The relationship of promotional opportunities and promotion of highly educated females with turnover intention and industry exit intention in the hospitality industry.

The relationship of promotional opportunities and promotion of highly educated

females with turnover intention and industry exit intention in the hospitality

industry

Master thesis

Judit Palinkas

Student number: 6347045

Supervisor: Dr. Stefan Mol

Second reader:

Master Business Studies University of Amsterdam

Table of content

Table of content ... 1

Abstract ... 2

Introduction ... 3

Literature review and hypotheses ... 4

Turnover and industry exit ... 4

Antecedents of turnover ... 6

Industry exits of females ... 10

Conceptual Model ... 16

Method ... 17

Data collection ... 17

The sample ... 17

Measures ... 19

Perceived promotional opportunities ... 19

Turnover intention ... 20

Industry leaving intention ... 21

Actual promotions ... 21

Demographic characteristics... 22

Employment situation ... 22

Data analysis and results... 23

Mean scores and bivariate correlations ... 23

Hypothesis testing... 25 Discussion... 34 Conclusions ... 38 Limitations ... 40 Managerial implications ... 41 Works Cited ... 42 Appendix ... 47 1

Abstract

Specific human resource concerns in the hospitality industry are the low presence of females in high positions, a high managerial turnover rate and a high industry leaving rate. This research is aimed at investigating the reasons of highly educated female graduates leaving the hotel industry in high numbers. Given that highly educated females are leaving the hotel industry in higher numbers than men, this paper argues that to a large extent this difference can be explained by fewer promotional opportunities and fewer actual promotions for women as compared to men. Data was collected from hospitality graduates; 253 respondents filled out the online questionnaire. It was found that gender directly affected industry leaving intentions, where males had higher intentions to leave the industry. While as an indirect effect females tended to wait longer for a promotion, this affected negatively their perception of promotional opportunities, which in turn translated into a higher intention to leave the organization or the industry; as another indirect effect actual promotion only – independent of promotional opportunities – mediated the relationship of gender and industry leaving intention, where being female was related to a higher intention to leave the industry. These indirect effects were only present among respondents with one or no promotion. It was suggested that the result of this study should be considered when developing human resource management policies in order to minimize employee turnover. Hospitality organizations should invest in career development of graduated female employees in order to decrease managerial turnover. Further research is necessary to discover the antecedents of turnover and industry leaving intention of males.

Introduction

The hospitality industry is a significant employer, according to a recent report (Oxford Economics, 2010) prepared by Oxford Economics; the hospitality industry is the fifth largest employer in the UK, directly employing over 2.4 million people. The hospitality industry is facing labor and skills shortages internationally (ISHC, 2006). This is due to the fact that jobs in the hospitality industry are characterized by high labor turnover and a high amount of industry exits; employees leaving for opportunities outside of the hospitality industry. Almost half of hospitality graduates left the industry within 10 years according to the study of Johns and McKechnie (1995); this number is even higher for female graduates (Blomme, Van Rheede, & Tromp, 2010b). Highly educated women form a substantial part of the workforce (Riley, 1991; Charlesworth, 1994; Deery & Iverson, 1996), thus their retention is an important challenge for the hospitality industry (Hoque, 1999; Walsh & Taylor, 2007). Around two thirds of hospitality graduates are female (HCITB, 1984; Ng & Pine, 2003), but only a few of them eventually make it to become general manager in a major hotel (Guerrier, 1986). The extant literature suggests that female graduates are facing bigger obstacles in their career advancement than males, in that women appear to have fewer promotional opportunities than men (Stroh, Brett, & Reilly, 1996; Lyness & Heilman, 2000). Promotional opportunities are negatively related to both turnover and turnover intentions (e.g. Blomme, van Rheede, & Tromp, 2010a; Walsh & Taylor, 2007; Lyness & Judiesch, 2001). This study argues that the lack of opportunities for women and their lack of actual promotions play a significant role in the turnover intentions and industry leaving intentions of female hospitality graduates.

Below, first the phenomenon of high turnover rates in the hotel industry is introduced, after which the possible causes of such turnover will be discussed.

Literature review and hypotheses

Turnover and industry exit

The phenomenon of high turnover is widely acknowledged in the hospitality literature. The average yearly labor turnover in all industries in the UK is approximately 16 % (CIPD, 2000) while in the hotel industry the estimates are 43 % (CERT, 1999). Turnover is defined by Macy (1983, p. 142) as “a permanent movement beyond the boundary of an organization”. Turnover can be voluntary or involuntary. An instance of voluntary turnover, or a quit, reflects an employee's decision to leave an organization, whereas an instance of involuntary turnover, or a discharge, reflects an employer's decision to terminate the employment relationship (Shaw, Delery, Jenkins & Gupta, 1998). Turnover can be seen positively and negatively by the employer (Deery & Shawn, 1997). On the positive side turnover can enhance labor quality by removing underperforming employees (Davidson & Wang, 2011). On the negative side turnover is a sign of employee dissatisfaction and a breakdown in the psychological contract between the employee and the employer (Blomme, van Rheede, & Tromp, 2010a). Regardless of the point of view turnover is associated with high costs for the organization. Direct costs are associated with replacing the employee such as the costs pertaining to the hiring and training process, but direct costs represent only a small part of the total cost of turnover. Indirect cost can occur in terms of lost productivity or a decrease in service quality (Hinkin & Tracey, 2006) and a decreased service quality can affect the reputation of service-oriented business (Hom &

Griffeth, 1995). Hinkin and Tracey (2006) found that the cost of turnover shows mostly in lost productivity. On average in the hospitality industry lost productivity accounts for 50% of the total cost of turnover, and in some positions this figure can reach 75%. The high turnover rate in the hospitality industry is often accepted as an industry characteristic (Phillips & Connell, 2003). In the hospitality industry many employees are seasonal or temporary; their departure is planned and unavoidable, and these departures are not caused by the breakdown of employee-employer relationship (Stalcup & Pearson, 2001). In the case of seasonal workers indirect cost are negligible.

Yet, voluntary turnover among managers in the hospitality industry is also high (Hiemstra, 1990). Their turnover cannot be explained by seasonal work, as managers are generally permanent employees. Thus the common explanations of high turnover in hospitality do not apply to managers. Turnover research in 64 four to five star Australian hotels showed annual turnover rates of 50,7% for operational employees and 39,2% for managerial employees; (Davidson, Timo, & Wnag, 2009). Managerial positions are mostly filled by highly educated employees, who successfully completed a higher education program at a bachelor`s or master`s level (Hoque, 1999; Blomme, van Rheede, & Tromp, 2010a). The cost of replacement of this highly educated managerial staff can be multiple times the cost of replacing a (seasonal) line employee. Hogan (1992) estimates the cost of an individual manager replacement in the hospitality industry at between 17000 USD and 20000 USD. Arthur Nathan the vice president of human resources at the Mirage Hotel and Casino in Las Vegas estimated this number to be 6600 USD (Stalcup & Pearson, 2001; interview, June 8, 1996). Hinkin and Tracey (2006) put the average cost of a replacement in the most complex hospitality jobs in their sample to an

average 10 000 USD. The high turnover rate among highly educated employees not only leads to higher staffing costs but also to a decrease in organizational competencies (Walton, 1985; Lado & Wilson, 1994) and an erosion of the company’s tacit knowledge (Coff, 1997).

