http://jms.sciedupress.com Journal of Management and Strategy Vol. 7, No. 1; 2016 Do Crises Change the Percieved Performance of Female Leaders versus Male Leaders from the Standpoint of Junior Level Employees? Emine Banu Çelebi1 & Ezgi Yildirim Saatci2 1 Social Entrepreneurship Research Center, Okan University, Formula 1 Yolu Tuzla Campus, Akfirat İstanbul, Turkey 2 Social Entrepreneurship Research Center, Department of Management, Okan University, Formula 1 Yolu Tuzla Campus, Akfirat İstanbul, Turkey Correspondence: Asst. Prof Ezgi Yildirim Saatci, Social Entrepreneurship Research Center, Department of Management, Okan University, Formula 1 Yolu Tuzla Campus, Akfirat İstanbul 34959 Turkey. Received: January 5, 2016 Accepted: January 25, 2016 Online Published: February 11, 2016 doi:10.5430/jms.v7n1p45 URL: http://dx.doi.org/10.5430/jms.v7n1p45 Abstract The difference between men and women was and will probably always remain to be an interesting research field for academics and practioners of all disciplines. From the management perspective, ass more and more women enter the workforce, the traditional workforce demographics are changing with an increasing tendency of women in leadership roles. Yet, the question of whether there are fundamental differences between the leadership behaviors of men and women remain crucial. This paper investigates the role of women managers versus male associates’ leading performances during the economic crises faced by the organizations through the lenses of junior level employees’ perception within an integrative approach of organizational theories of leadership, gender studies and crisis management. In accordance, the research question of “Do the junior level employees perceive women to be the “better leaders” in times of crisis and men to be the “better leaders” in times of tranquility? is addressed and two hypothesis are tested to find out this relationship on the axis of favorably evaluation and recommendation for leadership positions. Keywords: leadership, gender, employee perception, social identity, crises management 1. Introduction By the end of the 19th century, research on leadership theories revived and a variety of approaches such as the trait, behavioral and the contingency started to form. From these approaches, the transformational and the charismatic leadership were developed. These theories shared the idea that an appealing and commonly shared vision is a leader’s most important attribute (Strange and Mumford, 2002) and that this vision becomes even more important in a crisis situation (Berson et al., 2001). Furthermore, there is general agreement among theorists that charismatic leaders are more likely to emerge when a company’s performance is in a state of crisis (e.g., Conger & Kanungo, 1987; Yukl, 1998). Dimension of gender differences is also included within the domain of leadership in an effort to differentiate leader form others. Although some theorists found no significant differences in the leadership styles of men and women (e.g., van Engen et al., 2001,) and disregarded the gender stereotypes other theorists like Loden (1985) argued that men have a more masculine, agentic leadership style (with traits such as competitiveness, hierarchy and power), in contrast to women who prefer a more feminine, communal leadership style (with traits such as cooperation, equality and support). The field of gender and leadership has tried to assess the underrepresentation of women in management positions and the possible explanatory factors for this gender inequality such as (Ryan & Haslam, 2007). Some of these explanatory factors include; the obvious sexism in the work place (e.g., Schwartz, 1994), the perceived unsuitability between women’s capability as a leader and the necessities of true leadership (e.g., Arvey, 1979; Schein, 1973), women’s responsibilities at home and work – family conflicts (e.g., Schwartz, 1994) and women’s fear of success (e.g., Horner, 1972). Published by Sciedu Press 45 ISSN 1923-3965 E-ISSN 1923-3973 http://jms.scciedupress.com Journal of Management and SStrategy Vol. 7, No. 1; 2016 On the othher hand, sincce the end of the 20th centuury, the researrch on the geender differencces literature mmainly concentrateed around the qquestion of whhy women are nnot able to maakke it to the topp with “glass ceiling” phenommenon. (e.g., Ridgeway, 2001; AArfken et al., 2004; Maume, 2004) By 20000s “glass clifff phenomenon”” emerged expllaining that more aand more wommen are breakinng through the glass ceiling, bbut are only offfered positionss that are riskieer than the