2001). Because of the uncertainty paper flyers on campus to recruit participants, both of whether the IP and self-esteem are separate of which included the survey link. To be eligible to constructs, further research is essential. A greater participate in this study, individuals had to be an comprehension of this phenomenon is necessary undergraduate or graduate students. There was no to better treat the IP in order to help individuals time limit on the survey, and the survey respondent establish a higher self-esteem and sense of self- could start, stop, and restart the survey at his or worth (Langford & Clance, 1993). The goal of her convenience. The online survey took about 20 the current study was to expand on previous min to complete and included questions covering research to increase understanding of the IP by demographic, school, and health information. analyzing these related variables together in order Participation was voluntary; those who agreed to to advance treatment. participate were automatically entered into a ran- dom drawing for one of four $50 Target gift cards. Hypotheses First, we hypothesized that women would report Measures higher levels of the IP than men. Secondly, we Impostor phenomenon. The Clance Imposter Phe- believed participants who had low levels of mental nomenon Scale (CIPS; Clance, 1985a) evaluated health would have higher levels of the IP than imposter beliefs, characterized by a high-achieving SUMMER 2013 participants who had high levels of mental health individual who firmly believes he or she is a fraud. as defined by the Global Health Questionnaire. The 20-item scale used a rating scale ranging from PSI CHI JOURNAL OF Thirdly, we hypothesized that participants who 1 (not at all true) to 5 (very true). Higher scores indi- PSYCHOLOGICAL endorsed higher perfectionist ideals would also cated greater IP beliefs. An example item was “I can RESEARCH

76 COPYRIGHT 2013 BY PSI CHI, THE INTERNATIONAL HONOR SOCIETY IN PSYCHOLOGY (VOL. 18, NO. 2/ISSN 2164-8204) Cusack, Hughes, and Nuhu | Gender, Mental Health, and Imposter give the impression that I’m more competent than I Boucher, Davidson, & Munro, 1997) was used to really am.” Holmes, Kertay, Adamson, Holland, and measure perfectionism. An example item is “I try Clance (1993) used both a clinical and nonclinical to be perfect in everything I do.” A Likert-type sample in their study to evaluate the CIPS. Holmes scale was used ranging from 1 (strongly disagree) et al. (1993) found an alpha coefficient for the scale to 5 (strongly agree). Higher scores imply greater of .96, indicating strong internal consistency. For perfectionism. To establish validity evidence, CAPS the present study, an alpha coefficient of .91 was significantly correlates with depression and anxiety found for the CIPS scale. (Hewitt et al., 2002). Castro et al. (2004) found the Mental health. The abbreviated version of the scale to be reliable over a 1 week period. For this General Health Questionnaire (GHQ-12; Gold- study, an alpha coefficient of .89 was found. berg, 1972) was used to assess general wellbeing. Self-esteem. The Rosenberg Self-Esteem Scale It is a screening instrument to detect psychiatric (RSE; Rosenberg, 1965) consists of 10 items that disorders in community settings and nonpsychiat- measures self-esteem. An example item includes ric clinical settings. The questionnaire consists of “I feel that I’m a person of worth, at least on an 12 items. A typical question is “Have you recently equal basis with others.” A Likert-type scale was been feeling unhappy and depressed.” The scale used ranging from 1 (strongly disagree) to 5 (strongly ranged from 1 (not at all) to 4 (much more than agree). Higher scores are indicative of higher self- usual). Higher scores indicates better mental esteem. Convergent, discriminant, and predictive health. Hardy, Shapiro, Haynes, and Rick (1999) validity have been supported in numerous studies, reported an alpha coefficient of .89 for the scale such as the following. Hagborg (1996) found the and established both convergent and discriminant RSE correlated with similar self-esteem measures in adults and children. Rosenberg (1965) evaluated validity evidence for the GHQ by finding strong predictive validity and found self-esteem related correlations with other measures of mental and to many social and interpersonal consequences. physical health. For the current study, an alpha Reliability estimates are strong across ethnic groups coefficient of .88 was found for the GHQ, which (Hatcher & Hall, 2009). An alpha coefficient of .72 establishes further evidence for reliability. was found in this study. Perfectionism. The 14-item Child and Ado- Test anxiety. Instead of the TAI-20 (Spiel- lescent Perfectionism Scale (CAPS; Flett, Hewitt, berger, Gonzalez, Taylor, Algaze, & Anton, 1978), we used Taylor and Deane’s (2002) shortened, five TABLE 1 item version, the TAI-5. An example item is “I wish Participant Characteristics (N = 506) examinations did not bother me so much.” The TAI n % asks participants to rate their level of test anxiety on Age a 1 (almost never) to 4 (almost always). To establish 17–20 255 51.5 validity evidence, Marszalek (2009) found that the TAI-5 correlated with other measures of anxiety. 21–28 228 46.1 Taylor and Deane (2002) reported an alpha coef- 29–44 12 2.4 ficient of .86 for the short version of the scale. Test Race anxiety had an alpha coefficient of .93 for this study. White 326 64.6 Black 87 17.2 Results Asian 35 6.9 A t test was used to determine if there was a sig- Hispanic 17 3.4 nificant difference between gender and IP scores. We found there was a significant difference with Multiracial 39 7.7 women more likely to report IP feelings than Year in School men, t(504) = 3.44, p < .001, d = 0.42. Correlations First Year 67 13.3 were used to analyze the hypotheses regarding IP, Sophmore 98 19.5 mental health, test anxiety, perfectionism, and Junior 135 28.6 self-esteem. The first hypothesis of IP and mental Senior 128 25.4 health being correlated was supported. Mental SUMMER 2013 health was positively correlated with IP. The second Graduate Student–MS 41 8.2 hypothesis of IP and perfectionism being related PSI CHI Graduate Student–Doctoral 34 6.8 JOURNAL OF was also supported. Perfectionism was significantly PSYCHOLOGICAL RESEARCH

related to IP. Additionally, test anxiety was signifi- believe their new roles should be done perfectly. cantly related to IP. The hypothesis of a correlation Because women are more likely to have overwhelm- between IP and self-esteem was not supported. IP ing demands from multiple roles in their lives and self-esteem were not significantly related. See than men, a superwoman attitude, where a woman Table 2 for descriptive variables and Table 3 for a feels she must excel across all areas, may trigger correlation matrix. IP beliefs (Clance, 1985b). This idea originates We also hypothesized that mental health, from stereotypic gender roles. Thus, if women perfectionism, test anxiety, and self-esteem would internalize these gender roles, they may endorse predict IP scores. To evaluate this hypothesis, society’s belief that they are not as successful as men we used linear regression. The overall model was (Langford & Clance, 1993). A theoretical applica- significant, R 2 = .43, F(5, 486) = 73.16, p < .001. tion for these results may be further explained Gender, mental health, perfectionism, and test by the attribution theory previously presented by anxiety were the variables that were significant Clance and Imes (1978). predictors of the IP, whereas self-esteem was Mental health was significantly correlated not a predictor of the IP (p = .43). See Table 4 for with IP, with lower scores of mental health being this model. related to higher scores of the IP. This may indicate that someone with the IP is also likely to have Discussion lower mental health than someone without the IP; In this study, we examined if there were gender however, it is necessary to interpret these results differences in IP