Women’s Creative Performance: The Role of Impostor Phenomenon and Affirmative Action Perception
Sadaf Kaykha (12492604)
Amsterdam Business School, University of Amsterdam Executive Program in Management Studies – Management Track
Dr. Tanja Hentschel June 30, 2020
EBEC approval number: 20200313120302
Table of Contents
Statement of Originality page 3
Acknowledgements page 4
Abstract page 5
Introduction page 6
Theoretical framework page 8
Women’s creativity and creative performance page 8 Impostor phenomenon and women’s creative performance page 11 Affirmative action and women’s creative performance page 13
Affirmative action and impostor phenomenon page 15
Method page 16
Procedure and design page 16
Participants page 17
Measures page 19
Results page 21
Correlation analysis of variables page 21
Moderated mediation model page 24
Exploratory analysis page 26
Discussion page 27
Summary and discussion of results page 27
Limitations and future direction page 30
Suggestions for future research page 32
Practical implications page 33
Conclusion page 34
References page 35
Appendices page 41
Statement of Originality
This document is written by student Sadaf Kaykha who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Signature ___________________________________________
Acknowledgements
I would like to first of all thank my supervisor Dr. Tanja Hentschel for her guidance throughout this research. A word of appreciation to Rutger Mulder for his time and feedbacks while he was writing his own thesis. I am thankful to all my current and former colleagues and all my friends who took the time to participate in the surveys.
On a personal level, I would like to express my gratitude to my dearest Samrad Ghane for his encouragement, extreme patience, and unconditional support during the last two years.
Most importantly, I would like to thank my parents Manuchehr and Mashy for all the sacrifices they have made in their lives to bring me where I am today; and for setting the example to never stop learning because it’s never too late for new experiences. Finally, a very special thank you to my sister Morvarid who is my rock, always there for me no matter what.
Abstract
In order to enhance diversity and improve their creative performance, organizations seek to increase the presence of women in the workforce by implementing affirmative action policies, such as the introduction of gender recruitment quotas. It has been argued that affirmative action policies tend to remove the attention from women’s competences and focus it on their gender. This shift can further decrease women’s creative performance, which is already negatively impacted by gender stereotypes and normative expectations. Women’s internalization of gender stereotypes may also make them experience feelings of intellectual phoniness in the work environment. This so-called impostor phenomenon can undermine women’s creative self-efficacy and consequently negatively impact their self-perceived creative performance. The current study investigated the influence of impostor feelings and affirmative action perception on the creative performance of women in the work
environment. To this aim, a survey was conducted in the Netherlands amongst 903 highly educated employees, working contractually at least three days a week. The results showed that female gender predicted a lower self-assessed creative performance, and that impostor phenomenon mediated the negative relationship between gender and creative performance.
Moreover, the results revealed that affirmative action perception moderated women’s lower self-assessed creative performance. This effect was only found for medium and high values of affirmative action perception, but not for its low values. Besides, the results showed no moderation effect of affirmative action perception on women being more likely to experience impostor phenomenon. Theoretical and practical implications of the findings and future direction for research are elaborated.
Keywords: Gender, Creative Performance, Affirmative Action, Impostor Phenomenon, Imposter Feeling, Creativity, Diversity, Quota.
Introduction
In most industries, creativity and innovation are highly appreciated, needed and required assets on the work floor. As Friedman, Friedman and Leverton (2016) highlighted, in today’s knowledge based and globalized economy, the key advantage of organizations is the creative performance of their employees. Diverse teams, because of the variety of their background can explore larger area of information and a wider range of networks leading to an increased creativity and innovation (Friedman et al, 2016; Stahl, Maznevski, Voigt, &
Jonsen, 2010).
In order to increase their creative performance, organizations implement measures to enhance diversity (Gündemir, Martin, & Homan, 2019). Measures to increase the presence of women and minorities in the workforce are called affirmative action, an example of which is the introduction of mandatory gender quotas used in several countries, within firms and political institutions, to increase the presence of women in the workforce (Pande & Ford, 2012). Such initiatives are meant to support disadvantaged or underrepresented groups like women in organizations. But do affirmative action policies always deliver the intended results?
Affirmative action is a controversial practice, because women hired through such actions can be perceived privileged because of policy rather than because of their
qualifications (Calsamiglia, Franke, & Rey-Biel, 2013). Consequently, these women could have negative psychological experiences caused by being targets of prejudice and
reservations about their competence at work (Heilman, McCullough, & Gilbert, 1996). These women would, therefore, find themselves exposed to threatening situational cues, creating an identity-unsafe environment and increasing their vulnerability to stereotype threat (Davis, Spencer, & Steele, 2005). It can, hence, be argued that these psychological pressures and doubts under stereotypes threat can occupy women’s thoughts and reduce their task focus –
negatively impacting their creative performance (Cadinu, Maass, Rosabianca, & Kiesner, 2005). Moreover, as Heilman (1996) highlighted, affirmative action can have a negative impact on the self-evaluation and work attitude of the employees benefiting from them.
Consequently, women perceiving affirmative action could undervalue their own capabilities and as a result have an increased feeling of not belonging and not deserving their professional situation.
Clance and Imes (1978), based on their work with high achieving women, described the experience of intellectual insincerity, leading to feelings of not belonging and not deserving, as the impostor phenomenon. People experiencing the impostor phenomenon, attribute their own success to external factors and tend to have a lower self-perception of their own abilities such as creative performance (Kumar & Jagacinski, 2006). It can, thus, be argued that external factors such as affirmative action would negatively impact the self- perception of women, who consequently can have a lower likelihood to achieve their full creative performance potential.
As Dudău (2013) highlighted, a research gap exists in the relation between impostor phenomenon and creative performance. This gap seems to also exist in the research on the relation between affirmative action and impostor phenomenon. With a widespread
implementation of quotas and preferential recruitment measures for reaching diversity goals and diverse work floors, it has become increasingly important to better understand these relationships. Hence, the questions that this research will investigate are: Can impostor phenomenon explain the relation between gender and creative performance? Does affirmative action perception impact the creative performance of women in the organization? And may the impact of impostor phenomenon on women’s creative performance be due to affirmative action perception?
