1744-7941.12224.pdf

Asia Pacific Journal of Human Resources (2019) , 

doi:10.1111/1744-7941.12224

Effects of supervisor gender on promotability of female managers Hyondong Kim Dongguk University-Seoul, Korea Tong Hyouk Kang University of Memphis, USA

In order to more fully understand the importance of same-gender competition in female supervisor– subordinate working relationships, this study examined the effects of supervisor gender on promotion probabilities for Korean female managers with or without managerial qualifications (e.g. mentoring participation and job ranks). Using a balanced panel sample of 568 Korean female managers in each of four waves (in total, 2272 female managers over 7 years), we conducted a multinomial logistic regression analysis to estimate the promotability of female managers. Our findings showed that mentoring participation negatively affects promotion probability for female managers when they have female supervisors (vs male supervisors). Competitive interdependence can be exacerbated between female managers and female supervisors, especially when they are qualified to compete for the same resources and opportunities, which are limited for female managers and supervisors. Keywords: competition in same-gender relationships, job ranks, mentoring, multinomial logistic regression analysis, promotability of Korean female managers, supervisor gender Key points 1 Mentoring participation reduced promotion probabilities of female managers with female supervisors compared to male supervisors. 2 Competitive interdependence between female managers and female supervisors is exacerbated especially when managers are qualified to compete for resources and opportunities that are limited for the group of females. 3 Organizations should make efforts to create and sustain environments where their performance evaluations are based on job-related skills and abilities, and objective achievements, rather than gender-based perceptions.

Career progression opportunities providing females with the ability to access and occupy powerful positions have steadily increased, as female employees and managers have Correspondence: Hyondong Kim, Dongguk Business School, L419, Dongguk University-Seoul, 30 Pildong-ro, 1-gil, Jung-gu, Seoul 04620, Korea; e-mail: [email protected] Accepted for publication 19 February 2019. © 2019 Australian HR Institute

Asia Pacific Journal of Human Resources 

accumulated requisite working skills and experience (Abendroth et al. 2017). Females increasingly take part in supervisory roles with increasingly improved educational attainment and objective achievements. In female managers’ working relationships with supervisors in particular, females may undertake changes from traditional concepts of the supervisor–subordinate relationship, as their supervisors over the past decades have mostly been males. Interpersonal interactions between female supervisors and their female managers elicit different reactions from those involving male supervisors, because the category of gender seems to be a salient factor leading to differences in shaping and developing attributes and behaviors between females and males (Heilman and Parks-Stamm 2007). Gender roles, which are closely associated with how working individuals perceive themselves and others, may account for the evaluations and responses that supervisors and managers elicit toward each other (Sheppard and Aquino 2017). For female supervisors and managers, the shared gender identification between them defines characteristics of independent working relationships, which appear to be distinguishable from male colleagues: demographic similarity can lead to social comparison between each other or motivate them to foster gender identification (Heilman and Parks-Stamm 2007; Hoobler, Wayne and Lemmon 2009). Two contradictory perspectives – competition vs co-operation – can explain same-gender working interdependence in female supervisor– manager relations. (As this paper focuses on the supervision of female managers, the term ‘supervisor’ refers to higher level managers.) The view that women are depicted as jealous and unsupportive of the career progress of other females has been widely disseminated in mainstream media (Sheppard and Aquino 2017). The competitive natures of women’s work relationships shape and intensify the intentions of female supervisors to thwart the careers of female managers (Lee, Kesebir and Pillutla 2016). Many studies have found evidence of the ‘queen bee syndrome’, which refers to senior female leaders ignoring and even obstructing the career advancement of female managers (Lee, Kesebir and Pillutla 2016). Female supervisors also must confront invisible barriers – generally termed the ‘glass ceiling’ – to career advancement. Female supervisors are likely to disentangle themselves from gender identification in order to circumvent the career penalties incurred by gender membership. Moreover, the presence of hidden quotas for C-suite females forces female supervisors and female managers to engage in competitive interdependent relationships (Dezso, Ross and Uribe 2016). Because female supervisors may read signals of competence from female subordinates as threatening, they may react negatively to strongly competent female managers (Inesi and Cable 2015). By contrast, gender similarity fosters in-group identification, which defines and shapes the relationship characteristics between female managers and their female supervisors. Female supervisors are willing to support the career advancement of female managers that they perceive as belonging to the same gender category. Lee, Kesebir and Pillutla (2016) reported that female workers strongly reject competitive relationships with same-gender co-workers, because relational norms for female workers involve collaborative interdependence. In a similar vein, Vial, Brescoll and Napier (2018) found evidence of 2

© 2019 Australian HR Institute

Hyondong Kim and Tong Hyouk Kang

in-group favoritism between female supervisors and female managers indicating that female supervisors are more likely to make favorable assessments of female managers than their male counterparts. Existing studies have produced mixed findings regarding whether workplace relationships between female supervisors and female managers produce negative consequences or facilitate career success (Lee, Kesebir and Pillutla 2016). Testing female managers’ managerial qualifications makes the contribution of filling gaps in the relationships between supervisor gender and their promotability. First, female managers’ qualifications are hypothesized to impact their career outcomes, as they can either be seen as competitive threats or positive signals about gender identity (Inesi and Cable 2015). We defined female managers with qualifications as those receiving mentorship and those positioned in higher ranking jobs. Mentoring provides guidance regarding the behaviors that are appropriate for female managers in organizational contexts. Since females have limited opportunities for promotion to higher managerial levels, females in the higher ranks are seen as having the abilities and skills necessary to reach the upper echelons (Leslie, Manchester and Dahm 2017). Female managers who possess and demonstrate managerial qualifications are more likely to achieve highly and positively contribute to organizational outcomes. These female managers may believe that they deserve positive evaluations from female supervisors. However, in contrast to these beliefs, research has validated the counterintuitive prediction that female supervisors tend to penalize female managers with managerial qualifications, by triggering the comparison process between in-group and out-group members (Sheppard and Aquino 2017). Testing mentoring participation and job rank as proxies for managerial qualifications is a means of delineating the mechanisms through which female supervisors come to block or facilitate the career advancement of female managers. Second, this study used a sample of Korean female managers in order to determine the importance of supervisor gender in determining the promotability of female managers with competences and qualifications. The proportion of female workers in managerial and professional occupations has increased rapidly in Korea (Kim et al. 2010). However, the Global Leadership and Organizational Behavior Effectiveness project reported that Korea has the lowest scores for gender egalitarianism (House et al. 2004). Female managers in Korea encounter conflicts stemming from traditional notions of gender roles that diminish their opportunities for upward mobility in the workplace (Kim et al. 2010). As Asian females have steadily advanced into management positions, normative gender expectations in Asian societies have increasingly led to questions about whether women can effectively perform managerial roles (Kim et al. 2010, 2013). This study suggests that Asian companies should consider how supervisor gender relates to managerial qualifications in thwarting or facilitating the career prospects of female managers. We tested our hypotheses using a 2008–2014 dataset from the Korean Women Manager Panel (KWMP) collected by the Korea Women Development Institute (KWDI), a governmental research institution focused on enacting and institutionalizing governmental policies and legislation to endorse gender equality in workplaces. KWMP allows us to conduct longitudinal © 2019 Australian HR Institute

