The chapter revealed the result of data analysis, the finding of each hypothesis and discussion in this study. Descriptive statistics analysis showed the demographic information of respondents and the current condition of cabin crews’ job satisfaction and retention, afterward, the result of correlation analysis, linear regression between each dimension of job satisfaction and retention intention would be presented as well. Finally, hierarchical regression revealed the result of the relationship of job satisfaction and retention intention and the interaction effect of demographic variables (age, education level, position and tenure) between job satisfaction and retention intention.
Descriptive Analysis
The result of sample distribution, and job satisfaction and retention intention would be discussed in this section.
Sample Distribution
The demographic information in descriptive statistics was collected from 226 cabin crews, showing the distribution of gender, age, education level, position and tenure. The frequency and percentage of the demographic information were summarized in table 4.1.
As for the gender, the percentage of female (95.1%) was extremely higher than male (4.9%). Due to the nature of work at the airlines, the ratio between male and female was 1 to 9. In terms of age, since most female cabin crews would confront marriage pressure, health problem, family issue and so on, the result showed 170 respondents (75.2%) were below 32 years old and the age of 56 respondents (24.8%) were above 32. In the case of education level, 184 respondents (81.4%) owned bachelor degree while the percentage of associate (8.4%) and master (10.2%) was quite close. About the position, the ratio of management (35.4%) and non-management (64.6%) was almost one to two. The last but not the least, the tenure below 1 year of respondents was 65 (28.8%) while 58 respondents (25.7%) has worked for 1 year to below 3 years. 37 respondents (16.4%) has worked for 3 to below 7 years, as 66 respondents (29.2%) has worked for more than 7 years.
To sum up, the ratio of male and female was almost the same as previous distribution, approximately one to nine (Association of Wage-Earners, 2000) as a result of the characteristic of cabin crews. In hence, due to limited convenient and snowball sampling, major respondents are young female cabin crews, less than 32 years old with bachelor degree and entry level.
Table 4.1.
Descriptive Statistics on Sample Characteristics (N=226)
Demographic variables Category Frequency Percentage
Gender Male 11 4.9
After conducting descriptive statistics on pay, promotion, supervision, fringe benefits, contingent rewards, operating procedures, coworkers, nature of work, communication, total job satisfaction and retention, the result presented as the table 4.2 below.
Compared with different dimensions of job satisfaction, the scoring of minimum, maximum and standard deviation is quite the same. However, the percentage of minimum scoring is not identical. For instance, fringe benefits are much higher than other dimensions of job satisfaction, inferring that respondents are less satisfied with fringe benefits than the others.
On the other hand, although the maximum scoring of each dimension of job satisfaction is extremely high, quite few respondents satisfied with these sub-variables indeed. As for the total job satisfaction, the scoring shows that each respondent has different perception about it;
therefore, the standard deviation is little bit large (SD =18.051). Furthermore, the qualities of total job satisfaction scoring are 86, 100.5, 112 respectively, meaning that the thought of overall job satisfaction for most respondents are neutral, even partial to negative. About retention intention, even though the scoring is not much high, it represents that most respondents tends to retain in the same company.
To explore the current condition of overall job satisfaction for cabin crews in Taiwan airlines, the finding manifests ambivalent perception about job satisfaction. Additionally, even
if over half of cabin crews are not really satisfied with sub-variables, their overall retention intention exceeds half scoring. To explain this result, the author infers that there are other work-related or personal-work-related factors leading to resign or retain, or respondents are inclined to stay by comparing with other industries.
Table 4.2.
Descriptive Statistics on Each Dimension of JS and RI (N=226)
Min Max Median Mean SD
Pearson correlation and regression analysis was performed in order to determine the relationship between each dimension of job satisfaction and retention intention. The correlation finding (see table 4.3) showed that pay [r (224) = .396, p < .01], supervision [r (224) = .380, p
< .01], fringe benefit [r (224) = .394, p < .001], contingent rewards [r (224) = .376, p < .01], operating procedures [r (224) = .332, p < .01], coworkers [r (224) = .472, p < .01], nature of work [r (224) = .615, p < .01], communication [r (224) = .383, p < .01] and total job satisfaction [r (224) = .557, p < .01] have significant positive correlation with retention intention. Among them, pay, supervision, fringe benefits, contingent rewards, operation conditions and communication are modestly correlated with retention intention, whereas coworkers, nature of work and total job satisfaction showed moderate correlation with retention intention. On the other hand, there is no significant correlation between promotion [r (224) = .096, p = .149] and retention intention, which means no linear relationship between them.
To interpret reasonable possibility resulting in this consequence, the promotion system would be one of the affecting factors. Since the promotion system is diverse from different airlines, the perception of being promoted would be distinct from cabin crews. Therefore, it is
hard to be consistent with thought in terms of promotion to build the correlation between promotion and retention intention.