positions offered to men. While mmen turn these positions dowwn (in order nnot to fall shorrt in case of ffailure) women acccept it, and whhen failure doess occur, they hhave to face thee consequencees and get blammed much moree often than men (RRyan and Hasllam, 2004). Nevertheleess, recent reseearches suggesst that womenn are starting tto progress inn career paths (e.g., Dreher, 2003; Goodman eet al., 2003; RRyan & Haslamm, 2007). Yet, tthe participatioon rates are stiill below for mmost of the couuntries. For instancce, according tto Turkish Staatistical Institutte (2013) wommen representeed just over %30 of the 24 mmillion total labor forces in Turkeey as reported by World Bank Report and sstated in Figuree 1. TURKEY URBAN RURAAL NNON  NON AGRICULTURAL  AAGRICULTURAL  NON AGRICULTURAAL  NUMBER OF  UNEMPLOYMENT RATE  UNEMPLOYYMENT  NUMBER OF  UNEMPLOYMENT  UUNEMPLOYMENT  NUMBEER OF  UNEMPLOYMENNT  UNEMPLOYMENT RAATE  YEARS UNEMPLOYED (%) RATE (%) UNEMPLOYED RATE (%) RRATE (%) UNEMPLOYED RATE (%) (%) FEMALE 2004 622 11,0 19,6 5444 17,9 19,6 78 3,0 119,9 2005 642 11,2 18,7 5551 17,0 18,5 92 3,6 119,9 2006 658 11,1 17,9 5660 16,4 17,7 97 3,9 119,5 2007 660 11,0 17,3 5666 16,1 17,1 95 3,8 119,0 2008 734 11,6 18,1 6222 16,6 17,8 112 4,3 220,9 2009 979 14,3 21,9 8332 20,4 21,5 147 5,3 224,6 2010 959 13,0 20,2 8221 18,7 19,9 138 4,6 222,2 2011 885 11,3 17,7 7770 16,5 17,7 115 3,6 117,6 2012 883 10,8 16,4 7771 15,5 16,5 112 3,5 115,6 2013 1033 11,9 17,4 8885 16,4 17,3 148 4,5 118,1 2014* 1040 11,9 16,5 Source: Laabour Force Staatistics, 2004-22014 http://wwww.tuik.gov.tr/aalt_id=1007 Figure 1. Femmale laborforcce participationn in Turkey 20004-2014 When it coomes to glass ceiling, still women repressentation is noot significant wwith rates suchh as 8.5% of Board membershiip, 8.2% of ddirector in suppervision commmittees and oonly 7% of ddirector in corrporate managgement committeess are female (http://siterresources.worlddbank.org/TURRKEYEXTN/RResources/3617711_12688393345767/Femalee_LFP-en.pdf) as shown in FFigure 2. Source: Catalyst Knowledge Center, 2014 http://wwww.catalyst.org/kknowledge/2014-catalyst-census-women-booard-directors Figure 2. Percentagee of women paarticipation in ssenior level maanagement rolees Published byy Sciedu Press 46 ISSN 1923-39665 E-ISSN 19223-3973 http://jms.sciedupress.com Journal of Management and Strategy Vol. 7, No. 1; 2016 Due to the continued existence of differences, it is perceived and/or crucial to further investigate male / female contributions into organization studies, as a critical issue, and recognize it can add valuable and important contributions to the field, where traditional organization theories have still left much in vogue and especially during the time of crises. (Calas and Smircich, 2006) 2. Literature Review 2.1 Leadership Theories Leader based examples are prevalent throughout the history but only recently, from the beginning of the twentieth century, has leadership theories took its place in organizational behavior related researches on the grounds of power as “the capacity of individuals to overcome resistance on the part of others, to exert their will, and to produce results consistent with their interests and objectives” (Buchanan & Huczynski, 2004, p828). In order to define leadership related research studies, the first known approach was the trait approach that attempt to compare leaders from non-leaders with specific focus on superior qualities. Conveyed leadership characteristics included drive, motivation, honesty, integrity, self-confidence and cognitive ability (Kirkpatrick & Locke, 1991). The behavioral approach to leadership pioneered in the 1950s at the Ohio State University and the University of Michigan. This theory, in short, focused on the behavior kinds of people in leadership and identified that leadership behaviors could be grouped in two major types: consideration and initiating structure (Landy & Conte, 2004). Mutual trust, respect, and a certain warmth and rapport between the supervisor and group are included by consideration (Landy & Conte, 2004). In other words, how much the leader is concerned with his/her subordinates, and the manner in which he/she treats them. Starting structure included behavior in which the supervisor organizes and defines group activities and his relation to the group (Landy & Conte, 2004). In other words, how a leader succeeds in communicating to his/her subordinates that their task must be completed. The earlier researches in the area of leadership theories concentrated mainly on the traits or the behaviors of leaders, with