beliefs. Beyond this, correlations with caution, as these findings are correlations between IP with mental health, perfectionism, test based on participants who were recruited with con- anxiety, and self-esteem were assessed. We found venience and snowball sampling. Prior researchers that greater impostor beliefs were endorsed more revealed IP beliefs were correlated with depression so by women than by men. Significant results were (McGregor et al., 2008; Oleson et al., 2000; Oriel found between IP and mental health, perfection- et al., 2004; Ross et al., 2001). This is an important ism, and test anxiety, but self-esteem was not implication when treating an individual with the significantly related to IP. IP. Although depressive symptoms may be initially We found a significant difference between more pronounced than IP beliefs, it is imperative to gender and IP scores, which supports previous address all symptoms and recognize that cognitive research (Henning et al., 1998; King & Cooley, distortions, such as guilt and shame, are charac- 1995; Kumar & Jagacinski, 2006). Women are more teristic of both depression and the IP (Clance & likely to have IP beliefs than men. Whereas previous O’Toole, 1987). researchers were undecided on gender differences, Perfectionism was strongly related to IP scores we were not surprised to find that women reported with higher scores of perfectionism being linked greater IP beliefs than men. Women have more with higher scores of the IP. Test anxiety was also roles than men and are expected to excel at all of significantly correlated to IP scores with higher their roles equally, which can lead to impostor feel- scores of test anxiety being related to higher IP ings (Clance, 1985b; Clance et al., 1995). Harvey scores. These results were expected because prior and Katz (1985) asserted that people with the IP studies show perfectionistic beliefs to be one of the most commonly endorsed irrational beliefs TABLE 2 for people with the IP (Cromwell et al., 1990). Anxiety tends to be an underlying problem in both Descriptive Variables for IP, GHQ, CAPS, TAI, and RSE perfectionism and the IP (Cromwell et al., 1990); M SD Range thus, this correlation was anticipated. People who IP 58.68 13.87 26–98 endorse IP are notorious for striving for perfec- GHQ 33.25 6.52 13–47 tionism, and when individuals with the IP invest CAPS 42.34 10.09 15–69 excessive effort to meet perfection, it can cause TAI 10.92 4.59 5–20 additional anxiety (Parkman & Beard, 2008). Ele- ments of perfectionism are more pronounced and RSE 30.26 6.19 20–46 SUMMER 2013 common in people with IP feelings than people Note. Correlations for participants (N = 506) are presented above. Higher scores indicate greater magnitude. GHQ =General Health Questionnaire; CIPS = Clance Imposter who do not endorse IP, such as high standards PSI CHI Phenomenon Scale; CAPS = Child and Adolescent Perfectionism Scale; TAI = Test and self-critical thoughts (Thompson et al., 1998). JOURNAL OF Anxiety Inventory; RSE = Rosenberg Self-Esteem Scale. PSYCHOLOGICAL Test anxiety was also related to the IP (Kumar RESEARCH

& Jagacinski, 2006). This finding is not surprising to determine whether mental health, perfection- because the IP has roots in anxiety (Langford & ism, test anxiety, and self-esteem were significant Clance, 1993). A possible explanation for individu- predictors of impostor beliefs. Although the overall als with the IP to have elevated levels of test anxiety model was significant, only mental health, perfec- could be attributed to the fact that people who tionism, and test anxiety significantly predicted the endorse IP are likely to have lower levels of academic IP. This is not surprising, as IP scores have been efficacy (Thompson et al., 1998). Perfectionism and linked to more mental health problems in previ- test anxiety may be related because perfectionism ous research (Oriel et al., 2004; Ross & Krukowski, is embedded in anxiety issues and because individu- 2003; Ross et al., 2000). Specifically, depression