3

Asia Pacific Journal of Human Resources 

analysis that contains changes in job ranks of female managers over 7 years in four waves from 2008 to 2014, biannually. Using the KWMP dataset, this study can analyze how the interactions of supervisor gender with job ranks and mentoring participations determine the career tracks of female employers over years. Using the KWMP dataset, the current study can dig more deeply into the concept of competitive vs co-operative interdependence so as to provide further scholarly insight into how supervisors evaluate the promotability of female managers – particularly those with managerial qualifications – positively or negatively. We expect that the results of the current study present empirical evidence on the paradoxical role of female supervisors in the career prospects of female managers (Inesi and Cable 2015). Competition vs co-operation in relationships for female supervisors and managers The awareness of gender begins in childhood, and our understanding of gender constructs extends throughout our lives (Wood and Eagly 2015). People generally place themselves in one of two gender categories, male or female. The respective psychological traits ascribed to men and women cause and extend stereotypical beliefs about each gender’s respective characteristics and abilities (Wood and Eagly 2015). Many assume that managerial roles require masculine traits, and those who make such assumptions may view stereotypically female psychological traits (e.g. communal, nurturing, friendly, and concerned about others) as incongruent with such requirements and obstacles to their success in managerial positions (Heilman and Parks-Stamm 2007; Hoobler, Wayne and Lemmon 2009). Because gender is a salient social category on the individual level, demographic similarity between female supervisors and female managers causes them to perceive themselves as in-group members; however, the stereotyped beliefs that characterize females as unfit for managerial tasks and activities accompany this identification (Chattopadhyay, George and Lawrence 2004). The stereotypes associated with their gender lead female supervisors to distance themselves from their gender group in order to avoid the negative effects of gender stereotyping (Sheppard and Aquino 2017). Moreover, placing women in upper level positions enables organizations to fulfill diversity requirements, comply with laws, and support the values of diversity (Leslie, Manchester and Dahm 2017). Requirements to recruit and select females for upper echelon positions reveal the gender-matching processes used to produce and promote gender diversity in the workplace (Dezso, Ross and Uribe 2016). The likelihood of replacing female supervisors increases when female subordinates with sufficient managerial skills and abilities aspire to attain higher positions. Female supervisors feel threatened by successful female subordinates, as their successes may negatively impact the supervisors’ efforts to advance their own careers (Derks et al. 2011). Shared gender identity can trigger social comparison processes between female supervisors and female subordinates (Sheppard and Aquino 2017). When same gender comparisons exacerbate negative self-evaluations, female supervisors tend to penalize female subordinates so as to 4

© 2019 Australian HR Institute

Hyondong Kim and Tong Hyouk Kang

invalidate threatening comparisons (Sheppard and Aquino 2017). Consequently, female supervisors choose to reject in-group gender identification, as opposed to showing gender favoritism (Hoobler, Wayne and Lemmon 2009; Wood and Eagly 2015). Demographic similarity with female supervisors fosters gender identification, which is expected to mitigate the effects of gender stereotyping on the career progress of female managers. However, by contrast, as gender is a salient social category to individuals, demographic similarity between supervisors and managers can lead them to perceive themselves as in-group members (Chattopadhyay, George and Lawrence 2004). Women supervisors tend to make a positive valuation of female managers in order to enhance self-esteem as well as reduce uncertainty in their personal identity (Chattopadhyay, George and Lawrence 2004). Female supervisors are more likely to make commitments to female managers’ career development in order to enhance their own self-concepts (Wood and Eagly 2015). In this study, we regarded mentoring participation and high job rank as evidence of female managers’ competence and accomplishment on the job (Inesi and Cable 2015). Through mentoring relationships, female managers access resources that are critical to their career progress (Ramaswami et al. 2010). Females positioned in high management positions have demonstrated managerial capabilities across career paths. Female managers who have received mentoring or hold high-ranking jobs are regarded as preferred candidates for promotion to upper echelon positions. We expected that adopting the concept of competitive vs interdependence would enable us to identify conditions – mentoring participation and job rank – in which female supervisors can either become ‘queen bees’ or serve in committed coaching roles, particularly in relation to female subordinates with managerial qualifications. Hypotheses development Supervisor gender and mentoring program Female managers’ lack of access to senior leadership positions can be partially attributed to not receiving sufficient opportunities to support their career advancement (Abendroth et al. 2017). Female managers are thought to lack the experience and competence to maintain and develop motivations to achieve career success (Hoobler, Wayne and Lemmon 2009). Through participation in mentoring programs, female managers can access developmental opportunities to acquire requisite career-related skills and abilities, which can ameliorate gender stereotyping effects (Hoobler, Wayne and Lemmon 2009; McDonald and Westphal 2013). Demographic similarity with supervisors is expected to help female managers cope with the tendency to devalue their management skills and abilities. As female supervisors attach meanings or self-conceptions to gender, they are more likely than male supervisors to perceive that mentoring participation equips female managers with career-related skills and leadership competencies; this results in positive career consequences for female managers. © 2019 Australian HR Institute