paying little or no attention for the effect of situational or environmental variables. The contingency approach proposed to take into account the role of the situation in the field of leadership (Landy & Conte, 2004). Historically, Fiedler (1967) developed one of the first possible theories called the Least-preferred Coworker (LPC) Contingency Model or the Contingency Model of leadership effectiveness. The LPC Contingency Model “describes how the situation moderates the relationship between leadership effectiveness and a trait measure called the LPC” (Yukl, 1998). After 1980s, transformational and charismatic leadership theories gained significance as well. In 1978, Burns’ book “Leadership” created the base for transformational leadership theory. According to Burns (1978), transformational leadership is the back-and-forth between leaders and followers where each nurtures the other to higher levels of standards through motivation. The leader transforms the followers by appealing to their upright motives such as peace, trust, justice, and equality. A charismatic leader is defined to be the one who possess exceptional qualities which are accepted by their followers in an unconditional way with the guidance of leader’s mission, visions and tactics for achievement (Willner, 1984). Charisma “is most often associated with a personal attribute of a leader that almost hypnotizes followers and compels them to identify with and emulate the leader” (Landy & Conte, 2004, p.32). “These leaders represent revolutionary social forces, and they are responsible for significant societal transformations” (House & Baetz, 1979, p: 399). 2.2 Gender and Leadership From the gender point of view, leadership literature, until the 1980s, has assumed that leaders were men with special qualities. Furthermore, most of the research in this field has been conducted by men, whose subjects were men (Buchanan & Huczynski, 2004). It is argued that women are rendered ‘structurally powerless’ in being restricted to routine, low-profile jobs, as well as facing discrimination in promotion decisions. (Michelle et al 2010; Moran et al, 2011; Kelly et al. 2012; Lerbinger, 2012) There is considerable disagreement among researchers concerning the extent to which men and women differ on characteristics associated with leadership. On one side, many studies have not find significant differences between the leadership styles of men and women claim to not remain fundamental (e.g., Nieva & Gutek, 1981; Lefkoitz, 1994; Hyde & Plant, 1995; van Engen et al., 2001, Michelle et al 2010; Lerbinger, 2012). On the other side, the differences of gender related characteristics between the leadership styles are real and substantial (e.g., Gilligan, 1977; Ruble, 1983; Loden, 1985; Eagly, 1995 Moran et al, 2011; Kelly et al. 2012). Published by Sciedu Press 47 ISSN 1923-3965 E-ISSN 1923-3973 http://jms.sciedupress.com Journal of Management and Strategy Vol. 7, No. 1; 2016 Due to the natural and societal references associated to women –such as home keeper, child care taker and so on- not up to1990s, women were deeply scrutinized in the leadership research. Mann (1995) similarly argues that “women are underrepresented in leadership and management roles because they are less successful in acquiring power. Today, companies are affected by politics and external factors deeply; therefore many various sizes of crises happen during the life time of a company. Crises can cause many situations where some of them are fatal for the company but some of them show new opportunities. Seeing the benefit from the crises and acting in order to get minimum damage is the leader’s main challenge today. Despite of the leadership qualities, crises management ability is also required for today’s leaders. Crises management is different than the tranquility-time management, as in crises, leaders should emphasize intuitive way of managing with the help of analytical thinking. Multi-tasking, fast thinking and fast deciding abilities are more important in crises times where taking responsibilities of the decisions and acting according to the consequences of these actions are critical. In 1990, Eagly & Johnson conducted a meta-analysis on the laboratory studies done in the period from 1961-1987 and found that women were more likely to use a more democratic and interpersonally oriented style, compared with men who were more likely to use a relatively autocratic and task-oriented leadership style. However, no such differences were found when they examined field studies. This suggested that stereotypes might have been operating