was als with the IP are likely to view intelligence in a a strong predictor of the IP (Oleson et al., 2000). fixed viewpoint. If their skill sets are not aligned It would be interesting to discover what aspects of with their idea of intelligence, they may be unable mental health contribute to the IP in future studies. to internalize success (Langford & Clance, 1993). Additionally, Clance stated that many individuals Matthews and Clance (1985) posed the question of with the IP hold perfectionist beliefs (1985a). what constitutes real success and genuine achieve- Thompson et al. (1998) found that individuals ment. This leads one to question what individuals with the IP held higher standards and reported with IP define as success. Are they so high-achieving more anxiety than individuals without the IP. that perfection is the standard? If this is the case, it Because of this relationship between perfectionism could explain why people who endorse IP cannot and anxiety (Thompson et al., 1998), the fact that internalize their success because perfection is often these two variables significantly predicted IP scores unattainable. Along the same lines of test anxiety, in was expected. a broader sense, individuals with the IP are prone Self-esteem was not a predictor of the IP. to have high anxiety in general (Clance & O’Toole, This could be because IP and self-esteem are not 1987; Oriel et al., 2004). Personality traits such as independent constructs, as Fried-Buchalter (1992) neuroticism and introversion, which are related to and Cozzarelli and Major (1990) suggested. These anxiety, could be a potential explanation for the results are interesting for treating individuals with IP and perfectionism (Ross & Krukowski, 2003). impostor beliefs because clinicians should be aware No significant relationship was found between that mental health, perfectionism, and anxiety con- self-esteem and the IP. This was the opposite of tribute to more IP feelings. Treatment outcomes what was hypothesized because prior research may be more successful if the therapist is cognizant found low self-esteem as a predictor of the IP (Ross that these variables are closely related to each other & Krukowski, 2003; Sonnak & Towell, 2001). Some and predictors of the IP so that the individual researchers argue that there is too much overlap receives care for all of the potential comorbid in low self-esteem and the IP as psychological effects. Because self-esteem and the IP may not concepts to analyze the two concepts separately be exclusive constructs, treatment considering (Cozzarrelli & Major, 1990; Fried-Buchalter, 1992). The nonsignificant results could be explained TABLE 3 through bias with self-report measures and social desirability. Kolligan and Sternberg (1991) found Summary of Correlations, Means, and Standard that self-esteem was strongly, negatively correlated Deviations for Scores on IP, GHQ, CAPS, TAI, and RSE with fraudulent feelings and positively correlated Measure 1 2 3 4 5 with depression. Low self-esteem could be also 1. CIPS – -.48*** .48*** .50*** -.47 interconnected with the IP and depression to 2. GHQ – .30*** .32*** .03 yield significant results for the current study. Even 3. CAPS – .35*** -.03 though low self-esteem was not significantly related 4. TAI – -.01 to the IP, it is pertinent to recognize that other 5. RSE – research has found it correlated with the IP. This knowledge is applicable to treating the IP because M 58.68 33.25 42.34 10.92 30.26 if an individual has IP beliefs, it is important to SD 13.87 6.52 10.09 4.59 6.19 establish good self-esteem in hopes that they learn Note. Correlations for participants (N = 506) are presented above. Higher scores indicate SUMMER 2013 greater magnitude. GHQ =General Health Questionnaire; CIPS = Clance Imposter how to internalize accomplishments and self-worth Phenomenon Scale; CAPS = Child and Adolescent Perfectionism Scale; TAI = Test Anxiety PSI CHI (Langford & Clance, 1993). Inventory; RSE = Rosenberg Self-Esteem Scale. All analyses were two-tailed. * p < .05, ** p < .01, *** p < .001. JOURNAL OF Our sixth hypothesis was supported. We wanted PSYCHOLOGICAL RESEARCH