5

Asia Pacific Journal of Human Resources 

However, contrary to original expectations, female supervisors’ judgments of female managers are less favorable and more gender-biased, because they perceive themselves as competing for status and thus view their subordinates as threats (Lee, Kesebir and Pillutla 2016). The gender stereotypes embedded in male-dominated work contexts are explicitly and implicitly reflected in corporate decisions to appoint females to upper echelon positions. Corporations must comply with laws that aim to develop and sustain gender diversity in management positions; the persistence of gender-based beliefs leads corporations to maintain quotas for females in top management positions: the employees replacing departing senior female leaders are more likely to be of the same gender (Dezso, Ross and Uribe 2016; Tinsley et al. 2017). Through mentoring participation, female managers acquire sponsorship and coaching for career development, and enhance their senses of interpersonal relationship and emotional bonds. As female managers exposed to gender discrimination have feelings of anxiety and disturbing feelings, mentoring participation helps them cope with the negative effects of gender discriminating experiences (Ragins et al. 2017). Mentoring programs inspire female managers to articulate their sense of purpose and achievements in pursuing and advancing career paths. They also provide coaching in leadership competencies and instill confidence, thereby inspiring female subordinates to target senior leadership positions that they previously had deemed out of reach (Ramaswami et al. 2010). Gender is a salient cue that can be an important criterion during the selection and promotion decision process (Tinsley et al. 2017). As stereotypical beliefs prescribe female gender-type jobs that fit with feminine characteristics, selection and promotion decisions that favor female candidates are tilted toward functional positions in which female managers have previously achieved success (Lee, Kesebir and Pillutla 2016). Mentoring participation helps female managers quickly and accurately learn about proper behavioral norms in the context where most corporate elite members are male managers (McDonald and Westphal 2013). Female managers who receive mentoring are particularly likely to be included in candidate pools for powerful and influential positions (Tinsley et al. 2017). Female supervisors regard female managers who participate in mentoring programs as potential threats to their careers, and their relationships with same-gender managers are therefore marked by competition (Lee, Kesebir and Pillutla 2016). Since female managers with mentoring experience represent potential competition for female supervisors, female supervisors are more likely to be critical of their performance outcomes in seeking to define themselves as masculine and distance themselves from femininity (Hekman et al. 2017; Lee, Kesebir and Pillutla 2016). In order to neutralize threats to their own career aspirations, female supervisors penalize female managers who undergo mentoring experiences, thereby preserving their status against competitive feelings from female colleagues. Hypothesis 1: Mentoring participation negatively moderates the relationship between supervisor gender and the promotability of female subordinates. Female supervisors reduce the promotability of female subordinates who participate in mentoring programs over those who do not. 6

© 2019 Australian HR Institute

Hyondong Kim and Tong Hyouk Kang

Supervisor gender and job ranks The job ranks of female managers can be interpreted as evidence of competence, as access to powerful and influential positions have remained limited for female employees and managers. Gender stereotyping accounts for considerable differences in how management performance is evaluated as well as how rewards are allocated (Hekman et al. 2017). As compared with male counterparts with equivalent skills and talents, female managers are stereotyped as having fewer and lower level managerial capabilities, which reduces their likelihood of selection for promotion (Hoobler, Wayne and Lemmon 2009; Lyness and Judiesch 2014). Thereby, females in higher positions must have demonstrated significant capabilities and accumulated experiences throughout their careers, as they are prone to gender-biased perceptions about managerial qualifications. As female managers’ performance and qualifications can be observed and evaluated by organizations, job ranks can reveal female managers’ potentials. Females who hold higher ranking positions can be viewed as possessing the capability to ascend into senior leadership positions. In recent years, the increasing attention to gender diversity has boosted and sustained demand for female managers with management potential. Legal provisions impose senior female manager quotas so as to address the scarcity of females in upper level positions and achieve diversity goals (Tinsley et al. 2017; Zhang and Qu 2016). Corporations must identify and promote female managers who possess and develop the potential to ascend into upper level positions. However, relative to their male counterparts, female supervisors have limited opportunities to become part of the corporate elite. Gender-matching processes may play a significant role in ensuring gender parity in upper level positions; female candidates replace departing female top managers (Tinsley et al. 2017). People tend to think of themselves as belonging to either one of the two gender categories of men or women, as gender identity is salient and visible (Wood and Eagly 2015). Ascribing gender in-group attributes to themselves reinforces their belonging to the gender in-group and confers positive gender in-group identification (Wood and Eagly 2015). Female supervisors expect fierce competition with same-gender working colleagues for top level promotions, as gender discrimination against females in senior leadership positions is still prevalent (Abendroth et al. 2017; Tinsley et al. 2017). Female supervisors tend to view female managers in high-ranking positions as strong competitors for powerful positions, making intense competition between female managers with management potential and female supervisors likely. Female supervisors take advantage of their supervisorial positions to criticize the competence of female managers in high-ranking positions. Thus, we predicted that female supervisors would be negatively disposed toward the promotability of female managers with high job ranks, because the existence of genderbased quota can explicitly or implicitly constrain career development opportunities for female groups (Dezso, Ross and Uribe 2016). Hypothesis 2: Job rank negatively moderates the relationship between supervisor gender and the promotability of female managers. Female supervisors are more negatively disposed toward the promotability of female managers when those managers hold high-ranking positions. © 2019 Australian HR Institute

7

Asia Pacific Journal of Human Resources 

Methods Participants and procedure Female managers are underrepresented in Korean companies. More research is needed to promote career development for female managers. Since 2007, the KWDI has conducted the KWMP in order to gather comprehensive data about work- and family-related factors. The KWDI formed the KWMP research team, which included KWDI researchers, professors, and consultants, to confirm the reliability and validity of the KWMP questionnaire items. The KWMP provided their data to researchers so as to help identify and develop ways to advance the careers of female managers (Kim et al. 2010, 2013). The KWMP used a targeted-sampling strategy to identify female-manager samples. It also developed female-manager samples based on industries (manufacturing, retailing, finance, and service) and company sizes (100–299, 300–999, 1000–1999, and more than 2000 workers) (Kim et al. 2013) prior to conducting the survey in 2007. Those conducting the survey selected a sample of 2000 female managers from 300 companies. Between 2008 and 2014, the panel conducted a large-scale survey biannually in four waves. Of the 2000 female managers selected in total, 1806 responded to KWMP at wave 1 (2008). Of these 1806 female managers, as more than 1000 female managers changed occupations or quit to be housewives, temporary workers, or self-employed workers, only 824 female managers remained in KWMP by 2014 (Appendix 1). Of these 824 female managers, 348 took maternal leave and 862 changed their jobs: 256 female managers experienced career interruptions because they took maternal leave (89 females), changed their jobs (179 females), did both (61 females), or temporarily failed to make contact (49 females). The total number of female managers in the balanced panel KWMP is 2272 female managers (568 9 4 = 2272) over 7 years (2008, 2010, 2012, and 2014) at four waves – 568 of whom participated in each wave. With the balanced panel data of 2272 female managers over 7 years, we can analyze how interrelated relations among supervisor gender, mentoring participations, and job ranks affect the job-rank changes of those female managers. Measures Supervisor gender Supervisor gender served as a dummy variable. We coded gender as 1 for female supervisors and 0 for male supervisors. Mentoring participation We asked female managers whether they participated in mentoring programs and coded mentoring program participation as 1 for participation and 0 for no participation.