in lab studies but not in real organizations. In field studies, however, Eagly & Johnson (1990) determined that democratic and participative styles are preferred by women rather than the autocratic styles favored by men. Eagly’s (1995) gender role theory for instance, suggests that men and women are believed to have the traits that are necessary for the roles they generally occupy. These gender roles according to Eagly can be separated as the masculine attributes, which are named agentic, and the feminine attributes, which are called communal. Continuing, seeing as people strive to live up to their expectations, these stereotypes are likely to prove themselves to be right (Eagly et al., 2003). For this reason, women are challenged with difficulties to be able to become leaders, as they are not stereotypically perceived as leaders. For instance, women are more likely to use a transformational leadership style, which is seen as one of the most effective leadership styles in times of crisis ((Eagly et al., 2003). The newer methods for studying leadership, such as transformational, charismatic and visionary styles, brought forth another direction to the research on gender differences in leadership styles. The transformational leadership style is often seen as the feminine leadership style as it stresses on individual consideration (van Engen et al., 2001). This assumption is further confirmed by the meta-analysis confirmed by Eagly et al. (2003). From their study, they conclude that women make more use of transformational leadership style than men, and make more use of contingent rewords (which is part of the transformational leadership style). According to them, men on the other hand make more use of active and passive management by exception. Buchanan & Huczynski (2004), further state that in leadership roles men and women use different conversational styles which can lead to misunderstandings; women tend to listen and reflect more while men tend to talk more and give information. According to Landy & Conte (2004), it is clear that in the recent years the workplace is being transformed from one characterized by single contributors to one dominated by teams and interacting work groups. This in turn according to them suggests that people skills (e.g., communication, negotiation, conflict resolution) will become increasingly important and this would seem to favor women more as leaders. In his readings, Spector, (2006) also states that women would be more concerned with the feelings and emotional well-being of subordinates (consideration leadership style) while men would be more concerned with getting the job done (initiating structure leadership style) (Spector, 2006 p.6). Farrel (2005) argues that females are better leaders than men and when they actually go for the top they succeed better because of their female traits. 2.3 Social Identity Theory and Perception of Followers Ever since the early 1970s, research in the field of social psychology has made great advancement in the study of social cognition and intergroup relations. Social identity theory has become one of the most important research areas of this advancement. In 1972, Tajfel introduced the concept of social identity theory, to examine how people view themselves in intergroup concepts, how a system of social classifications creates and defines a person’s own place in society. Tajfel (1972) defined social identity as “the individual’s knowledge that he/she belongs to certain social groups together with some emotional and value significance to him/her of this group membership”. According to Tajfel (1972), social identity theory operates in two parts. Firstly, it defines and evaluates a person (for example, she is a female manager). These types of definitions and evaluations are made both by others and by the person themselves. Secondly, this theory prescribes appropriate behavior for them. They think and behave in Published by Sciedu Press 48 ISSN 1923-3965 E-ISSN 1923-3973 http://jms.sciedupress.com Journal of Management and Strategy Vol. 7, No. 1; 2016 characteristically “female manger” ways and this happens through social comparison (Buchanan & Huczynski, 2004). Furthermore, Tajfel (1972) states that in order to evaluate their own opinions and abilities, people not only compare themselves to others with whom they interact, but also compare their own group with similar, but distinct, out-groups. The elements used to make such comparisons are called social categorizations. Categorizations, according to Buchanan & Huczynski (2004), lead to assumptions of similarity