all the factors (the IP, mental health, perfection- IP and people who do not endorse IP, this would ism, and anxiety) could result in improvement of strengthen self-report data and offer less disputable an individual’s self-esteem as well. results. Additionally, as some researchers believe One of the positive outcomes of this study that the IP originates from gender stereotypes is that it furthers research on the IP, which is a (Clance et al., 1995), it might be worthwhile for relatively new psychological construct with just over future research to determine if there are differ- 30 years of research. It is valuable to gather ences between impostor feelings for women who information on psychological constructs that are identify themselves as feminists and those who correlated with the IP because it gives a more do not identify as feminists. Along similar lines of comprehensive understanding of the phenomenon. gender, it would be interesting to study IP beliefs Additionally, this study had good ethnic diversity. in a sample that consists of both lesbian and het- Although the majority of the sample was composed erosexual women. We predict that lesbian women of White American participants (64.4%), this would report lower IP scores than heterosexual percentage is just over half of the sample, which women because there is more fluidity in gender was of an adequate size (506 participants). There- roles in the lesbian, gay, bisexual, and transgender fore, these numbers show that our sample was (LGBT) community (Blashill & Powlishta, 2009). ethnically diverse. In conclusion, there is valuable information to Whereas the sample size and various racial be taken from this study. Gender, mental health, groups represented are positive aspects of this perfectionism, and test anxiety were all significantly study, the limitations should be kept in mind related to IP beliefs. Although self-esteem was not significantly correlated with the IP, this contradicts when interpreting the data. Ninety-two percent prior research, so further research should be con- of participants were between the ages of 17–24, so ducted for clarification. The IP is a relevant topic to these results may not be generalized to other age research because it is pervasive to many women in groups. Also, most of the participants in this study academic and professional settings, and associated were women (79.2%). It would be interesting to see with other serious psychological problems. if the results would replicate with a more gender balanced sample. Lastly, as always with self-report References data, it is critical to read results cautiously, as people Bernard, N. S., Dollinger, S. J., & Ramaniah, N. V. (2002). Applying may not respond accurately. the big five personality factors to the impostor phenomenon. Journal of Personality Assessment, 78, 221–233. doi:10.1207/ Because the IP is a recent phenomenon, results S15327752JPA7802_07 have remained inconsistent in what predicts the IP Blashill, A. J., & Powlishta, K. K. (2009). Gay stereotypes: The use of and which factors are associated with the IP. Thus, sexual orientation as a cue for gender-related attributes. Sex Roles, 61, 783–793. doi:10.1007/s11199-009-9684-7 further research is necessary. A future study might Caselman, T. D., Self, P. A., & Self, A. L. (2006). Adolescent attributes consider using physiological measurements in contributing to the impostor phenomenon. Journal of Adolescence, conjunction with self-report measures of stress and 29, 395–405. doi:10.1016/j.adolescence.2005.07.003 Castro, J., Gila, A., Gual, P., Lahortiga, F., Saura, B., & Toro, J. (2004). test-anxiety and contrast these differences between Perfectionism dimensions in children and adolescents with anorexia people who endorse IP and people who do not nervosa. Journal of Adolescent Health, 35, 392–398. doi:10.1016/j. endorse IP. If there were