The hotel industry is not only struggling with a high turnover rate in general, but highly educated people, occupying mostly managerial positions, are leaving the industry in large numbers. Blomme, Rheede and Tromp (2010a) collected data from the Alumni of the Hotelschool, The Hague including graduates under the age of 65. Blomme et al. (2010a) found that only 42.7% of alums still worked in the industry.

Having discussed the cost of turnover and the high turnover and industry exit present in the hospitality industry, it is important to understand the antecedents of turnover and industry exits especially for highly educated employees, as their turnover has even higher costs than non-graduate employees.

Antecedents of turnover

Phillips and Connell (2003) argue that individual businesses should work on maintaining a low level of turnover by gaining an understanding the causes of turnover and not to accept it as an industry problem. Companies can gain from understanding and maintaining a low level of turnover; these businesses can yield a competitive advantage over firms that accept turnover as an industry problem. In the following the antecedents of turnover in general and in the hospitality industry in specific will be discussed.

Griffeth, Hom and Gaertner (2000) updated an earlier meta-analysis of Hom and Griffeth (1995). This cross-industrial meta-analysis of turnover antecedents and correlates distinguishes both proximal precursors and distal determinants. The most important proximal precursors of

turnover were shown to be: job satisfaction, organizational commitment, job search, comparison of alternatives, withdrawal cognitions and turnover intentions. Griffeth et al. (2000) found moderate effect sizes for the following distal determinants: job content, stress, work group cohesion, autonomy, leadership and small effect sizes for distributive justice and promotional opportunities. However, findings from studies in other industries on turnover may not generalize to the hospitality industry (Woods & Macaulay, 1989).

Researchers investigating turnover antecedents in the hospitality industry mostly point to the role of stress, work-family conflict (as a specific stressor), and the impact of long, inflexible hours, low wages, intrinsic job features and promotional opportunities. According to a recent study by O `Neill and Davis (2010) employees reporting more interpersonal tension are less satisfied with their jobs and more likely to leave their jobs. In the same study hotel employees reported stressors on 40-62% of days. As a comparison in the study of Almeida and Horn (2004) a US national and diverse sample of subjects reported stressors on 25-44% of days. Thus hotel workers appear relatively more stressed as compared with other industries. According to the findings of O`Neill and Davis (2010) managers are significantly more stressed than line employees, and role stress has a stronger effect on job satisfaction for managers then for non-supervisory employees (Kim, Murrmann, & Lee, 2009). Job satisfaction is an important predictor of turnover (Griffeth, Hom, & Gaertner, 2000). Thus the experienced high level of stress provides a partial explanation for the high turnover rate in the hotel industry. Blomme at al. (2010b) found that work-family conflict and organizational support explain a large amount, namely 29% of variance in highly educated employees’ turnover intentions. Pavesic and Brymer (1990) and Sarabakhsh, Carson and Lindgren (1989) found uncompensated long, nontraditional

hours as the main reason for young graduated managers turnover. Regular schedule change causes dissatisfaction among hospitality employees (Poulston, 2009). Furthermore employees appear to have to work long hours without sufficient compensation. Davidson and Wang (2011) and Walsh and Taylor (2007) mentioned low wages in the hotel industry as a possible cause for high turnover and industry exit. The hotel industry has a high need for labor and offers lower wages than other industries (Australian Bureau of Statistics, 2008). In Australia, jobs in accommodation, cafes and restaurants had the lowest hourly earnings of 21.5 AUD for full time non-managerial employees, compared to 45,3 AUD in the mining sector and an overall average of 30,1 AUD for all Australian workers (Australian Bureau of Statistics, 2008). While hospitality has significantly lower pay than other industries most scholars agree that low wages are not the main reason for high turnover among highly educated employees (Walsh & Taylor, 2007). According to the findings of Walsh and Taylor intrinsic aspects of jobs, such as learning oriented relationships and challenging work are more important for graduates than extrinsic rewards. The most important aspects of challenging work were interesting work, being able to develop new skills and participation in decision making. Walsh and Taylor compared the degree to which respondents viewed certain job features present in their job as well as the degree to which each job feature was important to them. The largest difference was found between the importance of a challenging job and the degree to which the respondents’ current job was challenging. The second largest difference was found between the importance and presence of learning-oriented relationships at work. The difference regarding rewards was much smaller. They found that the more employees perceive their job as challenging the more they are committed to their organization. In order to have a challenging job, hospitality university

graduates are mostly looking for jobs that offer growth opportunities (Walsh & Taylor, 2007). If the above mentioned job features are present, together with fair compensation and competent leadership, managers` commitment levels rise, which in turn reduces the likelihood of them leaving the organization or the industry. Walsh and Taylor (2007) point out that their sample of Cornell University School of Hotel Administration graduates were unwilling to wait for a top level position; without having a clear ladder they were leaving their organization, and many the industry within a few years after their graduation. Walsh and Taylor (2007) conclude that it is the absence of opportunity that causes young managers to leave the hospitality industry. Stalcup and Pearson (2001) too identified career advancement issues as one of the main reasons for managerial turnover in hotels. In addition, promotional opportunities at other organizations are a pull factor and the cause of hospitality employees to leave their jobs (Brien, 2004; Davidson & Wang, 2011).

Hypothesis 1

There is a negative relationship between perceived promotional opportunities and turnover intentions of hospitality graduates.

Hypothesis 2

There is a negative relationship between perceived promotional opportunities and industry leaving intentions of hospitality graduates.

Industry exits of females

Having discussed the antecedents of turnover in the hospitality industry, this study is focused on identifying the antecedents responsible for the apparent higher number of females leaving the industry (HCITB, 1984; Ng & Pine, 2003; Blomme, van Rheede, & Tromp, 2010a; 2010b). Females’ higher industry exit is indicated by the following studies showing that there are more female hospitality graduates but more male graduates are working in the industry. According to HCITB (1984) almost 70% of the hotel and catering management students in the UK were female. The percentages of female students in the Department of Hotel and Tourism Management at the Hong Kong Polytechnic University between 1992 and 2002 were above 60% (Ng & Pine, 2003). According to the study of Blomme at al. (2010a) only 36.0% of the surveyed alumni who still worked in the industry were female. In the age category 22-33 about two thirds of the graduates who started to work in the industry still worked there, but in the age category 34-44 significantly more women had left the industry than men, 61,0% compared to 47,0% respectively (Blomme, Van Rheede, & Tromp, 2010b). Having shown females’ higher industry exit, this study is trying to shed light on the cause of this phenomenon.