8

© 2019 Australian HR Institute

Hyondong Kim and Tong Hyouk Kang

Job rank We measured job rank by asking female managers to select their current job rank from five possible categories: 1) junior manager; 2) manager; 3) junior director; 4) director; and 5) executive. Promotion/no change/demotion The KWMP included job-rank data in a 5-scaled single item (1 = junior manager; 2 = manager; 3 = junior director; 4 = director; 5 = executive). This study used job-rank data to gather information about promotion/no change/demotion for 2 years. We calculated jobrank changes by subtracting current job-rank data (2008, 2010, and 2012) from future job-rank data (2010, 2012, and 2014) as follows: promotion (current – past > 0), no change (current – past = 0), and demotion (current – past < 0). For example, one female manager was junior director in 2010 and had been promoted to director by 2012. As junior director was recorded as ‘3’ and director was recorded as ‘4’, subtracting junior director in 2010 from director in 2012 equaled ‘1’, which indicates a ‘promotion’. Another female manager was manager in 2012 and still remained manager in 2014. As manager was recorded ‘2’, subtracting position 2012 from 2014 equaled ‘0’, which refers to ‘no change’. Subtracting current job ranks from next-wave job ranks generated values ranging from 2 to +2. Negative values of job-rank changes, such as ‘ 2’ or ‘ 1’, are regarded as a demotion; a zero value of job-rank change refers to no change; and a positive value of job-rank changes, such as ‘2’ or ‘1’, is regarded as a promotion. The three dummy-coded variables – promotion/no change/demotion – represent job-rank changes, which are multinomial variables. Control variables The control variables included age, education, job functions, industry memberships, workplace discrimination, supervisor changes, ratios of female employees in workplace, CEO gender, and family unsupportive work environment. We measured education as a continuous variable: 1) high school graduate; 2) 2-year college graduate; 3) 4-year university graduate; 4) master’s degree; and 5) PhD. Female managers tend to tilt toward jobs or industries that are regarded as being a good fit for feminine characteristics. The KWMP had 18 job functions: 1) R&D; 2) management planning; 3) management information system; 4) supply chain management; 5) asset management; 6) law; 7) human resource management; 8) accounting/finance; 9) marketing; 10) domestic sales; 11) overseas sales; 12) advertising; 13) customer service; 14) operation management; 15) industrial technology; 16) manufacturing & engineering; 17) environmental management; and 18) quality management. We categorized job functions into four dummy-coded variables: R&D, management, marketing, and other functions. The management function included management planning, management information system, supply chain management, law, human resource management, and accounting/finance. The marketing function included domestic sales, overseas sales, marketing, customer service, and advertising. We categorized industry memberships into manufacturing, finance, retailing, and other industries. The © 2019 Australian HR Institute

9

Asia Pacific Journal of Human Resources 

career prospects of female managers were vulnerable to the extent to which HR decisions regarding the female managers were made fairly and without gender bias. The KWMP asked female managers whether they had experienced discrimination in five areas: hiring, job assignment, compensation, training/education, and promotion decisions. We measured each organization’s female employee ratios by dividing the total number of female employees and managers by the total number of employees and managers in the organization. We dummy coded the five workplace discrimination questions. The KWMP asked female subordinates’ whether their supervisors had changed. With this item, we measured whether supervisors had changed in the preceding 2 years. The ownership and traditionality values were closely associated with the assumptions about the traditional model – that work and family domains are incompatible for female employees and managers. As the KWMP dataset did not assess the ownership of the company and traditionality value, CEO gender and family unsupportive working environment from the KWMP dataset were used to represent company ownership and traditionality value. CEO ownership was dummy-coded as ‘1’ for a female CEO and ‘0’ for a male CEO. Items assessing family unsupportive work environment were adopted from Allen (2001): 1) long hours inside the office are the way to achieve advancement; 2) the way to advance in this company is to keep non-work matters out of the workplace; 3) attending to personal needs, such as taking time off for sick children, is frowned upon; and 4) work should be the primary priority in a person’s life. Results Descriptive statistics and correlation Table 1 presents the descriptive statistics and correlations for the study variables. Participant ages ranged from 23 to 60; individuals in their 30s and 40s accounted for 62% and 33% of the sample, respectively. One thousand two hundred twenty-seven female managers were university graduates (54.01%); 375 were college graduates (16.5%); and 334 had master’s degrees (14.70%). Those who held R&D positions accounted for 14.17% (322 managers) of the sample; 43.62% (991 managers) held management positions; and 28.65% (651 managers) held marketing positions. Work discrimination averaged 0.78: 25.5% of female managers had experienced discrimination in promotion decisions; 14.8%, 15.8%, and 15.9% of female managers had experienced discrimination in job assignments, compensation, and workplace training decisions, respectively. Twenty-five percent of female managers worked in manufacturing, 39% worked in finance, and only 6% worked in retailing. Female employees and managers accounted for 32% of all employees. Twenty-eight percent of female subordinates experienced supervisor changes over the 6-year and four-wave duration of the study. Only six percent of all CEOs were females. Three hundred thirty-four female managers had female supervisors (14.7%), and another 1938 female managers had men as supervisors (85.3%). One thousand one hundred eighty-nine female managers participated in mentoring programs (52.3%), while 10

© 2019 Australian HR Institute

© 2019 Australian HR Institute

.00 .03 .09 .02

.33 .33 .11 .04 .00 .01 .00 .01 .10 .00 .02 .00 .00 .00 .00 .03 .02

3

.01 .05 .12 .14

.33 .12 .01 .00 .01 .03 .00 .34 .00 .01 .00 .00 .00 .00 .02 .09

4

.02 .04 .21 .07

.35 .06 .01 .02 .03 .00 .06 .03 .03 .00 .00 .00 .00 .01 .04

5

.07 .09 .60 .01

.04 .01 .07 .08 .01 .11 .07 .08 .08 .13 .03 .04 .02 .04

6

.00 .01 .23 .08

.18 .01 .04 .07 .04 .10 .03 .05 .01 .07 .09 .00 .03

7

.00 .03 .02 .01

.25 .26 .08 .07 .06 .07 .02 .11 .06 .17 .01 .01

8

.06 .01 .04 .00

.39 .16 .01 .01 .01 .01 .05 .06 .01 .03 .03

9

.03 .06 .06 .00

.24 .11 .03 .02 .01 .19 .01 .20 .05 .01

Promotion/No change/Demotion is multinomial variable: Promotion = 1; No change = 0; Demotion =

.02 .07 .24 .05

.35 .50 1.00 .52

2

1.