significant differences in jadohealth.2003.11.094 Clance, P. R. (1985a). The superwoman aspect and the perfectionistic the physiological measurements, such as heart rate male. In The impostor phenomenon: When success makes you feel and blood pressure, between people who endorse like a fake (pp. 77–87). New York, NY: Bantam Books. Clance, P. R. (1985b). The impostor phenomenon: Overcoming the fear that haunts your success. Atlanta, GA: Peachtree. TABLE 4 Clance, P. R., Dingman, D., Reviere, S. L., & Strober, D. R. (1995). Predictors of Imposter Phenomenon Impostor phenomenon in an interpersonal/social context: Origins and treatment. Women & Therapy, 16, 79–96. doi:10.1300/ Variable B 95% CI β t p J015v16n04_07 Clance, P. R., & Imes, S. A. (1978). The impostor phenomenon in Gender 2.50 [0.19, 4.80] 0.74 2.12 .034 high achieving women: Dynamics and therapeutic intervention. Mental Health -0.61 [-0.76, -.045] -0.29 -7.78 .001 Psychotherapy: Theory, Research & Practice, 15, 241–247. doi:10.1037/h0086006 Perfectionism 0.39 [0.29, 0.49] 0.29 7.63 .001 Clance, P. R., & O’Toole, M. A. (1987). The impostor phenomenon: An internal barrier to empowerment and achievement. Women Test Anxiety 0.88 [-0.66, 1.11] 0.29 7.74 .001 SUMMER 2013 & Therapy, 6, 51–64. doi:10.1300/J015V06N03_05 Self-Esteem -0.51 [-0.30, 0.12] -0.03 -0.83 .407 Cozzarelli, C., & Major, B. (1990). Exploring the validity of the impostor phenomenon. Journal of Social and Clinical Psychology, 9, 401–407. PSI CHI Note. R2 = .43 (N = 506, p = .001). CI = Confidence Interval forB . This table shows 95% JOURNAL OF confidence intervals forB . doi: 10.1521/jscp.1990.9.4.401 PSYCHOLOGICAL Cromwell, B., Brown, N. W., Sanchez-Huceles, J., & Adair, F. L. (1990). RESEARCH

The impostor phenomenon and personality characteristics of high Marszalek, J. M. (2009). Validation of a TAI short form with an adolescent school honor students. Journal of Social Behavior and Personality, sample. The Journal of General Psychology, 136, 333–349. 5, 563–573. doi: 10.1080/00221300903228447 Deaux, D. (1976). Sex: A perspective on the attribution process. In Matthews, G., & Clance, P. R. (1985). Treatment of the impostor J. H. Harvey, W. J. Ickes, & R. F. Kidd (Eds.). New directions in phenomenon in psychotherapy clients. Psychotherapy in Private attribution research (Vol. 1, pp. 335–352). New York, NY: Halsted Practice, 3, 71–81. Press Division, Wiley. McGregor, L. N., Gee, D. E., & Posey, K. E. (2008). I feel like a fraud and it Flett, G. L., Hewitt, P. L., Boucher, D. J., Davidson, L. A., & Murno, Y. (1997). depresses me: The relation between the impostor phenomenon and The child-adolescent perfectionism scale: Development, validation, depression. Social Behavior and Personality, 36, 43–48. doi:10.2224/ and association with adjustment. Unpublished manuscript. sbp.2008.36.1.43 Fried-Buchalter, S. (1992). Fear of success, fear of failure, and Oleson, K. C., Poehlmann, K. M., Yost, J. H., Lynch, M. E., & Arkin, R. the impostor phenomenon: A factor analytic approach to convergent M. (2000). Subjective overachievement: Individual differences in and discriminant validity. Journal of Personality Assessment, self-doubt and concern with performance. Journal of Personality, 58, 368–379. doi:10.1207/s15327752jpa5802_13 68, 491–524. Goldberg, D. P. (1972). The detection of psychiatric illness by questionnaire Oriel, K., Plane, M. B., & Mundt, M. (2004). Family medicine residents and (Maudsley Monograph No. 21). Oxford, England: Oxford University the impostor phenomenon. Family Medicine, 36, 248–252. Press. Parkman, A., & Beard, R. (2008). Success planning and the impostor Hagborg, W. J. (1996). Scores of middle-school age students on the phenomenon in higher education. CUPA-HR Journal, 59, 29–36. Rosenberg Self-Esteem Scale. Psychological Reports, 78, 1071 Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, –1074. doi: 10.2466/pr0.1996.78.3c.1071 NJ: Princeton University Press. Hardy, G. E., Shapiro, D. A., Haynes, C. E., & Rick, J. E. (1999). Validation