Some studies suggest higher stress level among females (Kim, Murrmann, & Lee, 2009) as a cause of their higher turnover rate, but in a recent study, O`Neill and Davis (2010) found no differences in experienced stress level based on gender among hotel employees. Schwartz (1989) argues that women`s higher commitment to family responsibilities is the main reason for higher turnover. According to Walsh and Taylor (2007) poor work-family balance is often a reason especially for women to leave the hospitality industry. Stroh, Brett and Reilly (1996)

question the direct relationship between work-family balance and turnover. Blomme at al. (2010b) found that for men work-family conflict is explained by the lack of organizational support, while for women work-family conflict is explained by the lack of organizational support and by their dissatisfaction with workplace flexibility. Hoque (1999) also argues that women find it more difficult to maintain a healthy balance between work and family because of long hours and shift work at irregular hours.

Other studies argue for the role of promotional opportunities as the cause of a higher turnover rate among females. Blomme at al. (2010a) examined the moderating effect of gender on the relationship between affective commitment, job content, development opportunities, job security, intra-organizational mobility, work-family balance, work atmosphere, autonomy, salary, performance related pay, clear task description and promotional opportunities on the one hand and intentions to leave on the other. They found that gender does not moderate any of these relationships with one exception; the negative relationship between promotional opportunities and turnover intentions is moderated by gender so that this relationship is stronger for women than for men. Female managers were dissatisfied with their advancement opportunities and reported higher turnover intentions than men with comparable opportunities. This indicates that female managers have a lower ‘‘tolerance’’ for a lack of opportunity (Stroh, Brett, & Reilly, 1996). Women may perceive lack promotions as unfair or discriminatory and this lack of perceived fairness and organizational justice may lead to turnover (Poulston, 2009; Nadiri & Tanova, 2010; Radzi, Ramley, Salehuddin, Othman, & Jalis, 2009). This stronger relationship between promotions and turnover has been supported in the

other directions as well; according to the findings of Lyness and Judiesch (2001) promoted women are less likely to resign than promoted men.

Female managers reported less satisfaction with their promotional and career opportunities at their current organizations than males according to multiple studies (Lyness & Thompson, 1997; Miller & Wheler, 1992). Indeed despite the high percentage of female hotel management students, only a few of them eventually become general manager (Guerrier, 1986). The reasons for women not making it to the top of the hotel industry has been the subject of the below studies. Lyness and Heilman (2000) argue that women have to wrestle with a lack of person-job fit, where more negatively biased evaluations of female managers will occur when there is a greater perceived lack of fit between job requirements and attributes of women. Morrison, White and van Velsor (1987) introduced the term glass ceiling to describe the difficulties that women face in their career. Guerrier (1986) argues that women’s lack of success in becoming a general manager can be explained by the figurehead role of the position, where the general manager needs to represent the hotel, socialize with guests and peers, and in these informal settings men have greater social acceptability. Furthermore Guerrier argues for the importance of the characteristics of the typical career route leading to become a general manager. These characteristics involve progress through informal contacts; which puts females at a disadvantage, as several studies recognized `Old-boys` networks as a major obstacle for female managers (Guerrier, 1986; Diaz and Umbreit, 1995; Ng et al., 2002). Food and beverage experience is an important step in becoming a general manager, while this department is dominated by males. Gender stereotyping is also a barrier for females (Ng, 1995). Hicks (1990) interviewed hotel managers, and found that women were seen as temporary employees and

they had to prove their performance before being given a place on the board whereas males were able to obtain a place immediately after being hired. The assumption regarding men is that they will have support from their wives (Guerrier, 1986). Ng and Pine (2003) surveyed managers in Hong Kong hotels and found that women ranked lack of support systems at work and lack of equity in promotions as the top two obstacles for career advancement, while men ranked these second and fourth respectively. Both female and male managers ranked childcare responsibilities and being married as unimportant obstacles. Interestingly men ranked conflicts with family activities as number 9, while females ranked it as even less important, number 12. Gender stereotypes were the number one obstacle for women according to the answers on an open-ended question (Ng & Pine, 2003). Lyness and Heilman (2000) examined promotions in a 2 year period after a performance evaluation, in their cross-sectional study they found that performance ratings had more direct career consequences for female managers than for males and promoted women had higher performance ratings than promoted men suggesting that females have to accomplish more to be promoted.

Hypothesis 3

Gender influences actual promotions, with females having to wait longer to be promoted than males.

If females are indeed receiving fewer promotions and females have to accomplish more to be promoted (Lyness & Heilman, 2000), females will perceive their promotional opportunities negatively and will report less satisfaction with their promotional opportunities than males.

Hypothesis 4

There is a relationship between actual promotions and perceived promotional opportunities of hospitality graduates, where waiting longer for a promotion is associated with more negative perception of promotional opportunities.

Hypothesis 5

Gender influences perceived promotional opportunities with females having more negative perception of promotional opportunities than males. This relationship is mediated by actual promotions.

According to multiple studies turnover intentions are greater for females, but this difference is not significant after controlling for human capital variables (Miller & Wheler, 1992; Rosin & Korabik, 1995). Lyness and Thompson (1997) examined the turnover intentions of senior-level male and female managers. After matching the respondents on hierarchical level, age, line or staff job and performance ratings, they found no differences in turnover intentions. Kanter (1982) noted that some of the observed gender differences in work attitudes and behavior such as turnover are due to the fact that women’s jobs tend to be lower in the organizational hierarchy and that they have fewer career opportunities. Thus women do on the whole have higher turnover intentions, but after controlling for organizational level this difference disappears. Studies controlling for hierarchical level might just have removed the reason of the examined turnover intention, the lack of promotions. This study argues that because females receive fewer promotions than males, females will perceive their promotional

opportunities negatively and will report less satisfaction with their promotional opportunities than males; and perceived promotional opportunities play a significant role in their higher turnover intention.

Hypothesis 6

Gender influences turnover intentions, with females having higher turnover intentions than males. This relationship is mediated by actual promotions and perceived promotional opportunities.

As discussed above, female hospitality graduates tend to leave the hospitality industry in higher numbers than males, this is shown from the higher number of female graduates (HCITB, 1984; Ng & Pine, 2003). but less actually working in the industry (Blomme at al., 2010a; Blomme, Van Rheede, & Tromp, 2010b). It can be expected, that because female graduates are leaving the hospitality industry in higher numbers than males, females also express higher intention to leave the industry, than males. This study argues that females are leaving the industry due to lack of promotions and promotional opportunities in the hospitality.

Hypothesis 7

Gender influences industry leaving intentions, females having higher intentions to leave the hospitality industry than males. This relationship is mediated by actual promotions and perceived promotional opportunities.

Conceptual Model

Figure 1. Conceptual model of hypotheses

Method

Data collection

Data for this study were collected from hotel school graduates using self-administered questionnaire. The questionnaire was made using Qualtrics Survey Software and data was collected digitally by providing the link to the questionnaire to the respondents. This study recruited respondents through alumni organizations of hospitality schools.