.03 .00 .00 .01

.25 .03 .03 .08 .01 .01 .01 .00 .01 .05

10

.00 .05 .17 .34

.00 .03 .04 .05 .06 .05 .00 .04

11

.08 .02 .05 .08

.08 .00 .00 .01 .01 .00 .08

12

.17 .07 .09 .03

.16 .26 .00 .08 .14 .07

13

.06 .12 .11 .00

.45 .14 .39 .13 .03

14

.01 .12 .02 .02

.19 .52 .03 .14

15

.07 .03 .04 .02

.17 .05 .03

16

.03 .03 .09 .01

.01 .05

17

.03 .07 .01 .02

.01

18

20

21

22

.08 .02 .00 .04 .00 .11 .00 .03 .03 .35

19

n = 2272. All correlations above 0.04 are significant at p < 0.05; all correlations above 0.06 are significant at p < 0.01. Supervisor gender (1 = female, 0 = male), mentoring (1 = yes, 0 = no), job functions, industry memberships, supervisor changed (1 = change, 0 = no change) and CEO gender (1 = female, 0 = male) were dummy-coded (1 = yes, 0 = no).

a

.33 .33 .33 .35 .01 .02 .05 .06 .34 .30 .03 .03 .00 .00 .00 .00 .01 .02

1

.43 .43 .43 .43 4.93 .90 .35 .45 .45 .13 .44 1.20 .17 .43 .48 .23 .46 .24 .74

Mean SD

1 2008 .25 2 2010 .25 3 2012 .25 4 2014 .25 5 Age 37.88 6 Education 2.70 7 R&D .14 8 Management .28 9 Marketing .28 10 Others 11 Supervisor changed .28 12 Work discrimination .78 13 Female ratios .32 14 Manufacturing .25 15 Finance .38 16 Retailing .06 17 Other industries .31 18 CEO gender .06 19 Family unsupportive 2.98 work environment 20 Supervisor gender .14 21 Mentoring .52 22 Job ranks 3.17 23 Promotion/No change/ .22 Demotiona

Variables

Table 1 Descriptive statistics and correlation table

Hyondong Kim and Tong Hyouk Kang

11

Asia Pacific Journal of Human Resources 

1083 had no opportunity to take part in mentoring programs (47.7%). One hundred and seventy-five female managers had female supervisors and participated in mentoring programs (7.88%). The sample included 654 junior managers (28.79%), 847 managers (37.28%), 524 junior directors (23.06%), and 212 directors (9.33%); only 35 female managers held executive positions (1.54%). Job-change records ranged from +2 to 2, and 1550 managers experienced no change in status (68.22%). One hundred nine female managers had received demotions (5.9%) and 613 received promotions (27%). Multinomial logistic regression analysis Multinomial logistic regression analyses analyze nominal dependent variables. Nominal variables represent discrete sets of alternative choices without unique ordering (RabeHesketh and Skrondal 2012). We used a multinomial distribution to model probabilities that fell into multiple nominal variable categories. Given S = 3 categories, two intercepts and two coefficients are required to specify the probabilities of taking three-category values (Rabe-Hesketh and Skrondal 2012). We designated ‘No change’ as the base outcome for ‘Promotion’ and ‘Demotion’ by setting both the intercept and coefficient for those categories to zero. The parameters for No change can be used to compare between the alternatives of Promotion and Demotion, as they reveal the influence of supervisor gender and mentoring participation on promotions/demotions. Test of hypotheses Table 2 summarizes the multinomial logistic regression results. Model 1 analyzed the control variables and model 2 tested the main effects of independent variables, including supervisor gender, mentoring participation, and job rank. Model 3 examined the two-way interaction effects between supervisor gender, and mentoring participation as well as job rank on the promotability of female managers. We conducted a Hausman test to check the assumption of the independence of irrelevant alternatives (IIA). The Chi-square values for the three groups – Demotion (v2 = 5.422, p > 0.10), No change (v2 = 16.22, p > 0.10), and Promotion (v2 = 15.22, p > 0.10) – were not significant, which suggested no violations of IIA assumptions. Wald tests for combining alternatives produced statistically significant outcomes given a pair of alternatives (Demotion and No change: v2 = 68.06, p < 0.05; Demotion and Promotion: v2 = 48.90, p < 0.05; No change and Promotion: v2 = 42.44, p < 0.05). Analyzing the three categories of multinomial variables separately proved more valid than combining the categories. Of the control variables, work discrimination experience (promotion: B = 0.15, p < 0.001), education (Promotion: B = 0.22, p < 0.01), and supervisor changes (Promotion: B = 2.58, p < 0.001) were significantly related to promotion and demotion probabilities. Meanwhile, among the independent variables, neither supervisor gender (Promotion: B = 0.17, p > 0.10) nor mentoring participation (Promotion: B = 0.03, p > 0.10) was significantly related to promotion and demotion probabilities. Tests of the two-way interactions between supervisor gender and mentoring program participation 12

© 2019 Australian HR Institute

Hyondong Kim and Tong Hyouk Kang

showed negative values in predicting promotion probabilities (B = 0.77, p < 0.05). The fact that female managers participating in mentoring programs under female supervision had lower promotion probabilities partially supported hypothesis 1. However, supervisor gender and job rank had no significant interactions (B = 0.72, p > 0.05), failing to support hypothesis 2. An odds ratio is the ratio of the probability of a particular event divided by the baseline probability (Rabe-Hesketh and Skrondal 2012). Coefficients represent log oddsratios for the odds of each category vs the baseline category (Rabe-Hesketh and Skrondal 2012). A one-unit increase in independent variable x corresponds to the odds ratio for a particular category vs the base category. The graph in Figure 1 predicted probabilities for promotion when female managers had female supervisors and participated in mentoring programs. They plot the interactive effects between supervisor gender and mentoring participation on demotion, no change, and promotion probabilities, respectively. As shown in Figure 1, promotion probabilities remained flat, ranging from 26% to 28% when female subordinates did not have a chance to participate in mentoring programs. However, the interaction between mentoring participation and female supervisors reduced promotion probabilities from 28% to 9%, augmenting the negative effects on the promotability of female subordinates (Figure 1) and supporting hypothesis 1. Analyzing male managers Korean Women Manager Panel asked female participants to extend invitations to male counterparts who had equivalent skills and positions with them. Eight hundred male managers participated in KWMP in two waves, 2008 and 2010. After excluding missing variables, we had 621 male manager samples. According to the descriptive statistics analyses, the promotion rates of male managers were 25.81%, which did not significantly differ from the promotion rates of female managers, 26.61%. As the KWMP female samples were responsible for managerial and professional occupations, the promotion rates of males and females were similar. As the value of this study lies in the availability of comparison between female and male managers, similar promotion rates between female and male managers can serve as support for the study’s objective of examining the determinants of female manager promotability, not temporary workers or non-managerial employees. As compared with female managers, the analysis of male manager samples failed to show significant interaction effects of supervisor gender and job ranks as well as mentoring participations. Thus, the findings (see Appendix 2) of the current study indicate that supervisor gender and mentoring participation play crucial roles in the career development of female managers as compared with that of male managers. Discussion The results of the current study show that female supervisors reduced the promotion probability for female managers who were mentored; they also increased the chances for © 2019 Australian HR Institute