of Ross, S. R., & Krukowski, R. A. (2003). The impostor phenomenon and the General Health Questionnaire-12: Using a sample of employees maladaptive personality: Type and trait characteristics. Personality from England’s health care services. Psychological Assessment, 11, and Individual Differences, 34, 477–484. doi:10.1016/S0191- 159–165. doi:10.1037/1040-3590.11.2.159 9969(02)00067-3 Harvey, J. C., & Katz, C. (1985). The terrible secret. If I’m so successful, Ross, S. R., Stewart, J., Mugge, M., & Fultz, B. (2001). The impostor why do I feel like a fake? New York, NY: St. Martin’s Press. phenomenon, achievement dispositions, and the Five Factor Model. Hatcher, J., & Hall, L. A. (2009). Psychometric properties of the Rosenberg Personality and Individual Differences, 31, 1347–1355. doi:10.1016. Self-Esteem Scale in African American single mothers. Issues in Mental S0191-8869(00)00228-2 Health Nursing, 30, 70–77. doi:10.1080/01612840802595113 Spielberger, C. D., Gonzalez, H. P., Taylor, C. J., Algaze, B., & Anton, W. D. Henning, K., Ey, S., & Shaw, D. (1998). Perfectionism, the impostor (1978). Examination stress and test anxiety. In C. D. Spielberger & I. phenomenon and psychological adjustment in medical, dental, G. Sarason (Eds.), Stress and anxiety (pp. 167–191). New York, NY: nursing and pharmacy students. Medical Education, 32, 456–464. doi:10.1046/j.1365-2923.1998.00234x Hemisphere/Wiley. Hewitt, P. L., Caelian, C. F., Flett, G. L., Sherry, S. B., Collins, L., & Flynn, Sonnak, C., & Towell, T. (2001). The impostor phenomenon in British C. A. (2002). Perfectionism in children: Associations with depression, university students: Relationships between self-esteem, mental anxiety, and anger. Personality and Individual Differences, 32, 1049 health, parental rearing style and socioeconomic status. Personality –1061. doi:10.1016/S0191-8869(01)00109-X and Individual Differences, 31, 863–874. doi: 10.1016/S0191- Holmes, S. W., Kertay, L., Adamson, L. B., Holland, C. L., & Clance, P. 8869(00)00184-7 R. (1993). Measuring the impostor phenomenon: A comparison of Taylor, J., & Deane, F. P. (2002). Development of a short form of the Test Clance’s IP Scale and Harvey’s I-P Scale. Journal of Personality Anxiety Inventory (TAI). The Journal of General Psychology, 129, Assessment, 60, 48–59. 127–136. doi:10.1080/00221300209603133 Jarret, C. (2010). Feeling like a fraud. The Psychologist, 23, 380–383. Thompson, T., Davis, H., & Davidson, J. (1998). Attributional and affective King, J. E., & Cooley, E. L. (1995). Achievement orientation and the responses of impostors to academic success and failure outcomes. impostor phenomenon among college students. Contemporary Personality and Individual Differences, 25, 381–396. doi:10.1016/ Educational Psychology, 20, 304–312. doi:10.1006/ceps.1995.1019 S0191-8869(98)00065-8 Kolligan, J., & Sternberg, R. J. (1991). Perceived fraudulence in young adults: Is there an ‘impostor syndrome’? Journal of Personality Author Note. Claire E. Cusack, Psychology Department, Agnes Assessment, 56, 308–326. doi:10.1207/s15327752jpa5602_10 Scott College; Jennifer L. Hughes, Department of Psychology, Kumar, S., & Jagacinski, C. M. (2006). Impostors have goals too: The Agnes Scott College; Nadi Nuhu, Psychology Department, impostor phenomenon and its relationship to achievement goal Agnes Scott College theory. Personality and Individual Differences, 40, 147–157. Nadi Nuhu is now in the Clinical Psychology Program, doi:10.1016/j.paid.2005.05.014 Auburn University. Langford, J., & Clance, P. R. (1993). The impostor phenomenon: Recent Correspondence concerning this article should be research findings regarding dynamics, personality, and family patterns addressed to Jennifer L. Hughes, Department of Psychology, and their implications for treatment. Psychotherapy: Theory, Research, Agnes Scott College, Decatur, GA 30030. E-mail: jhughes@ Practice, Training, 30, 495–501. doi:10.1037/0033-3204.30.3.495 agnesscott.edu

SUMMER 2013

PSI CHI JOURNAL OF PSYCHOLOGICAL RESEARCH