The sample

A total of 253 respondents participated in the research. Respondents were excluded who have not graduated from hotel school (n=34). Due to insufficient questions answered further 24 respondents were excluded from the analysis. If answers were given for less than two questions the respondents were excluded.

The characteristics of the remaining sample of 195 respondents are provided in Table 1. Table 1 contains valid percentages, missing answers are excluded. Of the respondents 44.4% were male (n=63) and 55.6% were female (n=79). The mean age of the respondents is 32.6 years, ranging from 20 to 57 years of age. Most of the respondents 46.5% were single, 24.6% living together and 23.9% married; 75% had no children. Most respondents, 58% had a bachelor’s or a 3 to 4 year college degree (n=112); 30.1% had master’s degree (n=58). 25.8% of the respondents graduated in The Netherlands (n=48), 43.5% (n=81) in the rest of Europe and 30.6% (n=57) in the rest of the world.

Table 1 Demographic characteristics N % Gender Male 63 44.4% Female 79 55.6% Age 20-32 96 67.6% 33-45 30 21.1% 46-57 16 11.3% Marital status Single 66 46.5% Living together 35 24.6% Married 34 23.9%

Separated, Divorced, Widowed 7 4.9% Number of children no children 105 75.0% 1 13 9.3% 2 15 10.7% 3 6 4.3% 4 or more 1 0.7% Type of Education

2-year College Degree 14 7.3% 3 to 4 year College Degree, Bachelor Degree 112 58.0% Masters Degree, Post-Bachelor Degree 58 30.1% Doctoral, PGD, Professional 9 4.7% Country of

education

The Netherlands 48 25.8% Rest of Europe 81 43.5% Rest of the World 57 30.6% Year of graduation 2009 to 2013 81 43.1% 2004 to 2008 50 26.6% 2003 and earlier 57 30.3% Hospitality experience

less than 5 years 82 42.7% 5 to 10 years 61 31.8% more than 10 years 49 25.5% Country of

residence

The Netherlands 42 31.1% Rest of Europe (not The Netherlands) 41 30.4% Rest of the World (not Europe) 52 38.5% Employment

situation

Employed in hospitality 110 57.0% Employed but not in hospitality 51 26.4%

Not employed 32 16.6%

The year of graduation ranges from 1976 to 2013, most of the respondents graduated in 2012 (n=21). 42.7% (n=82) had been working in the hospitality industry for less than 5 years, 31.8% (n=61) between 5 and 10 years and 25.5% (n=49) for more than 10 years. 31.1% (n=42) reside in The Netherlands, 30.4% (n=41) in the rest of Europe and 38.5% (n=52) in the rest of

the world. 57% of the respondents are employed in the hospitality industry, 26.4% employed, but not in the industry and 16.6% were not employed.

Measures

Likert scales were employed for the following variables; promotional opportunities, turnover intention, and industry exit intention. On organization related issues Likert scales are acknowledged to be one of the most accurate methods in gathering respondents` opinions and attitudes (Brown, 1996). Perceived promotional opportunities, turnover intention and industry leaving intention were measured using 5 point Likert scale using the following answer choices; strongly disagree, disagree, neither agree or disagree, agree and strongly agree. To handle missing data in the above variables Hotdeck imputation (Myers, 2011) was used.

Perceived promotional opportunities

Perceived promotional opportunities was measured by using the 5 items used by Curry, Wakefield, Prince and Mueller (1986); it was coded in a way, that a higher score on the Likert scale indicated a more positive perception of promotional opportunities. Respondents were asked the following questions: How much do you agree or disagree with each of the following statements about your promotional opportunities at your current organization? 1. Promotions are regular. 2. I am in a dead-end job. (reverse coded) 3. There is an opportunity for advancement. 4. There is a good opportunity for advancement. 5. There is a good chance to get ahead. The perceived promotional opportunities variable was created as a mean of the above five items; the variable is reliable (Cronbach’s Alpha = 0.846). With 143 valid responses perceived promotional opportunities variable has a range of 3.40 with minimum value of 1.60

and a maximum value of 5; mean value of 3.63, standard deviation of 0.82 and a mode of 4. Means test showed no significant difference between respondents working outside or inside the hospitality industry; the F test’s significance was 0.167.

Turnover intention

Turnover intention was measured with three items adapted from Walsh, Ashford and Hill (1985) where a higher score on the Likert scale indicated higher turnover intention. These three items were: „As soon as I can find a better job, I will leave this organization.”; „I am actively looking for a job at another organization.”; „I am seriously thinking of quitting my job.” The turnover intention variable was created as mean of the above three items; the variable is reliable (Cronbach’s Alpha = 0.885). With 143 valid responses turnover intention variable has a range of 4 with minimum value of 1 and a maximum value of 5; mean value of 2.68, standard deviation of 1.19 and a mode of 1.

Among respondents working outside of the hospitality industry the average level of turnover intention was 2.30 while among respondents working in the hospitality industry it was higher, 2.87. The significance of this difference was tested using F-test which had a significance of 0.007. Respondents working in the hospitality industry have a significantly higher turnover intention than respondents working outside of the hospitality. (Table 2)

Table 2 - Means test of turnover intention by employment situation

Employment situation Mean N Std. Deviation Employed in the hospitality industry 2,8702 95 1,16214 Employed but not in the hospitality industry 2,3050 47 1,16683

Total 2,6831 142 1,18988

Industry leaving intention

Industry leaving intention was measured similarly to turnover intention; the wording was changed in the measures of Walsh, Ashford and Hill (1985). The following 3 items were used: „As soon as I can find a better job outside the hotel industry, I will leave the industry.”; „I am actively looking for a job outside of the hotel industry.”; „I am seriously thinking of quitting the hotel industry.”. Industry exit intention variable was created as a mean of the above items; the variable is reliable (Cronbach’s Alpha = 0.895). With 96 valid responses industry exit intention variable has a range of 4 with minimum value of 1 and a maximum value of 5; mean value of 2.40, standard deviation of 1.19 and a mode of 1.

Actual promotions

The following questions were asked about the present employment of each respondent: “When did you start working at your present employer?”; How many times have you been promoted at your present employer?”. Most of the respondents 53.1% have not received a promotion at their present organization (n=76), 26.6% received one (n=38), 7% received two (n=10) and 13.3% (n=19) received three or more promotions. Received promotions variable had a mean of 0.80, median of 0, standard deviation of 1.05 and a variance of 1.10. Length of employment was calculated by taking the difference between the starting date at present employer and the date on which the questionnaire was filled. Length of employment was measured in month; it had 144 valid responses, a minimum value of 0, a maximum value of 362, mean of 42.76 and a median of 22.