This study helps to provide a deeper understanding of the mechanisms contributing to a lower self-evaluated creative performance of women. More specifically the study offers a better view on the possible impact of affirmative action perception and impostor phenomenon on women’s creative performance. It will, therefore, set the ground for employers of various sectors to better consider eventual mitigation actions to counter the possible negative effects of affirmative action perception on women’s creative performance, and to develop strategies for reducing impostor feelings and maximizing women’s creative performance in the
workplace.
Theoretical framework Women’s creativity and creative performance
Throughout years of research about creativity, scholars agree on creativity in the workplace being one of the important factors of organizational performance, therefore, essential for firms and their long-term success. It seems, however, that there is no consensus on a single definition for creativity. In 1988, Amabile, after recognizing and reviewing different definitions that existed, described creativity as “the production of novel and useful ideas by an individual”. More recently, Anderson, Potočnik, and Zhou (2014) have confirmed the still persisting lack of agreement between researchers and have tried to remedy to the matter by providing an integrative definition as follows: “Creativity and innovation at work are the process, outcome, and products of attempts to develop and introduce new and
improved ways of doing things. The creativity stage of this process refers to idea generation, and innovation refers to the subsequent stage of implementing ideas toward better
procedures, practices, or products.”
Where both these definitions see creativity as the process and its end result, Findlay and Lumsden (1988) made a distinction between the creative process and creativity. They defined creativity as a combination of traits of an individual contributing to the creative
process, the result of which would be the creative performance also called creative accomplishment or creative achievement. I will use the term and concept of creative performance as a context or domain specific behavior or outcome (see also Baer, 2010).
Assuming that creative performance is a context and domain specific concept, the questions that arise are: Is creative performance a gender related phenomenon? And do men and women have similar creative performances?
Wood and Eagly (2009) explained the concept of gender roles as the expectations about men’s and women’s behaviors and attributes, as shared generally in the society. Gender roles further lead to the concept of gender stereotypes. As Heilman (2012) explained,
stereotypes are general ideas about certain group of people and the belief that those attributes should apply to each individual member of that group just because they belong to that
specific group. Gender stereotypes are, thus, fixed ideas about the characteristics of men and women. Stereotypically, men are seen as more agentic – self-assertive with an urge to master – and women are seen as more communal – selfless and concerned with others (Eagly &
Steffen, 1984). Heilman (2012) also clarified that gender stereotypes can either be
prescriptive – dictating what women and men should be like – or descriptive – outlining how they are like. Both descriptive and prescriptive stereotypes could have a negative impact on women’s creative performance.
In studies reviewed by Heilman (2012), stereotypes are established to be
automatically activated and widely shared, which makes them powerful and likely to be at the basis of people’s impressions and expectations. Stereotypes could lead to someone’s
disadvantage not based on their actions but rather based on their affiliation to a group. For example, in traditionally male gender-typed positions, some specific attributes are believed to be necessary for success (Heilman, 2012). Descriptive gender stereotypes promote that for such positions the attributes women are expected to possess would not match the attributes
considered to be necessary for success (Heilman, 2012). Creativity, for example, is strongly associated with stereotypically masculine-agentic qualities (Proudfoot, Kay, & Koval, 2015);
thus, expectations to have a good creative performance may be higher for men than for women. It can therefore be assumed that descriptive gender stereotypes may generate a negative expectation about the creative performance of women and hence can lead to gender bias.
Moreover, prescriptive gender stereotypes – should and should not – also promote gender bias by dictating normative expectations for both women and men’s behavior. For women, if the stereotypes are violated, the result would be disapproval and social rejection, which can take place in the form of lower evaluation by others such as their supervisor (Heilman, 2012). Considering that creativity is seen as a masculine-agentic quality, women working in positions requiring high creativity – by violating the stereotypes – might see a lower creative performance evaluation by their supervisor.
Furthermore, Wood and Eagly (2009) explained that when gender roles or gender stereotypes are accepted by people, they are then internalized to become part of their
perception of themselves, explaining the way they act. Therefore, when men may internalize the perception of their greater creative skills compared to women, it can lead to men
demonstrating higher creative performance than women. The same way, when women internalize the stereotype that they might have fewer creative skills than men, it can lead to women exhibiting lower creative performance than men.
In a review of gender differences in creativity, Baer and Kaufman (2008), explained that even though the validity of divergent thinking tests (e.g. Torrance Tests for Creative Thinking) to measure creativity are debated for years – and despite gender differences in scores on creativity tests not being conclusive – the available empirical evidence tends to show a higher score for women and girls than men and boys. However, when it comes to
real-world creative performance, the difference trend is reversed, and men’s creative performance is significantly higher than women. They further explained the lower creative performance of women being due to three social and structural factors. First, it is partly caused by the lack of supporting environment to encourage early talent to develop the needed expertise. Second, to some extend it results from the pressure and expectations of the society regarding women’s role in the family (e.g. as a mother). And the third cause is limitations and expectations set by men for the entry to many fields of work and to access their resources.
In sum, gender roles and stereotypes, describe and dictate normative expectations where men are seen as more agentic and women are seen as more communal. Stereotypes also present creativity as a masculine-agentic trait. Internalization of these gender biases enhances the incentive to self-stereotype in line with the societal expectations. These points, combined with the environmental support system from which men benefit, can enable them to have a higher creative performance than women. The first hypothesis of this research addresses, therefore, the relation between gender and creative performance as follows:
H1: Female gender is negatively related to creative performance.
Impostor phenomenon and women’s creative performance
In 1978, Clance and Imes developed the term impostor phenomenon to define the feeling of intellectual insincerity initially reported by high achieving women who feared to be impostors; in other words, women who felt like not belonging amongst the knowledgeable and brilliant people who were surrounding them in the workplace (Clance & O’Toole, 1987).
These women credited their successes to luck – being in the right place at the right time – knowing the right people, or to their interpersonal skills – charm and the ability to
communicate well – rather than to their own competence or capability. Moreover, they were terrified that others would realize that they were not as competent as they seemed to be.
Clance and O’Toole (1987) argued that men also indicated the presence of these impostor feelings but not openly and rather in confidential and anonymous settings. They suggested, however, that women – because of lack of support and their upbringing – express stronger impostor fears than men and are more negatively affected by their effects. They further elaborated that men when facing impostor fears, receive encouragement by their peers and superiors to overcome the fear and pursue success, which is often not the case for
women. Moreover, Clance and O’Toole (1987) explained that women’s childhood exposure to education encouraging stereotyped gender roles such as nurturing, puts them in a
conflictual situation, where they experience guilt when fulfilling their own needs. Such sentiments of guilt and internal conflict can enhance women’s impostor feelings when they occupy roles not in line with the feminine gender stereotypes.