Table 4.3.
Correlation Matrix of Job Satisfaction and Retention Intention
1 2 3 4 5 6 7 8 9 10
1. Pay -
2. Promotion .316*** -
3. Supervision .547*** .339*** -
4. Fringe Benefits .684*** .248*** .495*** -
5. CR .686*** .416*** .669*** .625*** -
6. OP .355*** .026 .430*** .421*** .450*** -
7. Coworkers .269*** -.027 .511*** .248*** .342*** .335*** -
8. NW .343*** .164** .430*** .315*** .418*** .339*** .479*** -
9. Communication .411*** .258*** .530*** .431*** .558*** .453*** .375*** .406*** -
10. TJS .765*** .466*** .806*** .737*** .851*** .597*** .559*** .630*** .715*** - 11. RI .396*** .096 .380*** .394*** .376*** .332*** .472*** .615*** .383*** .557***
Note. ** p < .01, *** p < .001; CR= Contingent Rewards; OP= Operating Procedures; NW= Nature of Work; TJS= Total Job Satisfaction; RI=
Retention Intention.
The result of regression model, there is a significantly positive relationship between fringe benefit (β = .145, p < .05), coworkers (β = .203, p < .01), nature of work (β = .444, p < .001) and retention intention separately, representing that fringe benefit, coworkers, and nature of work could predict or explain retention intention towards cabin crews. On the contrary, there is no significant relationship between pay (β = .136, p = .079), promotion (β = -.021, p =.715), supervision (β = -.059, p = .446), contingent rewards (β = -.066, p = .439), operating procedures (β = .029, p = .637), communication (β = .069, p = .291) and retention intention respectively.
Table 4.4.
Linear Regression for Each Dimension of JS and RI
Retention Intention
Contingent Rewards -.066 .439
Operating Procedures .029 .637
Coworkers .203 .002 intention by controlling demographic variables (age, education level, position and tenure) based on the research question, hierarchical regression analysis was employed and the result presented as table 4.5.
The consequence displayed that age, education level, position and tenure could explain the variance of retention intention for 16.4% (F = 10.847, p < .001) whereas controlling the demographic variables, job satisfaction could increase 32.1% variance of retention intention (F
= 41.431; p < .001). With controlling the demographic variables, there is a significant relationship between job satisfaction and retention intention (β = .567; p < .001), that is, the scoring of job satisfaction is higher, the possibility of intention to retain is much higher.
Table 4.5.
Result of Hierarchical Regression Analysis for JS on RI
Retention Intention
Model 1 (β) Model 2 (β)
Step 1
Age .167* .210*
Education Level -.039 -.051
Position -.005 -.008
Tenure .247* .214*
Step 2
Job Satisfaction .567***
R2 .164 .485
Adj R2 .149 .473
△R2 .164 .321
F 10.847*** 41.431***
△F 10.847*** 137.060***
Note. * p < .05. *** p < .001.
Interaction Effects of Demographic Variables between JS and RI
Based on the research purpose, hierarchical regression analysis was performed to answer the research question, “is there significant interaction effect of demographic variables, age, education level, position and tenure, on the relationship between job satisfaction and retention intention?”. The following sections would describe the result of interaction effect of each demographic variable in detail.
Age, Job Satisfaction and Retention Intention
Table 4.6 showed that the main effect of age and job satisfaction can explain 46.3%
variance of retention intention (F(2, 223) = 96.293, p < .001), whereas after controlling main effect, the interaction between age and job satisfaction can increase 0.4% variance of retention intention (F(1, 222) = 1.838, p = .177). However, there is no significant interaction effect between age and job satisfaction on retention intention.
Table 4.6.
Result of Hierarchical Regression Analysis for Interaction between Age and JS on RI Retention Intention
Age x Job Satisfaction -.162
Total R2 .468***
Note. *** p < .001.
Education Level, Job Satisfaction and Retention Intention
Table 4.7 showed that there is no significant interaction effect between education level and job satisfaction on retention intention (F(1, 222) = .075, p = .784). However, the main effect of education level and job satisfaction can explain 32.3% variance of retention intention (F(2, 223) = 53.117, p < .001).
Table 4.7.
Result of Hierarchical Regression Analysis for Interaction between EL and JS on RI Retention Intention
Education Level x Job Satisfaction .086
Total R2 .323***
Note. * p < .05. *** p < .001.
Position, Job Satisfaction and Retention Intention
To measure the effect of position and job satisfaction on retention intention, table 4.8 presented that there is no significant interaction effect between position and job satisfaction on retention intention (F(1, 222) = .171, p = .679). Nevertheless, the main effect of position and
job satisfaction can significantly explain 36.4% variance of retention intention (F(2, 223) = 63.897, p < .001).
Table 4.8.
