Determination and analysis of factors affecting turnover intentions of airline pilots in Turkey

In recent years, many researches have been conducted on turnover intention in different job groups. The fact behind this interest is that high personnel turnover rates lead to decrease in organizational efficiency. It has been realized personnel leaves bring high costs and significant losses to the companies. In the related studies, turnover intention(or intention to quit) is employed as the most important predictor of quits and factors affecting the intention to leave with their effects are examined. Despite these researches in different countries and occupations, no study has yet been conducted on turnover intentions of airline pilots in our country and in the world. In addition, the turnover rate of airline pilots in Turkey have significantly increased recently. Despite the rapid development of air transport in Turkey, pilot shortage leads to work-overload and fatigue in existing pilots. In the light of this information, factors affecting the turnover intention of airline pilots were examined in this study, considering both the academic gap and the practical need in our country. Data collected from 214 airline pilots working in different airlines in Turkey with survey method. The questionnaires were made both on paper and online. Just 5 invalid surveys were removed. The scales measuring factors were taken from articles in the literature. With a pilot survey, the reliability and clarity of scales were tested. The five-point Likert scales that ranged from "strongly disagree" to "strongly agree" was used to evaluate the questions. Structural equation modeling was employed for the analysis using AMOS software. "Stats Tool Package" that is an Excel tool was used for validity tests and SPSS Process Macro was used for advanced mediation- moderation effect tests. In the literature, hundreds of papers on the subject were examined and related factors were included in the theoretical model considering also the aviation sector in our country. In this context, the effects of work-family conflict, job satisfaction, job stress, organizational commitment, perceived nepotism and perceived job opportunities on intention to quit were investigated. Furthermore, the relationships between these each factor were tested. In the study, measurement model and structural model were analyzed with two stage-approach. At first, measurement model included 37 items describing seven constructs For the measurement model, confirmatory factor analysis was conducted to validate measurements and obtain final measurement model. For this purpose, each concept -called latent variable in structural equation modeling- was tested to provide reliability, construct, convergent, and discriminant validity, and also goodness of fitness. These analyses were conducted step by step for every latent variable via factor loadings, modification indices given by AMOS and Stats Tools Package. With elimination of some observed variables, measurement model including 30 items was finalized. This reveals the factor structure in the final measurement model fitted well to the data collected. In the analysis and results section, the structural model, the significance values of relationships, path coefficients(beta weights), factor loadings and mediation- moderation effect analysis were presented in detail. In conclusion, all of the results were evaluated and managerial implications were derived with the purpose of providing suggestions to airline companies for pilot retention. 21 hypotheses were tested including 11 basic relationships and 10 indirect effects(mediator-moderator), 19 of which were accepted. The results of hypothesis tests are presented in a table. The model explained substantial variance in intention to quit job(R2=0,79). This indicates the robustness and power of the model established. As a result of hypothesis tests, it was supported that work- family conflict and perceived nepotism positively influence job stress among airline pilots. Another findings are that job stress negatively influence organizational commitment and positively influence intention to quit. However, the hypothesis proposing job stress negatively influence job satisfaction was rejected. The hypothesis proposing organizational commitment positively influences job satisfaction was accepted as expected. Moreover, it was found that perceived nepotism has a significant and negative effect on job satisfaction. However, the hypothesis proposing nepotism has a negative influence directly on intention to quit was rejected. The positive effect of job satisfaction and organizational commitment on intention to quit job are verified. Another finding is perceived alternative job opportunities positively influence intention to leave. One of the remarkable findings is that job stress has the greatest effect on intention to quit. Besides job stress, job satisfaction, organizational commitment and perceived job opportunities, respectively, come to the degree of impact to intention to leave. At this point, it is essential to realize the fact that the pilot resignations are not dependent on just a single factor but it is a multidimensional issue. Perceived job opportunities and job stress are positively correlated, whereas job satisfaction and organizational commitment are negatively correlated with intention to leave. The effect of job stress is more than twice the effect of job satisfaction and organizational commitment. The effect of perceived job opportunities is half of the effect of job satisfaction and organizational commitment. It was measured that pilots have high organizational commitment, high job stress, very high work- family conflict, very high alternative job perception and medium job satisfaction. Therefore, it can be concluded organizational commitment has a significant blocking effect against a rise in pilot turnovers. Job stress is a critical issue for airline pilots since they carry million dolar aircrafts and hundreds of people's lives every day. Any stress-related pilot error can lead catastrophic consequences for human, economy and security. Therefore, it is critical to analyze