In addition to emphasizing that impostor phenomenon impacts women more than men, Clance and O’Toole (1987) explained how impostor phenomenon affects women’s accomplishments. They explicated that women’s internal conflicts – divided between being self-assertive or being concerned with others – and societal gender stereotypes defining and dictating their skills, in combination with impostor feelings, can prevent women from being empowered to reach higher levels of competence. Barriers that could negatively impact their creative performance. But how does impostor phenomenon relate to women’s belief about their own creative performance?
In a recent study, Intasao and Hao (2018) explained that employees who have a high level of creative self-efficacy – who believe that they can deliver creative work – are
perceived as being creative and receive better ratings for their creative performance from their supervisors. They confirmed in their research that beliefs about creativity positively influence creative performance. Linking the findings of Intasao and Hao (2018) to the
impostor phenomenon, it can be assumed that if creative self-efficacy has a positive influence
on creative performance, then impostor feelings and disbelief could have a negative influence on creative performance. Accordingly, considering that women experience more impostor feelings, the lack of self-efficacy which result from it may limit their creative performance.
To better understand the relation between impostor phenomenon and creative performance, Dudău (2013) conducted a study amongst high school adolescents. She
intended to understand whether students with higher impostor phenomenon symptoms could be less creative than their peers. With some caveat regarding the validity and fidelity of her measures and findings, she found that impostors did report an inferior self-perceived creative performance compared to their peers.
In sum, high achieving women may report experiencing feelings of intellectual phoniness in the work environment. These impostor feelings, even though also present for men, have more impact on women and on their self-perceived creative performance. This is partly due to their lack of creative self-efficacy. It can, therefore, be assumed that in the workplace, women are more likely than men to experience impostor phenomenon, and that this may explain why women believe they have a lower creative performance.
H2: Experiences of impostor phenomenon mediate the negative relationship between female gender and creative performance.
Affirmative action and women’s creative performance
The Equal Employment Opportunity Commission defines affirmative action policies as remedies to counter the discriminating effect of existing or previous policies and blockades to equal employment opportunities (Stewart & Shapiro, 2000). Affirmative action could be used, for example, to increase the number of women on the work floor or amongst the senior management level. As Stewart and Shapiro (2000) explained, the affected population by affirmative action can be divided into two groups. On the one hand, the “recipients” or
“beneficiaries” who based on some criteria other than merit (e.g. recruitment quotas for
women) receive preferential treatments. On the other hand, the “nonrecipients” who do not benefit from any preferential treatment and often feel left out or victimized by affirmative action. This latter feeling and the negative actions or behaviors that could derive from it by nonrecipients, can consequently put the beneficiaries in a difficult or victim position.
In a review of various studies of Heilman and her associates on affirmative action, Stewart and Shapiro (2000) emphasized that the negative consequences of perceiving affirmative action consistently occurred for women, but not for men. This finding led Heilman and her colleagues to highlight the prominent role of gender stereotypes and their internalization for both men and women (Stewart & Shapiro, 2000). They explained further that men – even if perceiving affirmative action – because of the confidence given to them by gender stereotypes, and the internalization of being able to perform in a gender-specific role according to the stereotypes, will approach the role with confidence and expectation of success. However, women – for the same reasons – will approach the role with lack of confidence and lower expectation of success (McMahon, 1982). Considering creativity, which is a masculine-agentic trait, it can be assumed that women perceiving affirmative action would approach creative tasks with less confidence and expect to achieve low creative performance.
Additionally, Stewart and Shapiro (2000) highlighted that contrary to what is intended, affirmative action can have negative consequences for the beneficiaries, all the more when they are aware of the presence of such policies. Women – when aware of the fact that they have been selected preferentially and not based on merit – have displayed negativity against other female applicants for an entry position (Heilman, Kaplow, Amato, & Stathatos, 1993); when given the choice, they have chosen less-demanding tasks (Heilman, Rivero, &
Brett, 1991); they took less credit for their performance, and have had more negative self- perceptions (Heilman, Simon, & Repper, 1987). All three above mentioned reactions from
women aware of having been selected based on affirmative action have had a negative impact on their level of task performance and willingness to continue performing (Heilman et al., 1987). From these findings of Heilman and her associates, it can be hypothesized that affirmative action – when noticeable – could negatively influence women’s self-perceived creative performance.
H3a: The negative relation between female gender and creative performance is stronger amongst individual with higher affirmative action perception.
Affirmative action and impostor phenomenon
Reviewing the literature, it seems that very few publications exist on the relation between affirmative action and impostor phenomenon. To argue the likelihood of a relationship, I will use the work of Heilman (1996) outlining three harmful effects of affirmative action for organizations and their employees, and I will suggest how they can each contribute to increase the impostor phenomenon. First, women benefiting from affirmative action – because they have not been chosen based on merit – experience high incompetence stigma from their peers and supervisors, possibly increasing their impostor feelings (Heilman, 1996). Second, being in presence of resentful nonrecipients of affirmative action can on the one hand, enhance the feeling of not belonging that impostors experience, and on the other hand, generate negative evaluations by others, that will in turn confirm the impostors’ belief of not being competent enough (Clance & Imes, 1978). Finally, women beneficiaries of affirmative action, when aware of the existence of such policies, can
experience negative self-evaluations that would undesirably impact their work attitudes and behavior, hence increase their impostor feeling (Heilman, 1996). Consequently, the
perception of affirmative action policies in firms can moderate the relationship between gender and impostor phenomenon, leading to below hypothesis:
H3b: The relation between female gender and impostor phenomenon is stronger amongst individual with higher affirmative action perception.
The conceptual model of this study, represented in Figure 1, postulates that gender affects creative performance; a relationship that is partially mediated by impostor
phenomenon. Additionally, affirmative action perception moderate both the relation between gender and creative performance as well as the relation between gender and impostor
phenomenon. Ultimately, the model represents a moderated mediation.