among those who are categorized together. It minimizes the superficial differences between members of the in-group and maximizes the differences between the in-group and out-groups (in terms of us and them). According to Tajfel and Turner (1986), the out-groups will tend to be stereotyped. Once this happens, those who are part of the in-group will have assumed social identity, and from this standpoint, they will view others. Continuing, in 2001, Hogg examined the social identity theory in relation to leadership. He describes leadership as“… how some individuals or cliques have disproportionate power and influence to set agenda, defines identity, and mobilizes people to achieve collective goals. The differential ability of some people to stamp their mark on attitudes, practices, decisions, and actions is endemic to all social groups—for example, nations, communities, organizations, committees, cliques, and families. Leaders are people who have disproportionate influence, through possession of consensual prestige or the exercise of power, or both, over the attitudes, behaviors, and destiny of group members.”(Hogg, 2001, 187) Furthermore, Hogg (2001) proposed that three core processes exist which operate in combination to increase the influence of social identity in leadership processes. These three core processes are prototypicality, social attraction, and attribution and information processing. Attribution is specifically important and can be seen though perception of followers/employees/subordinates. Until now, most of the research on the topic has been performed on working businesspeople. In addition to this, the objective of this research is to study junior level employees and to test what their perceptions as well as the preferences are regarding this glass ceiling. Their results would be extremely valuable as they will be the future leaders of top companies, who could possibly reduce, if not totally eliminate, the gender inequality and the predominant gender stereotyping. H : Compared to men, women are more favorably evaluated and more likely to be preferred for leadership 1 positions in times of crisis as perceived by junior level employees. H : Compared to women, men are more favorably evaluated and more likely to be preferred for leadership 2 positions in times of no crisis as perceived by junior level employees. 3. Methodology In order to investigate the hypotheses of the current study, a quantitative research is carried out. A method to explore the research question and hypotheses would be to conduct an online questionnaire to test the independent variable junior level employees’ perceived suitability of male or female manager as a manager/leader and the independent variable company performance (declining: crisis situation or improving: no crisis situation) on the dependent variables. The dependent variables are the respondents’ evaluation of the managers in the dimensions of (a) the managers’ leadership ability and (b) the managers’ suitability for the position according to the junior level employees’ perception. 3.1 Sample One-to-one questionnaire distribution and interviews were conducted based on convenience sampling over a period of 2 years with a total of 7 companies. After 20 sessions, at the end in total, 141 people have voluntarily participated in this experiment. All the respondents were junior employees who have less than 3 years of experience. Of the respondents 77 (55%) were female and 64 (45%) were male. The average age of the respondents was between the age group of 23 to 27 (52%). 3.2 Procedure, Measures and Data Analysis The questionnaire, which has been created for the current study, is based on Ryan & Haslam's (2005a) measure of the class cliff phenomenon. However, instead of a laboratory experiment (which they have used) an online experiment by means of a questionnaire is created, where respondents answered to one of the four versions of the seven pages questionnaire. The first page consisted of a job advertisement for a desirable senior management position, a financial director for a large, international firm. Published by Sciedu Press 49 ISSN 1923-3965 E-ISSN 1923-3973 http://jms.sciedupress.com Journal of Management and Strategy Vol. 7, No. 1; 2016 Second pages differed for inclining group (improving performance scenario with company’s stock values in growth trend graphs) and for declining (declining performance scenario with company’s stock values in recession trend graphs) Thereafter, participants were presented with a description and photograph of two managers for the advertised position. Both