Result of Hierarchical Regression Analysis for Interaction between Position and JS on RI Retention Intention
Position x Job Satisfaction -.081
Total R2 .365***
Note. ** p < .01. *** p < .001.
Tenure, Job satisfaction and Retention Intention
To analyze the effect of tenure and job satisfaction on retention intention, table 4.9 presented that there is no significant interaction effect between tenure and job satisfaction on retention intention (F(1, 222) = 3.103, p = .080). However, the main effect of tenure and job satisfaction can significantly explain 47.5% variance of retention intention (F(2, 223) = 101.014, p < .001).
Table 4.9.
Result of Hierarchical Regression Analysis for Interaction between Tenure and JS on RI Retention Intention
Tenure x Job Satisfaction -.199
Total R2 .483***
Note. *** p < .001.
Summary of Findings and Discussions
Through previous analysis approaches, descriptive statistics, Pearson correlation, linear and hierarchical regression, the hypotheses testing results summary listed as Table 4.10. As performing Pearson correlation, the research showed that exclude promotion, pay, supervision, fringe benefit, contingent rewards, operating procedures, coworkers, nature of work, communication and total job satisfaction has significant positive correlation with retention intention. Moreover, linear regression analysis revealed that there is a significant relationship between fringe benefit and retention intention, coworkers and retention intention, nature of work and retention intention. Furthermore, through hierarchical regression analysis, the result showed that there is a significant relationship between job satisfaction and retention intention by controlling demographic variables, but there is no significant interaction effect between demographic variables (age, education level, position, and tenure) and job satisfaction on retention intention.
Table 4.10.
Hypotheses Testing Results Summary
Hypothesis Result
H1 Job satisfaction has positive influence on retention intention. Supported H1-1 Pay has positive influence on retention intention. Supported H1-2 Promotion has positive influence on retention intention. Not
Supported H1-3 Supervision has positive influence on retention intention. Supported H1-4 Fringe benefits have positive influence on retention intention. Supported H1-5 Contingent rewards have positive influence on retention intention. Supported H1-6 Operating procedures have positive influence on retention intention. Supported H1-7 Coworkers have positive influence on retention intention. Supported H1-8 Nature of work has positive influence on retention intention. Supported H1-9 Communication has positive influence on retention intention. Supported
H2 Age has a significant interaction effect on the relationship between job satisfaction and retention intention.
Not Supported H3 Education level has a significant interaction effect on the
relationship between job satisfaction and retention intention.
Not Supported H4 Position has a significant interaction effect on the relationship
between job satisfaction and retention intention.
Not Supported H5 Tenure has a significant interaction effect on the relationship
between job satisfaction and retention intention.
Not Supported Job satisfaction was considered as one of the most important factors affecting retention intention in previous researches (Enu-Kwesi et al., 2014; Mitchell & Albright, 1972). As for the relationship between job satisfaction and retention intention, the result showed that hypothesis 1 was supported and was consistent with previous finding (Enu-Kwesi et al., 2014;
Kavitha et al., 2011). Moreover, the research found that job satisfaction has positive influence on retention intention by controlling age, education level, position, and tenure as well, that is, as cabin crews more satisfied with their job, the higher intention to retain in airlines.
As for the relationship between 9 dimensions of job satisfaction and retention intention,
pay, supervision, fringe benefit, contingent rewards, operating procedures, coworkers, nature of work, communication proved hypothesis 1-1 to 1-9 except 1-2 that each sub-variable has positive influence on retention intention. Among them, fringe benefit, coworkers, and nature of work has significant relationship with retention intention separately, representing that as better fringe benefit, interaction and cooperation among coworkers, and nature of work, the higher intention to stay towards cabin crew which was in accordance with previous researches (Chen & Lai, 2017; Dutta & Banerjee, 2014; Gaiduk & Gaiduk, 2009; Sinha & Sinha, 2012;
Steijn & Leisink, 2006). However, hypothesis 1-2 does not support, that is, promotion does not have significant relationship with retention intention. Comparing with previous findings (Baloch et al., 2014; Cotton & Tuttle, 1986; Huang et al., 2006; Khan & Aleem, 2014; Lazear, 1998), there is no consistent result of the relationship between promotion and retention or turnover intention.
Subject to previous findings of the relationship between each demographic variable and retention intention, the result showed the same consequence. People with larger age (Choong et al., 2013; Cotton & Tuttle, 1986; Govaerts et al., 2011), higher education level, managerial position (Emiroğlu et al., 2015), and higher seniority (Emiroğlu et al., 2015) would tend to retain in the same organization. However, as for the interaction effect between each demographic variable and job satisfaction on retention intention, hypothesis 2 to 5 are not supported, meaning that the relationship between job satisfaction and retention intention would not be influenced by age, education level, position, and tenure.