Figure 1
Conceptual Model
Method Procedure and design
Twelve students of part-time and full-time master’s in management studies have used a joint survey to collect data for their individual theses. Each student added items for each of their analyzed variables to create the survey. The sampling frame of this study was people working at least three days a week in the Netherlands. To collect the sample, the survey was distributed in the Netherlands to several organizations in various sectors (e.g. raw material, manufacturing, services, public and information) via email (see Appendix A). It was either distributed by the organization’s human resources department to all the employees or by the students to their direct network. A link to the survey was also posted on social media (e.g.
Gender Creative performance
Affirmative action perception
H1 H2 Impostor phenomenon H3a
H3b
LinkedIn in Appendix B, Twitter in Appendix C), aiming to have a snowball effect to optimize the number of participants.
Participants were told that we were interested in behavior in the workplace and that we would collect data with two surveys within a week interval. Moreover, participants were informed in an introductory note that they have to answer both surveys for their results to be usable. In the same note, participants were informed that they needed to provide their email address so that the second survey could be sent to them. They were also insured that their email address would not be used for any other purpose, and that in order to ensure
anonymity, there would be no results matched to the email address.
This study had a cross-sectional design with gender as the predictor, creative performance as the outcome variable, impostor phenomenon was proposed as the mediator and the moderator was expected to be affirmative action perception. To reduce the impact of common method bias, a two-wave panel design was used where the moderator and the mediator where measured at different times (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). At time 1, the predictor and the moderator were measured; at time 2 – which took place between one to three weeks later – the mediator and the outcome variable were measured.
Participants
The total respondent sample size was 1155, constituted of participants who answered at least 50% of questions of the first survey (T1) and at least one question of the second survey (T2). However, the full completion rate of the surveys was rather high as out of the 1155 respondents, 99.3% had answered all questions of T1 survey, and 98.8% had answered all questions of T2 survey. Given that the sample frame of this study required the participants to be current employees of an organization, respondents having identified as being
unemployed (5 cases), retiree (7 cases), student (30 cases) and self-employed (63 cases) were
excluded. Moreover, one participant indicated to be the sole employee of their own business and therefore was excluded. Additionally, to make sure employees were working at least three days a week as per the sample frame, all participants with contractual working hours below 24 hours (71 cases) were excluded. In total 177 participants did not fit the inclusion criteria and were excluded.
The remaining sample of 978 participant was checked for uncomplete data. 65 cases who gave a wrong answer to the attention question, which stated: “This is an attention check:
Please select "strongly agree" to respond to this statement.” were marked as non-completers (Aust, Diedenhofen, Ullrich, & Musch, 2013). Additionally, 11 cases had not answered at least one item related to affirmative action perception, 12 cases had not answered at least one of the creative performance items, and 12 cases had not answered any of the impostor
phenomenon items. Lastly, one case of incomplete age – used as a control variable – was detected. There was some overlap within the non-completers in terms of missing items, therefore the total of non-completers was 75 cases. The two groups of completers and non- completers were compared, and no significant differences were found in terms of gender, age, level of education, employment sector, marital status, organization size and years of work experience. All non-completers were excluded from the main analysis.
Finally, eight answers containing obvious typing errors were corrected1. The final sample used for analysis contained 903 participants, the demographics of which are presented in Table 1. It is to be noted that the majority of the respondents were Dutch (95.2%), and highly educated compared to the national statistics of the Netherlands, with 95.6% of the respondents having an HBO Bachelor’s level or higher versus 30% nationally (CBS, 2018).
1 In three cases, there was a text string added to the number of contractual weekly working hours, as a correction the string was deleted. In three cases, the number of contractual weekly working hours were typed twice in the same field, as a correction the repetition was deleted. In one case, there were ten zeros after the number of contractual weekly working hours, as a correction the zeros were removed. Finally, in one case, the respondent had entered the year of start of work instead of number of years of work experience, the year was replaced by the calculated number of years.
Table 1
Sociodemographic Characteristics of Participants (N=903)
N = 903
n %
Gender Men 379 42.0
Women 524 58.0
Level of education High School degree 40 4.4
College / HBO degree 524 58.0
University Bachelor's degree 53 5.9
University Master's degree 238 26.4
Post graduate degree (e.g. PhD) 48 5.3
Job sector Information (e.g. communication, high tech) 89 9.9 Manufacturing (e.g. industry, craft) 108 12.0 Public (e.g. education, transit, police, healthcare) 482 53.4
Raw materials (e.g. agriculture) 8 .9
Services (e.g. trade, insurance) 214 23.7
Marital status Divorced or separated 26 2.9
Married or domestic partnership 431 47.7
Serious relationship 246 27.2
Single 193 21.4
Widowed 7 .8
Nationality Dutch 860 95.2
Other - Unknown 56 6.2
Organization size Less than 20 employees 59 6.5
Between 21 and 99 employees 134 14.8
Between 100 and 499 employees 298 33.1
Between 500 and 1999 employees 167 18.7
More than 2000 employees 244 27.0
Leadership experience Current 195 21.6
Past 178 19.7
None 530 58.7
M SD
Age 39.75 12.72
Work experience 18.01 12.89
Note. Some participants had selected two nationalities (n=13).
Measures
For this study, gender was inquired as part of the demographic questions of the survey and validated measurement scales were used for creative performance and impostor
phenomenon. Thus, creative performance was measured with the below three items (α = .89) of the “Idea Generation” part of the 9-itemed scale of Janssen (2001), which he created based on the “Innovative Behavior Measure” scale of Scott and Bruce (1994): “How often do you perform these behaviors at work? a) Creating new ideas for improvement. b) Searching out new working methods/techniques. c) Generating original solutions to problems.” All three items were rated on a 7-point Likert scale (from 1 = “never” to 7 = “all the time”).
Impostor phenomenon was measured by the three below items (α = .86) of the 20- itemed “The Impostor Phenomenon” scale created by Clance and O’Toole (1987): a) “When people praise me for something I've accomplished, I'm afraid I won't be able to live up to their expectations of me in the future”. b) “I'm afraid people important to me may find out that I'm not as capable as they think I am”. c) “I often worry about not succeeding with a project or on an examination, even though others around me have considerable confidence that I will do well.” All three items were rated on a 7-point Likert scale (from 1 = “strongly disagree” to 7 = “strongly agree”).
Affirmative action perception was measured with two items (α = .93) created for the purpose of this survey rated on a 7-point Likert scale (from 1 = “strongly disagree” to 7 =
“strongly agree”). a) “I am aware that my organization has specific HR practices aimed at supporting the careers of female employees.” b) “I am aware that my organization has specific HR practices aimed at recruiting women.”