stressors for pilots in terms of aviation safety. In the study, two factors affecting job stress were studied. As a result, it was found that work- family conflict has a huge positive effect on pilot's stress. As it is known, night-day irregular flights and changing flight schedules seriously affect pilots' private and family lives. Approximately 60% of pilots were observed to have a work-family conflict in the survey. Besides work-family conflict, perceived nepotism(unprofessional and unfair practices in workplace) triggers directly pilots' stress. These two factors accounted for 56% of the the variance in job stress in pilots. Since the direct association between work- family conflict and intention to quit was not found, the indirect effect was examined. Identifying these mechanisms between these two factors will be helpful to understand why pilots exactly prefer to leave their organizations. Mediator variables are variables that are intermediate in the causal relation between two other variables. That is, mediator variables (the second variable in the causal chain) transmit changes in the first variable (the cause) to the third variable (the effect). In the case, four different (full)mediating effects between work-family conflict and intention to leave were identified. Job stress, job satisfaction, job stress - organizational commitment(serially), and job stress - organizational commitment - job satisfaction(serially) mediated the relationship between work-family conflict and intention to leave. In order to verify these mediation effects, firstly the significance of indirect effects were tested with bootstrapping method in AMOS. Then, structural model was run without mediator variables to see the direct effect between cause and effect variable. Since AMOS is not able to detect the mediation paths when there are multiple paths between cause(predictor) and effect(criterion) variable, Process Macro was employed to explore exactly mediation paths and mediator varibles. Via Process, the paths and variables having significant mediation effects were determined. To illustrate possible mediation effects, the path diagrams were presented. In addition, boostrapping lower and upper levels for confidence intervals for each mediator variables(or paths)were given to reveal their significance. These results mean without the mediator variables including job stress, job satisfaction and organizational commitment, the work-family conflict has a significant effect on intention to leave. However, when these variables are involved, work-family conflict shows its effect through these means. For example, pilots' work-family conflict increases work stress; which triggers their thinking about leaving the job. In another mediator effect, the work-family conflict reduces job satisfaction; which causes intent to leave. In the third mediating effect that is serial, work-family conflict increases work stress and work stress injures organizational commitment, which causes intent to leave. In the last mediation effect, it was observed that work-family conflict affects job stress, it affects organizational commitment, it affects job satisfaction and it affects intention to leave in a serial manner. As a result, although work- family conflict has no direct effect on intention to leave, it positively and remarkably influences intention to leave through affecting job stress, organizational commitment and job satisfaction. With these indirect effects, work- family conflict's degree of impact reaches organizational commitment's degree of impact crossing job satisfaction's degree of impact on intention to leave. Similarly, since the direct association between perceived nepotism and intention to quit was not found, the indirect effect was examined. The relationship between nepotism and intent to leave was found to be mediated in four different ways. It has been discovered that job stress, job satisfaction, job stress-organizational commitment(serially) and job stress-organizational commitment-job satisfaction(serially) play mediating role in this relationship. It was noted that perceived nepotism has a huge and direct negative effect on job satisfaction. Job satisfaction is directly explained by perceived nepotism and organizational commitment. Work- family conflict and job stress have negative indirect effects. Totally, these variables accounted for 78% of the the variance in job satisfaction of pilots. In the study, a moderation effect analysis was carried out. Moderation occurs when the magnitude and/or direction of the relation between two variables depend on the value of a third variable called a moderator variable. In the case, pilots were divided into two groups having low alternative job perceptions and high alternative job perceptions. The model was run in AMOS for these two groups. The regression weights and critical ratios were employed in Stats Tool Package to see group differences. It showed that perceived job opportunities have a moderating effect between organizational commitment and intent to leave. A table was presented to illustrate how intention to quit changes with organizational commitment under two categorical job perception level. The negative and significant effect of organisational commitment on intention to leave get lost and unsignificant under the condition of high perceived job opportunities. Therefore, airlines should focus more on approaches to reduce job stress and increase job satisfaction compared to developing organizational commitment for retention of pilots. In future studies section, some recommendations were given about how related studies can be more extensive. Including fatique and performance as additional factors in these models will be useful to understand concrete results of the intangible constructs mentioned. In addition, it will be beneficial to involve all stressors for airline pilots in terms of aviation safety and turnover intention researches. Since examined factors such as intention to quit and job satisfaction are relative concepts, the effects of pilots' characteristics and demographic factors(total flight hours, aircraft type, civil-military origin) needs to be researched as a further study.