Actual promotion was measured in months spent without promotions. To calculate the value of actual promotion variable the above two variables were used: received promotions and length of employment. The value of actual promotions variable was the length of present employment in months in case of no promotions. In case of promotions received the length of employment was divided into equal parts by promotions in order to calculate the value of the variable. It was calculated the following way: “Actual promotions” = (”received promotions” + 1) / “length of present employment in months”. Actual promotions variable with 143 valid responses has a mean of 22.52, median of 12.25, standard deviation of 27.37, a variance of 748.92 a range of 175 with a minimum value of 0 and a maximum value of 175.

Demographic characteristics

The following demographic characteristics are examined: gender, age, marital status, number of children, nationality and residence. Gender was coded as 1 for male, 2 for female; age of the respondents was calculated on the basis of the year of birth. For marital status the following options were given: single, living together, married, separated, divorced and widowed. Further data was collected about time spent in the hospitality industry (measured in months), level of education, country of the educational institute and year of diploma received. The summary of demographical characteristics is in Table 1.

Employment situation

To determine employment situation the following answer choices were given; “I am employed in the hospitality industry.”, “I am employed, but not in the hospitality industry.”, “I am not employed.” Most respondents, 57.0% were employed in the hospitality industry

(n=110), 26.4% were employed but not in the hospitality (n=51) and 16.6% were not employed (n=32). For the analysis only respondents working in the hospitality industry were included.

Data analysis and results

The Statistical Package for Social Sciences (IBM SPSS version 20 software) was used for statistical analyses. To test the hypotheses the Process procedure written by Hayes (2012) was used; it is a computational tool for path analysis-based moderation, mediation and their combinations.

Mean scores and bivariate correlations

Means, standard deviations and inter-correlations of study variables are presented in Table 3 with reliabilities on the diagonal. Strong positive correlations were found between received promotions and perceived promotional opportunities. Strong negative correlations were found between perceived promotional opportunities on the one hand and turnover intention and industry leaving intention on the other. There was a weaker negative correlation between received promotion and industry leaving intention and there were no significant correlation between received promotion and industry leaving intention. Gender showed no correlation with the study variables with two exceptions. A moderate positive correlation was found with actual promotion and a weaker negative correlation with industry leaving intention.

Table 3. Scale means, SDs, Intercorrelations and Reliabilities

Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14

1. Gender (1-male, 2-female) 1,61 0,49 (-)

2. Age 32,27 8,73 .02 (-)

3. Marital status 1,97 1,01 .01 .43*** (-)

4. Number of children 0,36 0,71 -.02 .35** .36*** (-)

5. Level of education 2,22 0,53 .07 -.08 -.09 -.21* (-)

6. Years since graduation 8,11 8,50 .03 .92*** .33** .25* -.11 (-)

7. Years in hospitality 9,80 7,41 .01 .74*** .20* .29** -.12 .73*** (-)

8. Number of past employments 2,01 1,46 -.16 .36*** .17 .19 .14 .35** .28** (-)

9. Length of present job in years 4,42 5,29 .14 .57*** .05 .27** -.12 .61*** .60*** -.07 (-)

10. Actual promotion 21.60 22.16 .28** .47*** .11 .43*** -.24* .50*** .47*** -.11 .77*** (-)

11. Received promotions 0.90 1.18 -.04 .37*** .16 .03 .16 .38*** .29** .04 .59*** .18 (-)

12. Perceived promotional opportunities 3,56 0,86 .04 .17 .05 -.17 .11 .17 .23* .05 .16 -.15 .36*** (.85)

13. Turnover intention 2,77 1,16 -.15 -.15 -.08 .10 -.13 -.16 -.07 .09 -.16 .05 -.13 -.52*** (.89)

14. Industry leaving intention 2,24 1,08 -.23* -.17 .00 .08 -.21* -.17 -.01 -.02 -.09 -.12 -.25* -.42*** .53*** (.90) Note. Correlations greater than r = I.35I are significant at p < .001 ***; correlations greater than r = I.26I are significant at p < .01 **

and correlations greater than r = I.19I are significant at p < .05 *. (one-tailed)

Hypothesis testing

To test the hypotheses Process computational tool (Hayes, 2012) was used. The advantage of using Process is that all hypotheses for a particular dependent variable can be tested in one model. Process offers tests for a great number of models of mediation; moderation and combinations thereof; models are available for parallel and serial mediation as well.

Model 6 was used from Process to test the hypotheses. Model 6 has one independent variable (X), one outcome variable (Y) and allows to model between 2 and 4 mediators operating in serial. Mediators (M) operating in serial are linked in a (presumed) causal chain, with the first mediator affecting the second, the second the third, and so forth. It is modeled in a way, that the independent variable is the causal antecedent of the mediators and the outcome variable. Model 6 estimates the total direct effect of an independent variable on a single outcome variable, as well as the total and all possible specific indirect effects of the independent variable on the outcome variable through multiple mediators. The conceptual model of model 6 with two mediators is in Figure 2.

Figure 2 – Conceptual Model of Model 6 from Process

Process allows for only one outcome variable, in order to test all hypotheses, Process was run two times; first with turnover intention, and second with industry leaving intention as the

outcome variable. The independent variable entered to Process was gender. According to the hypotheses, gender (X) is the causal antecedent of actual promotions (M1), perceived promotional opportunities (M2) and the outcome variables turnover intention and industry leaving intention (Y). The outcome variables, turnover intention and industry leaving intention are the farthest along in the causal chain and presumed to be ultimate consequences of gender. The two mediators – actual promotions and perceived promotional opportunities – are linked serially in the causal chain. Gender is modeled as causally influencing both actual promotions and perceived promotional opportunities, and perceived promotional opportunities is modeled as causally influenced by actual promotions. Indirect effects of gender on turnover intention or on industry leaving intention are estimated as the product of coefficients for variables linking gender to turnover intention or industry leaving intention through actual promotion and perceived promotional opportunities. There are three such indirect effects, one through actual promotions only, one through perceived promotional opportunities only, and one through both actual promotions and perceived promotional opportunities. These sum to yield the total indirect effect of gender on turnover intention or on industry exit intention. When added to the direct effect, the result is the total effect of gender on turnover intention or industry leaving intention. Process generates 95% bias corrected bootstrap confidence intervals for all indirect effects using 1000 bootstrap samples. Bootstrapped confidence intervals were used rather than the Normal theory-based Sobel test for inferences about indirect effects because of the unrealistic assumption the Sobel tests makes about the shape of the sampling distribution of the indirect effect (Preacher and Hayes, 2004; 2008; Hayes, 2009).

The conceptual models of the two runs of Process model 6 are presented in Figure 3 and Figure 4. Hypotheses 1, 3, 4, 5 and 6 were tested in the first run and hypotheses 2, 3, 4, 5 and 7 were tested in the second run. In both runs hypotheses 3, 4 and 5 were included; first these results will be discussed. The outputs of Process’ first and second runs are in Table 3 and Table 4.

There was a significant effect of gender on actual promotion, p=0.009, however there was no significant relationship between actual promotions and perceived promotional opportunities (p=0.107) or between gender and perceived promotional opportunities (p=0.470). Gender had an effect of 13.2139 on actual promotions, where females tend to spend longer time without promotion than males; the summary of all path coefficients can be seen in Table 7. Gender explains 8.86% of the variation in actual promotion (R square=0.0886), a summary of all R-squares are presented in Table 8.