Moreover, age, work experience, contractual working hours, actual working hours, level of education, organization size and language used to answer the survey were set as control variables in the main analysis. Indeed, it could be expected that younger women, with less working experience are more susceptible to experience impostor phenomenon.
Moreover, as Clance and O’Toole (1987) explained, impostor phenomenon has been more frequently observed in high achieving women, therefore the level of education and long
working hours could have an effect on the results. Furthermore, it could be expected that affirmative action is more frequent in bigger organization with well-structured human resource management, therefore organization size was controlled for (Richard, Roh, &
Pieper, 2013). Additionally, because the affirmative action perception items were created for the purpose of this study and were translated into Dutch, it appeared to be interesting to control for the language used while completing the survey at T1 containing the affirmative action perception items.
Results Correlation analysis of variables
Table 2 shows the descriptive statistics (means and standard deviations) and bivariate Pearson correlation of the main variable (gender, creative performance, impostor
phenomenon and affirmative action perception) as well as those of the control variables (age, work experience, contractual working hours, actual working hours, level of education, organization size and language used to answer T1 survey). The results showed that gender had a significant (p < .01) negative correlation with creative performance, affirmative action perception, age, work experience and weekly working hours. This negative correlation meant that women scored lower on these variables than men. Gender also had a significant (p <
.001) positive correlation with impostor phenomenon, meaning that women scored higher on this variable than men.
Moreover, creative performance had a significant (p < .001) positive correlation with affirmative action perception, level of education and weekly working hours. It also had a significant (p < .001) negative correlation with impostor phenomenon, meaning the higher the impostor phenomenon, the lower the creative performance. Furthermore, impostor phenomenon had a significant (p < .01) negative correlation with affirmative action, age and work experience, as well as a significant (p < .001) positive association with using English
language at T1. Finally, affirmative action perception had a significant (p < .01) positive correlation with age, organization size, work experience, and working hours, as well as a positive correlation (p < .05) with using English language at T1.
In sum, age, work experience, working hours, level of education and language used at T1 showed significant associations with at least two of the main variables. Therefore, they were used as control variables in the moderated mediation analysis. However, organization size correlated only with one main variable and therefore was not used as a control variable.
Descriptive Statistics and Correlations for Study Variables
Note. ***p < .001, **p < .01, *p < .05. Gender is coded as 0 men, 1 women. Language used is coded as 0 Dutch, 1 English.
Variable n M SD 1 2 3 4 5 6 7 8 9 10
1. Gender 903 .58 .49 -
2. Creative performance 903 4.67 .90 -.16*** -
3. Impostor phenomenon 903 3.24 1.45 .22*** -.15*** -
4. Affirmative action perception 903 3.47 1.68 -.15*** .15*** -.09** -
5. Age (covariate) 903 39.75 12.72 -.11** -.06 -.23*** .11** -
6. Level of education (covariate) 903 3.70 1.07 .04 .16*** .12** .02 -.16*** -
7. Language used T1 (covariate) 903 .08 .28 .04 .02 .12*** .07* -.07* .28*** -
8. Organization size (covariate) 902 3.45 1.22 .01 -.03 .05 .24*** .06 .11** .05 -
9. Work experience years (covariate) 870 18.01 12.89 -.10** -.05 -.24*** .13*** .95*** -.26*** -.09* .04 - 10. Contractual weekly working hours
(covariate) 903 34.80 4.87 -.30*** .13*** .07* .18*** -.25*** .18*** .11** .07* -.25*** - 11. Actual weekly working hours
(covariate) 902 38.32 7.94 -.28*** .23*** -.00 .30*** -.13*** .25*** .14*** .09** -.14*** .71***
Moderated mediation model
To investigate whether the indirect effect of gender on creative performance through impostor phenomenon depend on affirmative action, a moderated mediator analysis was performed using SPSS PROCESS macro model 8 with 5,000 bootstraps and 95% bias correction (Hayes, 2017). The outcome variable for this analysis was creative performance, the predictor variable was gender (coded as 0 = male, 1 = female), the mediator was impostor phenomenon, and the moderator was affirmative action perception.
Looking at the individual relations between each pair of variables, reported in Table 3, gender had a significant negative relationship with creative performance (β = -.17, SE = .07, CI [-.29, -.04]). This means that in this sample, women had a lower creative performance than men. This finding was in line with hypothesis 1 stating female gender is negatively related to creative performance.
Table 3
Mediation and Moderation Analyses
Note. N = 869. Work experience years had a number of missing answers which explain N.
Consequent
Creative performance (Y)
Impostor phenomenon (M)
Antecedent β SE p β SE p
Gender -.17 .07 .012 .62 .10 .000
Impostor phenomenon -.09 .02 .000 - - -
Affirmative action perception .15 .05 .001 -.09 .07 .198
Gender x Affirmative action -.11 .06 .065 .05 .10 .598
Constant 4.61 .33 .000 1.98 .52 .001
Age (covariate) -.01 .01 .127 .01 .01 .498
Work experience years (covariate) .01 .01 .401 -.03 .01 .027 Contractual weekly working hours (covariate) -.02 .01 .041 .04 .01 .007 Actual weekly working hours (covariate) .03 .01 .000 -.01 .01 .106
Level of education (covariate) .13 .03 .000 .05 .05 .279
Language used T1 (covariate) -.11 .11 .333 .50 .18 .004
R2 = .12
R2 = .12
F (10,858) = 11.66
p < .001 F (9,859) = 12.94 p < .001
Moreover, as reported in Table 3, gender had a significant positive relationship with impostor phenomenon (β = .62, SE = .10, CI [.42, .82]), meaning that in this sample, women reported experiencing higher impostor phenomenon than men. Furthermore, impostor phenomenon had a significant negative relationship with creative performance (β = -.09, SE
= .02, CI [-.13, -.05]). This means that in this sample, the higher the impostor phenomenon, the lower the creative performance. Consequently, women reported higher impostor feeling and lower creative performance than men. These results were in line with hypothesis 2 declaring that experiences of impostor phenomenon mediate the negative relationship between gender and creative performance.