13

Asia Pacific Journal of Human Resources 

Table 2 Supervisor gender, and mentoring and job rank Variables Promotion

Demotion

14

Constant 2010 2012 2014 Age Education R&D Management Marketing Supervisor changed Work discrimination Female ratios Manufacturing Finance Retailing CEO gender Family unsupportive work environment Supervisor gender Mentoring Job ranks Supervisor gender 9 Mentoring Supervisor gender 9 Job ranks Constant 2010 2012 2014 Age Education R&D Management Marketing

Model 1

Model 2

Model 3

6.23 (1.75)*** .05 (.15) 1.22 (.19)*** 1.98 (.21)*** 1.31 (.49)*** .22 (.06)** .02 (.18) .02 (.14) .21 (.14) 2.58 (.15)***

16.29 (2.37)*** .05 (.17) 1.48 (.21)*** 2.42 (.24)*** 6.04 (.70)*** .07 (.07) .08 (.20) .02 (16) .37 (.16)* 2.92 (.17)***

16.49 (2.36)* .05 (.17) 1.46 (.21)** 2.41 (.24)*** 6.09 (.71)*** .08 (.07) .06 (.20) .04 (.16) .39 (.16)* 2.91 (.17)***

.15 (.05)**

.12 (.05)*

.12 (.05)*

.29 (.31) .16 (.14) .19 (.14) .44 (.23) .19 (.22) .02 (.07)

.10 (.36) .58 (.16)*** .24 (.16) .88 (.27)*** .26 (.24) .02 (.27)

.09 (.36) .56 (.17)*** .24 (.16) .85 (.27)** .25 (.25) .02 (.08)

.17 (.17)

.06 (.63)

.03 (.12) 1.62 (.10)***

.14 (.13) 1.60 (.10)*** .77 (.34)*

.19 (.18)

7.88 (3.29)* .37 (.23) 4.09 (.76)*** 3.73 (.64)*** 1.45 (.91) .07 (.11) .45 (.34) .21 (.28) .20 (.29) 1.78 (.30)***

18.20 (3.93)*** .40 (.23) 3.80 (.77)*** 3.39 (.65)*** 4.83 (1.13)*** .07 (.12) .48 (.36) .22 (.30) .26 (.30) 1.52 (.32)***

17.86 (3.95)** .39 (.23) 3.83 (.77)*** 3.43 (.65)*** 4.80 (1.13)*** .07 (.12) .56 (.36) .25 (.30) .31 (.30) 1.52 (.32)***

© 2019 Australian HR Institute

Hyondong Kim and Tong Hyouk Kang

Table 2 (continued) Variables Supervisor changed Work discrimination Female ratio Manufacturing Finance Retailing CEO gender Family unsupportive work environment Supervisor gender Mentoring Job ranks Supervisor gender 9 Mentoring Supervisor gender 9 Job ranks LR Chi2(df) Log likelihood

Model 1

Model 2

Model 3

.08 (.08)

.07 (.09)

.08 (.08)

.88 (.71) .14 (.30) .08 (.30) .67 (.43) .04 (.50) .02 (.15)

.86 (.71) .12 (.30) .08 (.30) .63 (.43) .01 (.50) .02 (.15)

.84 (.66) .26 (.29) .17 (.29) 1.05 (.40)*** .07 (.49) 1.05 (.40)*

.63 (.29)*** .01 (.22) 1.01 (.19)***

1.22 (1.03) .02 (.24) 1.16 (.22)*** .04 (.56) .72 (.37)

639.39 (34) 1407.134

1090.97 (40) 1181.343

1099.80 (44) 1176.930

N = 2272. +p < 0.10, *p < 0.05, **p < 0.01, ***p < 0.001. Supervisor and CEO gender (1: female and 0: male), mentoring, job functions and industry memberships were dummy-coded (1: yes; 0: no). No change (0) is base outcome for Demotion ( 1) and Promotion (1).

demotion when female managers occupied upper- and middle-management positions. Female supervisors appear to embrace competitive attitudes toward female candidates for promotion, as the social comparison process is activated for same-gendered female working relations. Female supervisors are less likely to be supportive of the career prospects of women managers, since female managers who participate in mentoring are more qualified to replace supervisors of the same gender (Derks et al. 2011). For female supervisors, female managers who demonstrate managerial qualifications appear to be seen as competing for smaller chances of career success. The penalties that female managers confront in the labor market take into account the significance of the social comparison process with female supervisors. © 2019 Australian HR Institute

15

Asia Pacific Journal of Human Resources 

Research implications It is counterintuitive that female managers who demonstrate managerial qualifications are penalized more than less competent females (Inesi and Cable 2015). The results of the current study can be inferred from the competitive interdependency model that accounts for mixed implications about working relationships between female supervisors and female managers: female supervisors may view competent female managers as a target of social comparison, so those female managers may in turn be threatening to the female supervisors’ career prospects. Female managers who show strong credentials and qualifications appear to be salient and may be victims of penalization from female supervisors. In this study, competitive interdependence theory may predict that the promotability of female managers is contingent upon the supervisor gender and manager qualifications. Our research has compared the career consequences of female managers’ working relations with same- and different-gender supervisors. Female employees and managers who aspire to top management positions confront implicit quotas that intensify competition with female peers (Dezso, Ross and Uribe 2016; Hekman et al. 2017). Competent, conscientious, and credible female managers are more likely to replace other females in upper level positions, as upper level managers target the small number of females occupying senior leadership positions (Dezso, Ross and Uribe 2016). Female supervisors face the potential risk of losing their status as participating in mentoring programs extends room for female managers to ascend to powerful positions in organizations (McDonald and Westphal 2013; Sheppard and Aquino 2017). Female supervisors and female managers may, as a result, engage in zero-sum competition: female managers can ascend to top management positions as replacements for women supervisors; female managers and female supervisors may, therefore, competitively target the small number of positions available to them. In this way, same-gender supervisory relationships can hinder the career success of female managers who participate in mentoring programs. An important extension of this research would be to indicate the presence of competitive interdependence between female supervisors and female managers: female managers’ qualifications may have threatening effects on the career prospects of female supervisors,

Figure 1 Predicted probabilities of promotion based on supervisor gender and mentoring 16