these managers were highly experienced, currently holding Financial Director positions and both having earned MBAs from prestigious business schools. The photographs were chosen after pilot testing so that Managers 1 and 2 were as similar as possible on all dimensions other than gender: they were both white and of equivalent age and attractiveness. The statements are designed to measure perceptions for: (a) the managers' leadership ability (‘The manager would be a good leader’, ‘The manager has the skills and experience to lead other people’, ‘The manager has clear leadership credentials’) and (b) the managers' suitability for the position (‘The manager's past experience is relevant to the position’, ‘The manager will bring the required skills to the job’, ‘The manager is suitable for this position’). Participants were then asked to evaluate each of the managers by indicating their level of agreement with six statements (from 1, do not agree at all, to 7, agree completely). The statements were designed to measure perceptions of (a) the managers' leadership ability (‘The manager would be a good leader’, ‘The manager has the skills and experience to lead other people’, ‘The manager has clear leadership credentials’) and (b) the managers' suitability for the position (‘The manager's past experience is relevant to the position’, ‘The manager will bring the required skills to the job’, ‘The manager is suitable for this position’). (Ryan & Haslam, 2008, p 530) 3.3 Results The data gathered in the experimental questionnaire is analyzed with the statistical program SPSS 21.0 for Windows. For testing each dependent variable separately a ANOVA was run, then the interaction of the independent variables and the covariates were checked against the six dependent, to see whether it provided any significant results. If indeed the relationship of the interactions was significant, a further analysis was conducted to measure the degree of the interaction by using a LSD post-hoc test. The results of ANOVA test were not significant for variables “the manager would be a good leader” (p-value = 0.751), “The candidate has the skills and experience to lead other people” (p-value = 0.471),”The candidate has clear leadership credentials” (p-value = 0.303) and “The candidates past experience is relevant to the position” (p-value = 0.159). However 2 variables were found significant in ANOVA as: The candidate will bring the required skills to the job (p-value = 0.087) with a post hoc test as shown in Table 1. Table 1. LSD Post-Hoc Test for bring the required skills to the job Multiple Comparisons bring the required skills to the job LSD (I) merge of (J) merge of Mean Std. Sig. 95% Confidence Difference Error Interval (I-J) Lower Upper Bound Bound female male and crisis .19 .183 .309 -.17 .55 and crisis female and no crisis .08 .176 .641 -.27 .43 male and no crisis -.18 .187 .334 -.55 .19 male and female and crisis -.19 .183 .309 -.55 .17 crisis female and no crisis -.10 .177 .556 -.45 .25 male and no crisis -.37 .188 .053 -.74 .00 female female and crisis -.08 .176 .641 -.43 .27 and no crisis male and crisis .10 .177 .556 -.25 .45 male and no crisis -.26 .182 .149 -.62 .10 male and female and crisis .18 .187 .334 -.19 .55 no crisis male and crisis .37 .188 .053 .00 .74 female and no crisis .26 .182 .149 -.10 .62 Published by Sciedu Press 50 ISSN 1923-3965 E-ISSN 1923-3973 http://jms.sciedupress.com Journal of Management and Strategy Vol. 7, No. 1; 2016 The results of Table 1 shows that the mean of the male candidate and crisis situation (mean = 2.47) is significantly different (p-value = 0.053) from the mean of male and no crisis (mean = 2.81). These results indicate that when respondents are evaluating if the candidate (male or female) will bring the required skills to the job, the presence of a crisis has an effect only on the male candidate. Furthermore, these results also show that in a crisis situation male candidate’s means are ranked lower than the overall mean by respondents when evaluating if the candidate will bring the required skills to the job. These findings suggest that male candidates are more likely to bring the required skills to the job when the company is not in a crisis situation, whereas they are less likely to bring the required skills to the job when the company is in a crisis situation. The candidate is suitable for this position (p-value = 0.077) with a post hoc test as shown in Table 2. Table 2. LSD Post-Hoc Test for “Suitable for this position” suitable for this position LSD (I) merge of (J) merge of Mean Difference Std. Sig. 