Figure 3 – Conceptual model of hypotheses tested in the first run of Process

In the first run of Process turnover intention was entered as the outcome variable (Figure 3.). Process estimated the direct effect of gender on turnover intention, as well as the indirect

effects of gender on turnover intention through actual promotion and perceived promotional opportunities. The output of the first run of process is in Table 3.

There was no significant total effect of gender on turnover intention, the significance of the total effect was 0.118; modern thinking about mediation analysis does not require evidence of a total effect prior to the estimation of direct and indirect effects (Cerin & MacKinnon, 2009; Hayes, 2009; Rucker, Preacher, Tormala, & Petty, 2011). There was no significant direct or indirect effect of gender on turnover intention, the significance of the direct effect of gender on turnover intention was p = 0.125; the bootstrap confidence interval for the total indirect effect was between -0.3586 and 0.2363. There was no significant serial mediation found, the bootstrap confidence interval for the serial mediation was -0.0018 to 0.2031. There was no evidence for parallel mediation, where the two mediators were tested for being affected by gender, in turn affecting turnover intention, but not transmitting their effect on the other mediator. The bootstrap confidence intervals for the parallel mediation were between -0.1393 to 0.1431 for actual promotion as a mediator and were -0.3702 to 0.1433 for perceived promotional opportunities as a mediator. The summary of all direct and indirect effects is in Table 1.

Perceived promotional opportunities had a significant effect of -0.6507 on turnover intention (p=0.000), this relationship was negative; the better promotional opportunities the respondent perceived the less turnover intention was shown. The proposed antecedents - perceived promotional opportunities, actual promotion and gender - explained 25.97 % of the variation in turnover intention (R-square = 0.2597, p=0.000, Table 8).

Figure 4 – Conceptual model of hypotheses tested in the second run of Process

In the second run hypothesis 2, 3, 4, 5 and 7 were tested; the conceptual model is presented in Figure 4. Model 6 from Process estimated the direct effect of gender on industry leaving intention, as well as the indirect effects of gender on industry leaving intention through actual promotion and perceived promotional opportunities. Table 4 contains the output of the second run of Process in SPSS.

The total effect of gender on industry leaving intention was not significant (p=0.063), however the direct effect of gender on industry leaving intention was significant (p=0.032), the effect size was -0.5118. In contrary to hypothesis 7 this relationship was negative, males tended to have higher industry leaving intention than females. There was no indirect effect of gender on industry leaving intention; the bootstrap confidence interval for the total indirect effect was -0.2203 to 0.2749. There was no significant serial mediation found, the bootstrap confidence interval for the serial mediation was -0.0010 to 0.1695. There was no evidence for parallel mediation. The bootstrap confidence interval for the parallel mediation was -0.0159 to 0.2460 for actual promotion as a mediator and was between -0.3203 and 0.0958 for perceived

promotional opportunities as a mediator. The summary of all direct and indirect effects is in Table 1.)

The proposed variables - perceived promotional opportunities, gender and actual promotion - explained 23.16% of the variation in industry leaving intention, (R-square=0.2316, p=0.000). Perceived promotional opportunities had a significant effect on industry leaving intention (p=0.000), the effect size was -0.4977; the better the respondent perceived promotional opportunities the more likely they expressed low intention to leave the hospitality industry. There was no significant direct effect of actual promotions on industry leaving intention (p=0.294).

Exploratory analyses suggested that the relationship between actual promotion and perceived promotional opportunities was moderated by the number of promotions received. It can be expected that the more promotion was received the less important the frequency of the promotions became. Hicks (1990) found that females had to prove their performance before given the same place as men; this suggests women are facing more obstacle than men in obtaining their fist promotion. The hypotheses were tested again only among respondents with one or no promotion (sample size: 58). Model 6 from Process was used; the output of Process’ two runs are presented in Table 5 and Table 6.

Gender had a significant effect (effect size: 14.6035) on actual promotions (p=0.007), 12.34% of the variance in actual promotions can be explained by gender (R square=0.1234, p=007), where females tend to spend longer time without promotion than males. This

explained variance is greater than among all respondents, where it was only 8.86% (summary of R-squares is in Table 8).

Among respondents with less than two promotions actual promotions had a significant effect (p=0.004) on perceived promotional opportunities, the size of this effect was -0.0176, the longer the respondents spent without promotions the more negative their perception of promotional opportunities were, this relationship was not significant among all respondents,. There was no significant direct effect of gender on perceived promotional opportunities (p=0.244) however actual promotions and gender together explained 13.93% of the variance of perceived promotional opportunities (R-square=0.1339, p= 0.016), while this was not significant among all respondents.

Perceived promotional opportunities was a significant predictor of both turnover intention (p=0.000) and industry leaving intention (p=0.014). Perceived promotional opportunities had an effect of -0.6188 on turnover intention and an effect of -0.3990 on industry leaving intention; a better perception of promotional opportunities was related to a lower intention to leave the organization or the industry. Actual promotions had no significant direct effect on either turnover intention (p=0.264) or industry leaving intention (p=0.136). The proposed antecedents – perceived promotional opportunities, gender and actual promotion – explained 35.52% of the variation of turnover intention (R-square=0.355, p=0.000) and 23.49% of the variation of industry leaving intention (R-square=0.235, p=0.002) (R-squares are in Table 8).

Gender had no significant total effect on turnover intention (p=0.161) or industry leaving intention (p=0.114), but had a significant (p=0.036) negative direct effect on industry leaving

intention with an effect size of -0.6201; gender had no significant – but very close to significant - direct effect on turnover intention (p=0.060) with an effect size of -0.5127. There was no significant total indirect effect of gender on neither turnover intention (p=0.093) nor industry leaving intention (0.150), but there were significant specific indirect effects of gender on both turnover intention and industry leaving intention evidenced by bootstrap confidence intervals that do not contain zero.

Figure 5 - Conceptual model of the first run of Process with path coefficients

Gender had an indirect effect on turnover intention; this serial mediation (ind2 in Table 5) carries the effect of gender to turnover intention through actual promotion and perceived promotional opportunities. It was estimated by Process as the product of the following coefficients, ga=14.6035, ao=-0.0176 and ot=-0.6188 or 0.1587, with a 95% bootstrap confidence interval of 0.0519 to 0.3720. The conceptual model with the path coefficients is in Figure 5. Longer time spent without promotion resulting from being female translates into more negative perception of promotional opportunities which in turn leads to higher intention to leave the organization. There was no evidence for parallel mediation between gender and

turnover intention; the bootstrap confidence intervals for the parallel mediation were between -0.0448 and 0.3850 for actual promotion as a mediator and were between -0.5197 and 0.0993 for perceived promotional opportunities as a mediator.