In line with hypothesis 3a stating that the negative relation between female gender and creative performance is stronger amongst individual with higher affirmative action perception, the interaction of gender and affirmative action perception on creative
performance was marginally significant (p = .065). The conditional effect reported in Table 4 shows that for low values of affirmative action perception, there was no significant negative effect of gender on creative performance – thus, men and women indicated equal levels of creative performance. However, for medium and high values of affirmative action perception, the direct effect of gender on creative performance was significant. This means that
affirmative action perception, only for its medium and high values, had a moderating role on the negative direct effect of gender on creative performance.
Table 4
Conditional Direct Effect Analysis
Conditional direct effect(s) of gender on creative performance
Affirmative action perception effects SE p LLCI ULCI
-1.00 -.05 .09 .551 -.23 .12
.00 -.16 .07 .012 -.29 -.04
1.00 -.28 .09 .002 -.45 -.11
Note. N = 869. CI = confidence interval; LL = lower limit; UL = upper limit. Affirmative action perception was mean centered. Work experience years had a number of missing answers which explain N.
Finally, as reported in Table 3, the interaction of gender and affirmative action perception on the impostor phenomenon was not significant (p = .598). Consequently, affirmative action perception did not moderate the positive relationship of gender with the impostor phenomenon. This result lead to rejecting hypothesis 3b which states that the positive relation between female gender and impostor phenomenon is stronger amongst individual with higher affirmative action perception.
To conclude, as presented in Table 5, the index of moderated mediation was not significant (CI [-.02, .01]). Therefore, there was no moderated mediation or conditional indirect effect of affirmative action in this study.
Table 5
Moderated Mediation Analysis
Index of moderated mediation
Affirmative action perception Index SE LLCI ULCI
-.01 .01 -.02 .01
Note. N = 869. CI = confidence interval; LL = lower limit; UL = upper limit. Affirmative action perception was mean centered. Work experience years had a number of missing answers which explain N.
Exploratory analysis
Given that affirmative action perception had a significant positive correlation with number of contractual weekly working hours (r = .18, p < .001), it can be argued that full time employees may perceive affirmative action differently from employees working only three days a week. Therefore, as an exploratory analysis, the moderation effect of affirmative action perception on the relations between both gender and creative performance, as well as gender and impostor phenomenon was examined for fulltime employees with a contractual weekly working hours of 32 or more.
This moderated mediator analysis was performed using the same SPSS PROCESS macro model 8 with 5,000 bootstraps and 95% bias correction (Hayes, 2017) as in the main
analysis. This analysis, too, showed that gender had a significant negative relationship with creative performance (β = -.19, SE = .07, CI [-.32, -.05]), gender had a significant positive relationship with impostor phenomenon (β = .59, SE = .11, CI [.38, .81]) and impostor phenomenon had a significant negative relationship with creative performance (β = -.09, SE
= .02, CI [-.13, -.04]). However, the interaction of gender and affirmative action perception on creative performance was not significant (p = .477), and the interaction of gender and affirmative action perception on the impostor phenomenon was also not significant (p = .963). Similarly, as the main analysis, the index of moderated mediation was not significant (CI [-.02, .02]). Thus, in this exploratory analysis there is also no moderation effect of affirmative action and no moderated mediation or conditional indirect effect of affirmative action.
Discussion Summary and discussion of results
Around the world, affirmative action, such as gender recruitment quotas are used in organizations – public or private – to increase diversity on the work floor. Recruiting women through affirmative action policies can prompt other employees to behave differently towards them, affecting these women’s sense of belonging, and consequently their creative
performance. The present study intended to address and examine the relation between impostor phenomenon and creative performance, as well as the relation between affirmative action perception and impostor phenomenon. To do so, this study first sought to establish that there is a relation between gender and creative performance, and then investigated whether impostor phenomenon could explain this relation. The study also explored whether
affirmative action had an influence on the relation between gender and creative performance.
The results of this study – conducted in the Netherlands amongst employees of various organizations on a payroll who worked at least three days a week – indicated that
women’s creative performance was indeed lower than men’s creative performance. The results also confirmed, that women were more likely to experience impostor feelings, a phenomenon that could account for their self-reported low creative performance. Besides, regarding the influence of affirmative action, the results of the study showed a small trend that women who perceived medium or high levels of affirmative action perception reported a lower creative performance than men. However, for lower levels of affirmative action
perception, men and women indicated equal levels of creative performance. Finally, contrary to the expectations, the results did not indicate that women with higher affirmative action perception experienced more impostor feelings.
In line with previous research (Baer & Kaufman, 2008), the current study showed that women’s self-reported creative performance is notably lower than men’s self-reported
creative performance. The fact that these results are replicated in a large sample of diverse employees in the Netherlands, shows that issues pushing women to self-rate their creative performance lower than men, may persist across countries and cultures. A lower self- evaluation may possibly find its roots in a lack of creative self-efficacy, caused by
internalization of gender stereotypes and societal norms. In this respect, this study extends the research on women’s self-limiting behaviors, such as lack of self-efficacy and self-
stereotyping (Dickerson & Taylor, 2000), to the literature on creativity.
Additionally, the current study showed women are more likely to experience that impostor feelings could account for their self-reported low creative performance. In line with Clance and Imes (1978), who originally postulated impostor phenomenon to be mostly present among women, the current study showed that women experienced impostor feelings more than men. Impostor feelings were, previously, associated with lower creative
performance among a sample of Romanian high school students (Dudău, 2013). The present study extends these past findings to a large sample of diverse, highly educated, professionals
in the Netherlands, in which women experiencing impostor feelings had a lower creative performance than men. It can be, thus, argued that because women who experience impostor phenomenon are critical to their own work and performance in general (Hoang, 2013), they would accordingly rate their creative performance lower. Alternatively, it may also be argued that women with higher imposter feelings may lack confidence to perform creatively at work, and that their lower self-reported creative performance may, in fact, reflect their true creative output in their working environment. All in all, this study links the impostor phenomenon and its consequences to the study of creative performance and creativity.
Moreover, the results of the study showed a small trend regarding the influence of perceived affirmative action on the relation between gender and creative performance.
Indeed, for medium and high levels of affirmative action perception, women reported lower creative performance than men, however for low level of affirmative action perception women and men reported similar creative performance. It can, thus, be stated that the higher the affirmative action perception of women, the lower their self-assessed creative
performance. These findings, emphasizing the importance of the perception of affirmative action, concur with the expectations established based on previous research. The current study, therefore, empirically supports the studies of Heilman and her associates who indicated that affirmative action has negative effects on the behavior of women who are aware that they are benefiting from such policies (Heilman et al., 1987, 1991, 1993).