95% Confidence Interval (I-J) Error Lower Upper Bound Bound female and male and crisis .10 .167 .545 -.23 .43 crisis female and no .02 .161 .912 -.30 .34 crisis male and no -.36* .171 .037 -.70 -.02 crisis male and crisis female and -.10 .167 .545 -.43 .23 crisis female and no -.08 .162 .606 -.40 .24 crisis male and no -.46* .173 .008 -.80 -.12 crisis female and no female and -.02 .161 .912 -.34 .30 crisis crisis male and crisis .08 .162 .606 -.24 .40 male and no -.38* .166 .025 -.71 -.05 crisis male and no female and .36* .171 .037 .02 .70 crisis crisis male and crisis .46* .173 .008 .12 .80 female and no .38* .166 .025 .05 .71 crisis Based on observed means. The error term is Mean Square (Error) = .483. *. The mean difference is significant at the 0.05 level. The results of Table 3 shows that the mean of the male candidate and no crisis situation (mean = 2.87) is significantly different from the mean of male and crisis (mean = 2.44, p-value = 0.008), the mean of female candidate and crisis (mean = 2.54, p-value = 0.037) and the mean of female candidate and no crisis (mean = 2.52, p-value = 0.025). These results indicate that when respondents are evaluating if the candidate (male or female) is suitable for this position, the male candidate under a no crisis situation is evaluated as significantly higher than all the other means. The female candidate is rated indifferent for company performance in both crisis and no crisis Published by Sciedu Press 51 ISSN 1923-3965 E-ISSN 1923-3973 http://jms.scciedupress.com Journal of Management and SStrategy Vol. 7, No. 1; 2016 situation (tthis is especiallly visible fromm figure 3, which shows an aallmost horizonttal line when eevaluating the female candidate).. These findingg suggest that male candidatees are more likkely to be suittable for this joob position whhen the company iss not in a crisiss situation, wheereas they are lless likely to bbe suitable for tthis job positioon when the company is in a crisis. Figuree 3. The candiddate is suitable for this positioon Furthermorre, since the coonceptual moddel of the curreent study, baseed on the reseaarch conductedd by Ryan & HHaslam (2005a), suuggested two factors: (1) thhe potential caandidates’ leaddership abilityy and (2) the ppotential canddidates’ suitability for the positionn., The Bartlettt’s test of Sphhericity (p < .000) and KMOO test (KMO = .817) were used and verified thaat the factor annalysis is feassible to be testted and in ordeer to rotate thee solution indiicated that of tthe six variables, tthe first three vvariables correllated with eachh other and the last three variiables correlateed with each otther, as a result forming two factoors (this is alsoo in accordancee with the theorry of Ryan & HHaslam, 2005aa). Factor 1 wwas named as LLeadership Abbility, (which mmeasured the total values oof the first threee variables annd that divided byy three) belongging to the caandidates' leaddership ability and factor 2 was named as Suitability, ((which measured tthe total valuess of the last thrree variables annd that dividedd by three) beloonging to the ccandidates' suittability for the posiition. The sufficiiently high valuues of Cronbacch α in each factor indicate aa high reliabilitty of the two exxamined factorrs. The total variannce explained bby the two factors (68.996 %)) is adequatelyy satisfying. The resultss of the ANOVVA indicate thaat the interactioon of companyy performance and the gendeer of the candiddate on the dependdent variable ‘thhe leadership aability of the caandidate’ were insignificant ((p-value = 0.6332). However, results of the ANOVA inddicate that the interaction off company peerformance annd the gender of the candidate oon factor ‘suitaability” is significant (p-valuee = 0.050) Published byy Sciedu Press 52 ISSN 1923-39665 E-ISSN 19223-3973 http://jms.sciedupress.com Journal of Management and Strategy Vol. 7, No. 1; 2016 Table 3. LSD Post-Hoc Test for “Suitability” Suitability LSD (I) merge of (J) merge of Mean Std. Sig. 95% Confidence Difference (I-J) Error Interval Lower Upper Bound Bound female and male and crisis .1277 .15376 .408 -.1763 .4318 crisis female and no .0405 .14780 .785 -.2518 .3327 crisis male and no -.2491 .15618 .113 -.5579 .0597 crisis male and crisis female and -.1277 .15376 .408 -.4318 .1763 crisis female and no -.0873 .14895 .559 -.3818 .2073 crisis male and no -.3768* .15728 .018 -.6878 -.0658 crisis female and no female and -.0405 .14780 .785 -.3327 .2518 crisis crisis male and crisis .0873 .14895 .559 -.2073 .3818 male and no -.2896 .15145 .058 -.5891 .0099 crisis male and no