© 2019 Australian HR Institute

Hyondong Kim and Tong Hyouk Kang

as competitive interdependence between females’ working relations is exacerbated by social comparisons with the same gender. Future research should examine when and where female managers’ competitions with same gender co-workers are exacerbated and can come to have harmful effects on their career outcomes (Lee, Kesebir and Pillutla 2016). Practical implications Female supervisors who are interested in mentoring participation of female managers may be expected to improve their managerial talents and capabilities. However, the concept of competitive interdependence raises a question as to whether mentoring can be beneficial or problematic for the career tracks of female managers; thereby, in contrast to original expectations, mentoring experiences may intensify competitive interdependence between female supervisors and female managers (Lee, Kesebir and Pillutla 2016). When institutionalizing a mentoring program within an organization aimed at maintaining and promoting the status and power of female employees and managers, female managers’ mentoring experiences can be an unexpectedly negative externality to the future career prospects of women in such positions (Inesi and Cable 2015; Lee, Kesebir and Pillutla 2016). When female supervisors come to appreciate the effectiveness of mentoring program for female managers, the actions and efforts of female supervisors should acquire recognitions from organizations. Female supervisors may exploit the advantages of a mentoring program to capitalize on the potentials of female managers: coaching and sponsorship from mentoring programs provide opportunities for female managers to learn and accumulate career-related skills and abilities. When female supervisors are supportive of female managers’ mentoring participations, they should believe their activities to be helpful or acceptable to mentoring benefits for female managers to be evaluated favorably (Sheppard and Aquino 2017). Mentoring programs need to be planned and implemented so as to increase female managers’ contributions to team task accomplishments and organizational interests that benefit female supervisors’ career prospects, rather than triggering a rivalry between them. The results of this study indicate that female managers tend to experience career interruptions. Of the 824 female manager samples in wave 4 (2014), 256 female managers could not participate in KWMP during waves 1–3 (2008–2012). The total numbers of maternal leave recipients were 348 of the 824 female managers who participated in wave 4 (2014). Of 348 female managers, 89 experienced career interruptions, which was one-third of the total number of non-participants, 256 female managers. Work–family conflicts may be a source of career changes for other non-participants who reported changing their jobs and occupations, becoming unemployed, starting a business, or being hired as temporaries. Imbalanced states between work and family domains are underlying assumptions about traditional work–family relations. As females are regarded as being responsible for housework and childcare, female managers experience high levels of family interferences with company works. As female managers spend more time and resources to care for family, they are less likely to be viewed as capable of taking managerial leadership © 2019 Australian HR Institute

17

Asia Pacific Journal of Human Resources 

positions (Hoobler, Wayne and Lemmon 2009). Furthermore, female supervisors regard female managers as potential competitors to their career progress. Female supervisors may leverage work–family conflict as shaping and exacerbating negative perceptions of female managers as unfit for managerial roles, as they share gender attributes with female managers. Under female supervisors, family conflicts with company work are more likely to be highlighted as major causes for the lack of fit to promotions. Work–family support programs are designed and implemented to help female managers cope with family interference with company work. However, if competitive interdependence between female supervisors and female managers persists within organizations, a company’s efforts to tackle the work–family challenges of female managers may not be effective. While it is challenging, organizations provide education and training about the existence of higher levels of family interference with work for female managers. After diluting negative perceptions about work–family relations for female managers, work–family support programs can be effective in increasing the possibility of promotions and supporting their career development. Limitations and future research directions This study contributes to scholarly understanding of the ways in which organizations facilitate the career development of female managers by highlighting the importance of supervisors in the promotion probabilities of female managers; however, some limitations warrant attention. First, this study used a male-manager sample to test the significance of the gender difference in the effects of supervisor gender and mentoring programs on the promotability of company managers. Comparing male and female manager samples may help identify the significance of supervisor gender and mentoring programs in career prospects of female managers. However, as KWMP had male manager samples in 2008 and 2012, caution should be taken to interpret the comparison between male and female managers. Second, testing mentoring effectiveness requires an examination of various mentoring program characteristics, such as mentoring function and mentor type. The KWMP dataset did not survey mentoring function variables. However, existing mentoring studies have demonstrated significant interactions between mentoring function and mentor gender (Ragins et al. 2017), and additional investigations of mentoring function could help more fully explain how mentor gender affects female managers’ promotability. A variety of factors that affect mentoring effectiveness, such as time spent mentoring, the frequency of interactions between mentors and proteges, and the mentors’ communication and coaching skills, must be considered in testing mentoring effectiveness. Third, the KWMP asked female managers about their mentoring experience with their direct supervisors, but female managers can also receive mentoring from other members of senior management or even from outside their companies, and mentors other than direct supervisors likely have varying impacts on the promotability of female managers. Fourth, the gendered attributes of each industry context impact the career trajectories of female managers (Ramaswami et al. 2010). The KWMP surveyed employees in four industry membership 18

© 2019 Australian HR Institute

Hyondong Kim and Tong Hyouk Kang

categories: manufacturing, finance, retailing, and others. In order to more fully explain the importance of industry context in mentoring effectiveness for the career success of female subordinates, future studies should consider a variety of industries in which gender stereotypes prevail. Fifth, detailed information about supervisor job types would facilitate a more comprehensive examination of the effects of supervisor gender and mentoring on the promotability of female managers. Finally, supervisors’ perceptions of the promotability of female managers depend on national contexts. In countries with low gender egalitarianism, like South Korea, supervisors have more negative perceptions of female subordinates’ job-related competencies than in countries with high gender egalitarianism, like Sweden. Since levels of gender egalitarianism significantly affect supervisor roles in female subordinate promotability, future studies should compare countries with high and low gender egalitarianism in order to analyze the relationships between supervisor gender, mentoring programs, and promotability for female subordinates. Conclusions The competitive nature of female working relationships poses a dilemma in which the institutionalization of a mentoring program within an organization can support or thwart the career progress of female managers. Gender similarity categorizes female supervisors and female managers into an in-group; but sharing attributes induce feelings of competition between female supervisors and female managers, while it enacts and fosters interdependence (Sheppard and Aquino 2017). The findings of this study highlight the negative impacts of mentoring participation on the career advancement of female managers when their supervisors are females. The results of the study note the ‘queen bee syndrome’, in which powerful women at the top levels of management are not supportive of female managers attempting to climb the ladder. Planning and implementing mentoring programs to facilitate female manager career advancement need to mitigate competitive elements in female supervisor–manager interdependent relationships. Our study explores why female working relationships are characterized as competitive, although a variety of programs and policies are enacted and implemented to target gender equity in organizations. As female managers and supervisors are particularly concerned about the negative perceptions and biased judgments about their gender identity, ‘female’, organizations make efforts to create and sustain environments where their performance evaluations are based on job-related skills and abilities as well as objective achievements, rather than gender-based perceptions. Hyondong Kim (PhD, Ohio State Univ) is a professor in Dongguk University-Seoul. Hyondong Kim teaches organizational behavior and human resource management. Tong Hyouk Kang is a PhD candidate in the Deparment of Management, Fogelman College of Business and Economics, University of Memphis. His research interests include corporate strategy, internationl strategy, and entrepreneurship..