Lastly, the results of the study showed no influence of affirmative action on the relation between gender and impostor phenomenon. These results contradict the assumptions made by linking impostor phenomenon to three harmful effects of affirmative action that were previously outlined by Heilman (1996). It is possible that in this study, women
beneficiaries of affirmative action might not have been exposed to incompetence stigma that would have triggered an increased impostor feeling. Similarly, women might not have had
resentful peers and difficult working relationships that might have contributed to an increased feeling of impostor. It is also possible that the perception of affirmative action was not salient enough to generate negative self-evaluation and increased impostor feeling amongst the participants. Finally, on a practical level, it can be argued that affirmative action perception is a complex is a complex construct that requires a more elaborate operationalization.
Additional methodological issues that could have caused a suboptimal measurement of the affirmative action perception are elaborated in the limitations section.
Limitations and future direction
One limitation of this research is the use of two items for the measurement of affirmative action perception that were specifically created for this study without any pre- testing for validation. Even if it is common practice to create items when existing instruments do not cover a construct of interest, running a pilot is recommended to ensure the questions designed are well understood by respondents (Saunders & Lewis, 2012). Indeed, it seems that the two questions of this survey had a complex formulation and some participants
(approximately 1%) mentioned in their comments that the question was not clear to them. It is desirable to conduct further research on the impact of affirmative action with a different scale – containing more than two items – that would ideally be pre-tested for validity.
Additionally, the data collection of this study was carried out through a survey, asking participants to share their self-perceptions and self-evaluations. For creative performance, self-rating as opposed to evaluation by others (e.g. supervisors, peers) is a controversial method. As Ng and Feldman (2012) described in their meta-analysis, some contextual and personal factors could influence the self-rating of creative performance. Besides, the validity of self-rated creative performance could be compromised due to social desirability bias (Ng
& Feldman, 2012). Additionally, as Hardies, Breesch, and Branson (2012) highlighted, men tend to be overconfident which creates a mismatch between what they believe being true and
what the reality is. Men’s overconfidence could have the impact of increasing the self- evaluation of their creative performance. Therefore, in order to avoid common method bias, percept-percept inflation, and construct validity concerns (Anderson et al, 2014), future studies should seek to replicate the present findings, using peer or supervisor ratings as a measure of creative performance.
Furthermore, the current relational study, using a survey design, does not allow for any causal conclusion (Trochim, Donnelly & Arora, 2016). To better establish whether impostor phenomenon explains why women self-evaluate having a lower creative
performance, an experimental study is needed. Similarly, to investigate whether affirmative action perception is the reason for women to self-rate a lower creative performance, further research with an experimental design could be conducted. As conducting this type of studies in naturalistic settings may pose ethical challenges, lab studies may be a more viable option.
Moreover, even though the selected sample contains participants, working at least three days a week, there was no specific question about the number of employers that each respondent was working for. Affirmative action policies may not be perceived by employees who spend little time in a single organization; or not appear important to employees who cover several part-time jobs with potentially other priorities. It would be, therefore,
interesting for future research to focus on employees working at least three days a week but for one organization only.
Furthermore, the education level of the sample was much higher than the national average of the Netherlands. With 95.6% of the respondents having an HBO Bachelor’s level or higher versus 30% nationally (CBS, 2018), the sample is not truly representative of the population. It can be argued that employees with higher level of education can be more often engaged in creative tasks at work, and therefore perceive their creative performance to be higher. Future research should aim at having a sample with more balanced education level.
Finally, it is to be noted that the data collection of this study started two weeks after the start of the Covid-19 lockdown, in a period when respondents who did not have essential professions were forced to work from home or not to work at all (Rijksoverheid, 2020). The general mindset and changes in working conditions and habits, in addition to the uncertainty of the situation, could have influenced some responses. Approximately 2.1% of the
participants mentioned in the comment section that their working habits, work location, working hours and mood have been impacted by the situation caused by Covid-19 measures.
Suggestions for future research
The current study showed only a marginal association between affirmative action and lower self-report of creative performance among women, compared to men. It also showed no influence of perceived affirmative action on women experiencing more impostor
phenomenon than men. However, the perception of affirmative action in this study is dependent on the judgement of participants – hence subjective. To ensure objectivity on the perception of affirmative action future research should analyze the level of impostor feeling and creative performance for women who were openly hired through affirmative action, in organizations where the policies are commonly known and publicly announced.
Furthermore, it is crucial for any organization, to have a full understanding of factors influencing the creative performance of its employees. Depending on the sector, women occupy a large part of the workforce. Understanding environmental and psychological elements that impact their creative performance, would enable the organizations to find the best way to address and to mitigate decrease in creative performance. Future research should, therefore, investigate what other factors could impact the effects of affirmative action, and consequently women’s creative performance. For example, does women’s organizational engagement and job fit have an effect on their affirmative action perception? Does the gender distribution, for instance the dominance of male gender, increase affirmative action
perception or impostor feeling for women? Does the presence of measures to increase a diverse and inclusive organizational culture impact the effect of affirmative action policies for women?
Practical implications
Impostor phenomenon was initially observed in psychological therapy sessions, and even in recent research it is mostly mentioned as a psychological and personality trait (Sakulku, Alexander, 2011). However, as its impacts are affecting the creative performance of women in the workspace, employers need to find ways to counterbalance and diminish impostor feelings. For example, adopting an appropriate management style can help decrease impostor feelings. As Badawy, Gazdag, Bentley, and Brouer (2018) highlight, employers and employees can benefit from a continuous effort to suppress impostor phenomenon tendencies with an appropriate mentoring and a constructive relationship.
To prevent women from underrating their own creative performance, organizations need to implement measures to reduce gender stereotypes; and more specifically to improve gender fit for the male-type positions that women occupy. To achieve this goal, organizations should reduce managerial bias by assigning organizational responsibilities (e.g. diversity committees, diversity taskforces, and diversity departments), demanding accountability from decision makers, and promoting transparency. The results of such measures are found to be more favorable, when they are combined with diversity trainings, focused on gender
similarities rather than differences (Kalev, Dobbin, & Kelly, 2006; Caleo & Heilman, 2019).