female and .2491 .15618 .113 -.0597 .5579 crisis crisis male and crisis .3768* .15728 .018 .0658 .6878 female and no .2896 .15145 .058 -.0099 .5891 crisis Based on observed means. The error term is Mean Square (Error) = .408. *. The mean difference is significant at the 0.05 level. The results of Table 3 shows that the mean of the male candidate and no crisis situation (mean = 2.9063) is significantly different (p-value = 0.018) from the mean of male candidate and crisis (mean = 2.5294) and the mean of female candidate and no crisis (mean = 2.6167, p-value = 0.058). These results indicate that when respondents are evaluating if the candidate (male or female) is suitable for this position, the male candidate under a no crisis situation is evaluated (1) higher than a male candidate under crisis situation and (2) higher than a female under no crisis situation. The female candidate is not affected by the presence of a crisis (this is especially visible from Figure 4). Finally, there are gender effects in a crisis situation. However, in a non crisis situation, there are indeed significant differences between a male and female candidate. Overall, male candidates are significantly rated higher than female candidates. Published by Sciedu Press 53 ISSN 1923-3965 E-ISSN 1923-3973 http://jms.scciedupress.com Journal of Management and SStrategy Vol. 7, No. 1; 2016 Figurre 4. Suitabilityy of candidate to the positionn These finding suggest thaat male candiddates are more likely to be suuitable for thiss job position wwhen the comppany is not in a crrisis situation, whereas they are less likely to be suitablee for this job pposition when the company is in a crisis situattion. 4. Discussiions and Concclusions According to the results mentioned aboove, when exaamining the ressults of the deependent variabble ‘the suitabiility of the candidaate’ in greater detail, the folllowing can be concluded reggarding the hyppotheses testingg. The results for the dependent variable (factoor 2), on the basis of ANOVVA and LSD ppost-hoc testingg shows that wwhen respondennts are evaluating if the candidaate (male or ffemale) will ‘wwill be suitablle for this possition’, only thhe male candiddate is evaluated wwith significannt results. The female candiddate is rated inddifferent for coompany perforrmance in bothh crisis and no crissis situation. These findiings imply thaat male candidaates are more llikely to ‘be suuitable for the position’ whenn the companyy is not in a crisis situation, wheereas they are less likely to ‘be suitable foor the positionn’ when the coompany is in aa crisis situation. FFor the above mmentioned reassons, there is eevidence to suppport hypothessis two which states that commpared to women, men will be mmore favorablyy evaluated andd more likely to be recommmended for leaddership positionns in a company wwhen its performmance is improoving rather thhan declining. AAnd as a resultt, reject hypothhesis one. Although tthe other depeendent variablee ‘candidates leadership abiility’ showed nno significant results, and bby this neither hyppothesis one oor two could bbe supported, tthe dependentt variable did show a patterrn for the interraction between thhe company performance andd the dependennt variable whicch is somewhaat in line with hypothesis twoo. This dependent variable showeed that responddents were morre inclined to rrate a candidatte (which is gennderless as theere was no significcance for the ggender of the candidate) higgher when coompany performmance was immproving ratheer than declining. AAlthough hypoothesis two dooes take genderr of the candiddate into it, thee remaining paart of hypothessis two states that a candidate is more likely too be recommennded when the performance oof the companyy is improvingg rather than declinning. Even thoough there is nno support for hypothesis twwo as it does nnot take gendeer into accounnt, it is important tto note this paattern of the reespondents whho rate candidaates higher in a no crisis situuation and ratee them lower in a ccrisis situation. In line wiith hypothesis two, evaluatiions of the twwo candidatess by respondeents revealed that when commpany performancce was not in aa crisis, the maale candidate wwas seen (a) ass more suitablee for the positiion and (b) as hhaving more leadeership ability, tthan an identiccally qualified female candiddate. This findiing also partiallly supports thee glass cliff phenoomenon, that beeing participannts’ ideal leaderr in a healthy ccompany is asssociated with ““think managerr-think Published byy Sciedu Press 54 ISSN 1923-39665 E-ISSN 19223-3973