© 2019 Australian HR Institute

19

Asia Pacific Journal of Human Resources 

References Abendroth A-K, S Melzer, A Kalev and D Tomaskovic-Devey (2017) Women at work: women’s access to power and the gender earnings gap. ILR Review 70, 190–222. Allen TD (2001) Family-supportive work environments: the role of organizational perceptions. Journal of Vocational Behavior 58, 414–435. Chattopadhyay P, E George and SA Lawrence (2004) Why does dissimilarity matter? Exploring selfcategorization, self-enhancement, and uncertainty reduction. Journal of Applied Psychology 89, 892–900. Derks B, C Van Laar, N Ellemers and de Groot K (2011) Gender-bias primes elicit queen-bee responses among senior policewomen. Psychological Science 22, 1243–1249. Dezso CL, DG Ross and J Uribe (2016) Is there an implicit quota on women in top management? A large-sample statistical analysis. Strategic Management Journal 37, 98–115. Heilman DE and EJ Parks-Stamm (2007) Gender stereotypes in the workplace: obstacles to women’s career progress. In SJ Correll (ed) Social psychology of gender, 47–77. Emerald Group Publishing Limited, Bingley, UK. Hekman DR, SK Johnson, M-D Foo and W Yang (2017) Does diversity-valuing behavior result in diminished performance ratings for non-white and female leaders? Academy of Management Journal 60, 774–797. Hoobler JM, SJ Wayne and G Lemmon (2009) Bosses’ perceptions of family-work conflict and women’s promotability: glass ceiling effects. Academy of Management Journal 52, 939–957. House RJ, PJ Hanges, M Javidan, PW Dorfman and V Gupta (2004) Culture, leadership, and organizations: the GLOBE study of 62 societies. Sage Publication, Thousand Oaks, CA. Inesi ME and DM Cable (2015) When accomplishments come back to haunt you: the negative effect of competence signals on women’s performance evaluations. Personnel Psychology 68, 615–657. Kim J, T-M Lee, I-S Yang, N-J Kim, S-H Kang and H-J Min (2010) Korea women manager panel 2010. Korean Women’s Development Institute, Seoul, Republic of Korea. Kim N-J, J-S Kim, T-M Lee and M-J Kang (2013) Korea women manager panel 2013. Korean Women’s Development Institute, Seoul, Republic of Korea. Lee SY, S Kesebir and NM Pillutla (2016) Gender differences in response to competition with samegender coworkers: a relational perspective. Journal of Personality and Social Psychology 110, 869– 886. Leslie LM, CF Manchester and PC Dahm (2017) Why and when does the gender gap reverse? Diversity goals and the pay premium for high potential women. Academy of Management Journal 60, 402–432. Lyness KS and MK Judiesch (2014) Gender egalitarianism and work-life balance for managers: multisource perspectives in 36 countries. Applied Psychology: An International Review 63, 96–129. McDonald ML and JD Westphal (2013) Access denied: low mentoring of women and minority first-time directors and its negative effects on appointments to additional boards. Academy of Management Journal 56, 1169–1193. Rabe-Hesketh S and A Skrondal (2012) Multilevel and longitudinal modeling using Stata Volume II: categorical responses, counts, and survival. Stata Press, College Station, TX.

20

© 2019 Australian HR Institute

Hyondong Kim and Tong Hyouk Kang

Ragins BR, K Ehrhardt, KS Lyness, DD Murphy and JF Chapman (2017) Anchoring relationships at work: high-quality mentors and other supportive work relationships as buffers to ambient racial discrimination. Personnel Psychology 70, 211–256. Ramaswami A, G Dreher, R Bretz and C Wiethoff (2010) Gender, mentoring, and career success: the importance of organizational context. Personnel Psychology 63, 385–405. Sheppard LD and K Aquino (2017) Sisters at arms: a theory of female same-sex conflict and its problematization in organizations. Journal of Management 43, 691–715. Tinsley CH, JB Wage, BGM Main and CA O’Reilly (2017) Gender diversity on US corporate boards: are we running in place? ILR Review 70, 166–189. Vial AC, VL Brescoll and JL Napier (2018) Differential support for female supervisors among men and women. Journal of Applied Psychology 103, 215–227. Wood W and AH Eagly (2015) Two traditions of research on gender identity. Sex Roles 73, 461– 473. Zhang Y and H Qu (2016) The impact of CEO succession with gender change on firm performance and successor early departure: evidence from China’s publicly listed companies in 1997–2010. Academy of Management Journal 59, 1845–1868.

Appendix 1 KWMP female sample statistics

Year (total numbers)

Status

Numbers

2008 (1806)

Management positions Change occupations Unemployed Started own business Temporary Management positions Change occupations Unemployed Started own business Temporary Management positions Change occupations Unemployed Started own business Temporary Management positions Change occupations Unemployed Started own business Temporary

1736 41 27 1 1 1275 76 90 4 5 904 78 74 5 7 824 49 49 3 5

2010 (1450)

2012 (1078)

2014 (930)

© 2019 Australian HR Institute

21

Asia Pacific Journal of Human Resources 

Appendix 2 Male managers Promotion

Demotion

Constant Age Education R&D Management Marketing Work discrimination Male ratio Manufacturing Finance Retailing CEO gender Family unsupportive work environment Supervisor gender Mentoring Job ranks Supervisor gender 9 Mentoring Supervisor gender 9 Job ranks Constant Age Education R&D Management Marketing Work discrimination Male ratio Manufacturing Finance Retailing CEO gender Family unsupportive work environment Supervisor gender Mentoring Job ranks Supervisor gender 9 Mentoring Supervisor gender 9 Job ranks LR chi2 (df) Log likelihood

16.67 (4.09)*** 5.70 (1.23)*** .03 (.15) .81 (.56) .66 (.48) .67 (.51) .14 (.13) 1.15 (.64) .34 (.25) .78 (.30)*** 1.11 (.51)*** .16 (.39) .41 (.17)** 1.50 (1.44) .04 (.20) 1.07 (.18)*** 1.49 (.81) .04 (.45) 15.01 (1893.29) 1.60 (4.08) .32 (.47) .79 (2766.10) 16.32 (1893.24) 15.81 (1893.24) .37 (.24) .39 (2.14) .92 (.78) .76 (1.28) 14.99 (2073.84) 15.31 (2056.71) .82 (.53) 12.46 (8142.96) .07 (.64) .71 (.50) .46 (5381.03) .63 (2459.29) 89.88 (36) 359.26

N = 621. +p < 0.10, *p < 0.05, **p < 0.01, ***p < 0.001. Supervisor and CEO gender (1 = female, 0 = male), mentoring, job functions and industry memberships were dummy-coded (1 = yes, 0 = no). No change (= 0) is base outcome for Demotion (= 1) and Promotion (= 1). 22

© 2019 Australian HR Institute