Figure 6 - Conceptual model of the second run of Process with path coefficients

Gender had two specific indirect effects on industry leaving intention. The first effect carries the effect of gender on industry leaving intention through actual promotions only, bypassing perceived promotional opportunities. This indirect effect (ind 1 in Table 6) is the product of ga=14.6035 and ai=0.0112 or 0.1631 with 95% bootstrap confidence interval of 0.0074 to 0.3798 (path coefficients are presented in Figure 6). Females tended to wait longer for a promotion than males and this was associated with higher industry leaving intention, independent of perceived promotional opportunities. The other indirect effect of gender passes through both actual promotion and perceived promotional opportunities. It is estimated as the product of ga=14.6035, ao=-0.0176 and oi=-0.3990 or 0.1023, with a 95% bootstrap confidence interval of 0.0256 to 0.3116 (ind 2 in table 6). The longer time spent without promotion resulting from being females translates into a more negative perception of promotional opportunities which in turn leads to a higher intention to leave the hospitality industry.

Perceived promotional opportunities alone did not mediate the relationship between gender and industry leaving intention, the bootstrap interval was -0.4480 to 0.0417. The summary of all direct and indirect effect sizes is in Table 2.

Discussion

This study has similar finding to Walsh and Taylor (2007) and Stalcup and Pearson (2001) who identified the absence of career advancement opportunities as the main reason for managerial turnover and industry exit. Hypothesis 1 and 2 were supported, hospitality graduates who perceived better promotional opportunities tend to have lower turnover intention and lower industry leave intention. As hypothesized actual promotion was affected by gender, females needed to wait longer to be promoted than males. Gender explained a higher percentage of variance in actual promotions among respondents with less than two promotions than among all respondents; this as align with Hicks (1990) findings that females had to prove their performance before provided with similar opportunities to men. Gender had a significant effect on actual promotion among all respondents and among respondents with less than two promotions as well; hypothesis 3 was supported.

There was no relationship between actual promotion and perceived promotional opportunities when looking at all respondents working in hospitality industry, however among respondents with less than two promotions actual promotion was a significant predictor of perceived promotional opportunities. Actual promotion was measured in months spent without promotions. It might be the case, that the amount of time spent without promotion does not

affect perceived promotional opportunities, however it also can be that there is a relationship, but it was not shown due to measurement or sampling errors. The time spent since the last promotion was only estimated by dividing the employment time into equal periods. This means that a respondent who was promoted a month ago and a respondent who was promoted a year ago would have the same value of actual promotion if respondents had the same length of employment and amount of promotions. The longer employment time and the more promotion the respondent received the bigger the difference can be between the estimation and the real time spent without promotion. It is also possible that employees with more promotions are less sensitive on how long they are without promotions in their view of promotional opportunities. Regardless of the causes, hypothesis 4 was only partly supported as the relationship between actual promotion and perceived promotional opportunities was significant among respondents with less than two promotions only. The longer a respondent had to wait to be promoted the more negative their view of promotional opportunities was.

Gender was not directly related to perceived promotional opportunities, however gender seemed to affect perceived promotional opportunities through actual promotions among respondents with less than two promotions; there was partial support for hypothesis 5.

Gender showed no direct relationship with turnover intention. There was also no indirect effect of gender on turnover intention among all respondents employed in the hospitality industry, but among respondents with less than two promotions there was a significant indirect effect; where gender had an indirect effect on turnover intention through actual promotions and perceived promotional opportunities. Being female resulted in longer time spent without

promotion, which translated into a more negative perception of promotional opportunities; in turn this led to higher intention to leave the organization. The above described serial mediation is in align with the hypothesis, however as a direct effect, gender showed a nearly significant negative effect on turnover intention. This suggests that among females with one or no promotion the variance of turnover intention can be explained to a great proportion by actual promotions and perceived promotional opportunities while for males these effects are smaller or not present. This study contrary to previous findings (e.g. Lyness and Thompson (1997); Rosin & Korabik (1995)) showed no higher turnover intention for females than males; it might be due to the characteristics of the sample, most of the respondents were quite young, more than two thirds of the respondents fell in the age-category of 20-32. It might be, that in an older age, as females are falling behind more in their career compared to males, they are tend to express higher turnover intention. It is also possible, that previous finding no longer apply for this younger generation, and females no longer have higher intention to leave. Partial support was found for hypothesis 6; gender had only an indirect effect on turnover intention and only among respondents with less than two promotions.

Contrary to hypotheses 7, gender had a negative relationship with industry leaving intention; males tended to have higher industry leave intention than females. There was no indirect effect of gender on industry leaving intention among all respondents employed in the hospitality industry, but among respondents with less than two promotions there were two significant specific indirect effects. The first effect is supportive of hypothesis 7, the effect of gender on industry leaving intention passes through both actual promotion and perceived

promotional opportunities. The longer time spent without promotion resulting from being females translates into a more negative perception of promotional opportunities which in turn leads to a higher intention to leave the hospitality industry. The other indirect effect carries the effect of gender on industry leaving intention through actual promotions only, bypassing perceived promotional opportunities. Females tended to wait longer for a promotion than males and this was associated with higher industry leaving intention, independent of perceived promotional opportunities. This specific indirect effect was not significant for turnover intention as outcome variable. The explanation can be that female graduates compare their opportunities with other female hospitality employees when perceiving their promotional opportunities, but a pull factor from other industries may have an important role when considering intentions to leave. Gender was related to actual promotions only among respondents working in the hospitality industry and not among respondents employed outside of the industry, suggesting more equal opportunities outside of the hospitality industry. This may have provided females with a stronger pull factor from other industries than from the hospitality industry, resulting that females waiting longer for promotions expressed higher intentions to leave the industry, independent of their perception of promotional opportunities. Thus only partial support was found for hypothesis 7 as males had higher intentions to leave the industry, however as an indirect effect being female was related to higher intentions to leave through actual promotions and perceived promotional opportunities among respondents with no or one promotion. This finding suggests that the industry leaving intention of females can be explained to a great extent by actual promotions and perceived promotional opportunities while males have different underlying factors when considering industry leaving

intention, the reasons behind these intentions were not explained by the variables of this study.

The above finding suggests that there is no difference of turnover intention of males and females, but the lack of promotions can cause high turnover intention for females through negative perception of promotional opportunities. There was a significant difference between females and males industry leaving intention, males tended to have higher intentions to leave the industry. This study could not explain this difference, the study variables – actual promotion and perceived promotional opportunities – provided explanation for higher industry leave intention of females and only among respondents with few or no promotions.