Furthermore, additional steps should be undertaken to reduce the negative impact of affirmative action policies. These steps may include: (1) paying attention to the language used to describe a role in the organization and the characteristics for success in a specific role; (2) highlighting in a prominent manner that gender bias is not the norm in the
organization and is not accepted; and (3) emphasizing the qualifications and competences of
women in the organization. More importantly, finding ways to have an organization culture welcoming and fitting women, is a viable and long-term replacement for affirmative action policies. Examples of actions towards a culture embracing women is the limitation of masculine cues and traditions, and the establishment of family friendly programs –
parenthood leave of absence, flexible working – that are available to both men and women without any backlashes (Heilman & Caleo, 2018).
Conclusion
Gender equality is the fifth United Nations sustainable development goal (UN, 2020), which emphasizes, among other things, the importance of integration and inclusion of women in the workforce. For women to be present on the work floor and treated equally to men, it is important that employers and managers understand the negative effects of affirmative action policies on women’s creative performance. It is also important for employers to realize that environmental and psychological factors can trigger impostor feelings among women, hence negatively affecting their creative performance. Recognizing the relations between these factors and seeking to mitigate their negative effects would support gender equality in the workspace.
References
Amabile, T. M. (1988). A model of creativity and innovation in organizations. Research in Organizational Behavior, 10(1), 123-167.
Anderson, N., Potočnik, K., & Zhou, J. (2014). Innovation and creativity in organizations: A state-of-the-science review, prospective commentary, and guiding
framework. Journal of Management, 40(5), 1297-1333.
Aust, F., Diedenhofen, B., Ullrich, S., & Musch, J. (2013). Seriousness checks are useful to improve data validity in online research. Behavior Research Methods, 45(2), 527-535.
Badawy, R. L., Gazdag, B. A., Bentley, J. R., & Brouer, R. L. (2018). Are all impostors created equal? Exploring gender differences in the impostor phenomenon- performance link. Personality and Individual Differences, 131, 156-163.
Baer, J. (2010). Is creativity domain specific. In J. C. Kaufman & R. J. Stenberg (Eds.), The Cambridge handbook of creativity (pp 321-341).
Baer, J., & Kaufman, J. C. (2008). Gender differences in creativity. The Journal of Creative Behavior, 42(2), 75-105.
Cadinu, M., Maass, A., Rosabianca, A., & Kiesner, J. (2005). Why do women underperform under stereotype threat? Psychological Science, 16(7), 572-578.
Caleo, S., & Heilman, M. E. (2019). What could go wrong? Some unintended consequences of gender bias interventions. Archives of Scientific Psychology, 7(1), 71-80.
Calsamiglia, C., Franke, J., & Rey-Biel, P. (2013). The incentive effects of affirmative action in a real-effort tournament. Journal of Public Economics, 98, 15-31.
CBS. (2018). Society – Figures – Education. Trends in the Netherlands 2018. Retrieved from: https://longreads.cbs.nl/trends18-eng/society/figures/education/
Clance, P. R., & Imes, S. A. (1978). The impostor phenomenon in high achieving women:
Dynamics and therapeutic intervention. Psychotherapy: Theory, Research &
Practice, 15(3), 241-247.
Clance, P. R., & O’Toole, M. A. (1987). The impostor phenomenon: An internal barrier to empowerment and achievement. Women & Therapy, 6(3), 51-64.
Davies, P. G., Spencer, S. J., & Steele, C. M. (2005). Clearing the air: Identity safety
moderates the effects of stereotype threat on women's leadership aspirations. Journal of Personality and Social Psychology, 88(2), 276-287.
Dickerson, A., & Taylor, M. A. (2000). Self-Limiting Behavior in Women: Self-Esteem and Self-Efficacy as Predictors. Group & Organization Management, 25(2), 191-210.
Dudău, D. P. (2013, December 27). The impostor phenomenon and its relation to creative potential among high school students: Could fraudulence feelings regarding
creativity be justified? [Conference paper]. The International Conference "Education and Creativity for a Knowledge-Based Society", Psychology, 2013.
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2394007#page=28
Eagly, A. H., & Steffen, V. J. (1984). Gender stereotypes stem from the distribution of women and men into social roles. Journal of Personality and Social
Psychology, 46(4), 735-754.
Findlay, C. S., & Lumsden, C. J. (1988). The creative mind: Toward an evolutionary theory of discovery and innovation. Journal of Social and Biological Structures, 11(1), 3-55.
Friedman, H. H., Friedman, L. W., & Leverton, C. (2016). Increase diversity to boost creativity and enhance problem solving. Psychosociological Issues in Human Resource Management, 4(2), 7-33.
Gündemir, S., Martin, A. E., & Homan, A. C. (2019). Understanding diversity ideologies from the target's perspective: A review and future directions. Frontiers in
Psychology, 10(282).
Hardies, K., Breesch, D., & Branson, J. (2012). Male and female auditors' overconfidence.
Managerial Auditing Journal, 27(1), 105-118.
Hayes, A. F. (2017). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford publications.
Heilman, M. E. (1996). Affirmative action's contradictory consequences. Journal of Social Issues, 52(4), 105-109.
Heilman, M. E. (2012). Gender stereotypes and workplace bias. Research in Organizational Behavior, 32, 113-135.
Heilman, M. E., & Caleo, S. (2018). Combatting gender discrimination: A lack of fit framework. Group Processes & Intergroup Relations, 21(5), 725-744.
Heilman, M. E., Kaplow, S. R., Amato, M. A., & Stathatos, P. (1993). When similarity is a liability: Effects of sex-based preferential selection on reactions to like-sex and different-sex others. Journal of Applied Psychology, 78(6), 917-927.
Heilman, M. E., McCullough, W. F., & Gilbert, D. (1996). The other side of affirmative action: Reactions of nonbeneficiaries to sex-based preferential selection. Journal of Applied Psychology, 81(4), 346-357.
Heilman, M. E., Rivero, J. C., & Brett, J. F. (1991). Skirting the competence issue: Effects of sex-based preferential selection on task choices of women and men. Journal of Applied Psychology, 76(1), 99-105.
Heilman, M. E., Simon, M. C., & Repper, D. P. (1987). Intentionally favored, unintentionally harmed? The impact of gender based preferential selection on self-perceptions and self-evaluations. Journal of Applied Psychology, 72(1), 62-68.