Conclusions

As pointed out in the literature review the hospitality industry in characterized by high labor turnover and industry exits. In this study too, graduates employed in the hospitality industry had a significantly higher turnover intention than respondents working in other industries. As shown in earlier studies (Blomme, van Rheede & Tromp, 2010a; Walsh & Taylor, 2007; Lyness & Judiesch, 2001) promotional opportunity is negatively related to turnover intentions. In this study too, perceived promotional opportunities was strongly related to both turnover intention and industry leaving intention, however actual promotion had no direct effect on neither turnover intention nor industry leaving intention. This study aimed to investigate the reasons of higher turnover and industry exit of females but it was found that males tended to have higher industry leave intention than females and gender had no significant effect on turnover

intention. The respondents in the sample were quite young (more than two-thirds of the respondents fell into the 20-32 age group); it might be that among young graduates females are the ones who tend to stay in the industry and this changes only when females are presented with unsatisfying promotional opportunities later on their careers. So female’s industry exit intention might “catch up” later with male’s because of the unequal opportunities females are presented with.

Males had higher intention to leave the industry than females; the study variables could not give explanation for this phenomenon. As an indirect effect being female was related to higher intentions to leave, where being female was related to longer time spent without promotions, this led to a less positive perception of promotional opportunities, which in turn was related to higher intentions to leave. These findings suggest that females and males have different underlying factors when considering intentions to leave. For females the intentions to leave can be explained by the lack of promotions and opportunities while for males the motivations are different.

Limitations

Limitation of the applied computerized online self-administered questionnaire technique is that the respondent’s honesty and seriousness cannot be verified; on the other hand participants may tend to give more frank and truthful answers (Konradt, Hertel & Joder, 2003). The sample size (n=77) was quite small especially when only respondents with less than two promotions (n=58) were included; thus the study findings might be not replicable. There is a danger of common method bias because the predictor and the criterion variables were obtained from the same source. According to Cote and Buckley (1987) on average 26.3% of the variance in a typical research measure might be due to systematic measurement error, such as common method biases. It is difficult to prove causal relationships as all the variables were measured at the same time. It seems likely that perceived promotional opportunities had an effect on turnover intention and industry leaving intentions; but it can be said definitely only if it is tested in a longitudinal study. Also as industry leave and turnover was measured in intentions it cannot be said that actual industry exit was higher among males than females or that gender had no effect on actual turnover; the decision of an individual to actually leave the organization might be dependent on additional or other factors. A longitudinal study in which hospitality graduates are followed throughout their career would yield valuable results.

Managerial implications

This research showed that contrary to previous findings women have lower intentions to leave the hospitality industry, and nearly significant lower intentions to leave the organization than males. It was also shown that actual promotions and perceived promotional opportunities were explaining higher intentions to leave of females. These findings can be beneficial for human resource professionals in the hospitality industry when managing employee retention. Most of the respondents were in their twenties, so it can be said that at the beginning of their career women tend to have low intentions to leave. Hospitality employers and human resource professionals should work on maintaining this low intention to leave by providing equal promotional, developmental opportunities and a clear career path for their highly educated female employees.

Works Cited

Almeida, M. C., & Horn, D. M. (2004). Is daily life more stressful during middle adulthood? The University

of Chicago Press, 425-451.

Australian Bureau of Statistics. (2008). Employee earnings and hours (Cat. No. 6306.0). Commonwealth

of Australia .

Blomme, R. J., van Rheede, A., & Tromp, D. M. (2010a). The use of the psychological contract to explain turnover intentions in the hospitality industry: a research study on the impact of gender on the turnover intentions of highly educated employees. The International Journal of Human Resource Management ,

21 (1), 144 – 162.

Blomme, R. J., Van Rheede, A., & Tromp, D. M. (2010b). Work-family conflict as a cause for turnover intentions in the hospitality industry. Tourism and Hospitality Research , 269-285.

Brien, A. (2004). “Do I want a job in hospitality? Only till I get a real job!”. In Proceedings of the New

Zealand Tourism and Hospitality Research Conference Edited by: Smith, K. A. and Schott, C. Wellington,

New Zealand

Brown, J. D. (1996). Using Surveys in Language Programs. Cambridge University Press .

Cerin, E., & MacKinnon, D. P. (2009). A commentary on current practice in mediating variable analysis in behavioral nutrition and physical activity. Public Health Nutrition, 12, 1182-1188.

CERT. (1999). Hospitality 2005: A Human Resource Strategy. CERT .

Charlesworth , S . ( 1994 ) Productivity in the Hospitality Industry: A Consultative Approach to the

Development of Performance Indicators . Canberra, Australia: Commonwealth department of Industrial

Relations, Equal Pay Unit .

CIPD. (2000). Labour Turnover. Chartered Institute of Personnel and Development (CIPD) .

Coff, R. W. (1997). Human assets and management dilemmas: Coping with hazards on the road to resource-based theory. Academy of Management Review (22), 374-402.

Cote, J. A., & Buckley, R. (1987). Estimating trait, method, and error variance: Generalizing across 70 construct validation studies. Journal of Marketing Research (24), 315-318.

Curry, J., Wakefield, D., Prince, J., & Mueller, C. (1986). On the causal ordering of job satisfaction and organizational commitment. Academy of Management Journal, 29 , 847-58.

Davidson, C. G., & Wang, Y. (2011). Sustainable Labor Practices? Hotel Human Resource Managers Views on turnover and Skill Shortages. Journal of Human Resources in Hospitality & Tourism (10:3), 235-253. Davidson, M. C., Timo, N., & Wnag, Y. (2009). How much does labor turnover cost? A case study of Australian four- and five- star hotels. International Journal of Contemporary Hospitality Management , 451-466.

Deery, M.A., Iverson, R.D. (1996), "Enhancing productivity: intervention strategies for employee turnover", in Johns, N. (Eds), Productivity Management in Hospitality and Tourism, Cassell, London, 68-95.

Deery, M. A., & Shawn, R. N. (1997). An exploratory analysis of turnover culture in the hotel industry of Australia. International Journal of Hospitality Management (16 (4)), 375-392.

Diaz, P.E., Umbreit W.T. (1995). Women leaders—a new beginning. Hospitality Research Journal, 18. 47-60

Griffeth, R. W., Hom, P. W., & Gaertner, S. (2000). A Meta-Analysis of Antecedents and Correlates of Employee Turnover: Update, Moderator Tests, and Research Implications for the Next Millennium.

Journal of Management (26), 463-488.

Guerrier, Y. (1986). Hotel manager-an unsuitable job for a woman? The Service Industries Journal , 6 (2), 227.

Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium.

Communication Monographs, 76, 408-420.

Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling. Retrieved from

http://www.afhayes.com/public/process2012.pdf

HCITB, H. a. (1984). Women`s Path to Management. London.

Hicks, L. (1990). Excluded Women: How can this Happen in the Hotel World? The Service Industries

Journal , 348-363.

Hiemstra, S. J. (1990). Employment policies and practices in the lodging industry. International Journal of

Hospitality Management ( 9), 207-221.

Hinkin, T., & Tracey, B. (2006). Development and use of a web-based tool to measure the costs of employee turnover: Preliminary findings. CHR Reports, Ithaca, NY: Cornell University Center for

Hospitality Research. , 6, no. 6.

Hogan, J. J. (1992). Turnover and what to do about it. . The Cornell Hotel and Restaurant Quarterly, 40-45.