Problem Definition
Turnover in engineering is a considerable bother for managers due to considerable implications of the matter. High turnover rates are associated with increased expenses on recruitment and training of new personnel. Moreover, high turnover rates lead to a necessity of keeping a sizeable human relation (HR) department, which is also associated with increased costs.
When speaking about the price paid for retention, it is also vital to consider that after recruitment, the efficiency of new employees immediately after hiring is significantly lower than the performance of their predecessors. New employees need time to adapt to the new working environment and acquire knowledge about the norms and business processes of the company. Moreover, the firm pays for the losses in productivity associated with the time a position stays unoccupied. Therefore, the prevention of high turnover rates is of extreme importance for HR managers to decrease associated costs.
The consequences of low retention rates are not limited by financial implications. The matter may also be associated with nonpecuniary costs, such as low employee morale and decreased workplace satisfaction. Therefore, the primary task of management personnel in engineering is to employ efficient strategies for improving retention rates. The present paper argues that one of the most efficient strategies for addressing the problem of high turnover rates is to use predictive hiring techniques by screening candidates for possible risk factors. However, HR managers are often unaware of risk factors of low retention intentions specific to engineering. Therefore, the purpose of the present paper is to introduce data-driven criteria for predictive hiring in engineering.
Methodologies
Variables
In order to arrive at relevant conclusions, it is vital to identify the variables that should be analyzed. The dependent variable for the present project is retention intention, which was quantified by the number of years in the current position. The self-reported average salary in the current position was assumed to influence loyalty to the company. The number of children was identified as an independent variable that may influence retention intentions due to decreased risk tolerance associated with having children. Self-reported GPA in college was also considered to have positive correlations with the dependent variable because it was assumed that a higher GPA is associated with constancy and a sense of purpose. Gender was also understood as a modifier for lower turnover intentions.
Sample and Data Collection
In order to answer the research question, a sample of 50 random engineering employees in one organization was surveyed online. The participants were given a link to the survey, and they had to complete it within one week after the link was issued. The surveys did not include any personal information, such as names, addresses, and phone numbers, to ensure the anonymity of the survey. The data was stored on a private personal computer protected by a password. Descriptive statistics of the sample is demonstrated in Tables 1 and 2 below.
Table 1. Sample’s Descriptive Statistics.
Table 2. Retention by Gender.
Identified Test
Three types of tests were performed using the collected data. Since the aim of the study is to determine if there are any interrelationships between the variables, correlation analysis was performed. Pearson’s correlation coefficient is a measure of the linear correlation between two variables. Moreover, multiple regression was used to identify which variables have an impact on the dependent variable and confirm the results of the correlation analysis. Finally, a two-sample t-test was used to compare the means of retention intentions in females and males. The latest update of Minitab-19 was used to analyze the data. The test was designed to test the following hypotheses.
- Hypothesis 1. There is a positive linear correlation between retention intentions and self-reported average annual salary.
- Hypothesis 2. There is a positive linear correlation between retention intentions and the number of children a person has.
- Hypothesis 3. There is a positive linear correlation between retention intentions and the college GPA of the candidate.
- Hypothesis 4. Female candidates have higher retention intentions than male candidates.
Data Statistical Analysis
Correlation Analysis
Correlation analysis was performed to quantify the interdependencies between all the variables, excluding gender. The matrix plot for the test with quantified results is demonstrated in Figure 1 below. The results show that there is a positive linear correlation between retention and salary and between retention and the number of children. However, there was no statistically significant connection between retention and GPA. Moreover, the results demonstrated that there is a medium correlation between salary and the number of children.
Regression Analyses
Retention versus Salary
The results for regression analysis for retention versus salary are demonstrated in Figure 2. Adjusted R2 value suggests that 60.19% of changes in retention intentions are attributed to increases in salary. In other words, the test confirms the results of the correlation analysis in the previous section.
Retention versus Number of Children
The results for regression analysis for retention versus the number of children is summarized in Figure 3. The results can be interpreted that 52.78% of changes in the number of years in the company is attributed to the changes in the number of children. The results support the idea suggested by correlation analysis.
Retention versus GPA
The results of the regression analysis for retention versus GPA also confirm the results of correlation analysis. The test demonstrates that 0% of changes in retention are attributed to changes in GPA. The output of Minitab-19 is presented in Figure 4 below.
Two-Sample T-Test
In order to identify the differences in means between the male and female populations in terms of retention intentions, a two-sample t-test with unequal variances was performed with a 95% confidence interval. The results demonstrate that there is a statistically significant difference between the two samples (p=0.006). The difference in means was estimated to be 2.3, which means that on average, women work in one company for 2.3 years longer than men do.
Results and Discussion
The results of the analyses described in the present paper support three out of four hypotheses concerning the retention intentions among engineering employees. First, the findings suggest that higher salaries are associated with lower retention intentions in the analyzed sample. Second, the results demonstrate that the number of children a person has influences his or her decision to change jobs. Third, females are less likely to leave a firm than men. Finally, Hypothesis 3 was not confirmed, meaning that there is no correlation between GPA and the number of years a person worked in the company. These results can be transformed into criteria for HR managers to utilize predictive hiring.
When recruiting engineering personal, HR managers may consider asking for the number of children a person has and avoid considering GPA. A high college GPA may be misleading since the statistical analysis shows that it does not influence retention intentions and employee performance (see Figure 1 for correlations between GPA and salary). At the same time, the number of children seems to be correlated with decreased risk tolerance, which may be associated with decreased turnover intentions.
At the same time, the findings suggest that females are more likely to be loyal to the company. Even though it may be unethical for gender to be a reason for hiring, the findings can be used by HR managers for targeting interventions for the reduction of turnover rates. In other words, HR managers should consider retention interventions to be focused on men since they are more likely to job. HR managers may also choose to target employees with lower salaries because the findings suggest that people with lower annual income are more likely to look for another job.
While the validity and reliability of methods are high, there are several limitations to the findings. First, the sample size is limited to 50 representatives of one engineering company, which negatively affects the generalizability of findings. Second, the number of identified variables may be a drawback, since the identified correlations may be indirect. For instance, the number of children usually depends on a person’s age.
However, age was not used as a control variable to adjust the findings. Third, the paper fails to review literature in the field and identify its place in current research. The research results may have no contribution to current knowledge since other researchers may have suggested the findings. In short, the study has considerable limitations, and its results should be used with caution.
Future research should focus on addressing the week parts of the present paper. Generalizability can be improved by increasing the sample size and diversifying sample population in terms of geographical and demographical characteristics. The same design can be used for replicated studies applied to different organizations. Other independent variables should be identified and tested to find other direct and indirect correlations with retention intentions. Finally, the same study can be applied to employees from other industries to determine if HR managers dealing with other professions can use the same criteria.
Conclusion
Employee turnover in engineering is a considerable bother for managers in the industry since it negatively affects the performance of the company. Predictive hiring is one of the methods for addressing the problem. The present paper analyzed four criteria that can be used by HR managers to decrease turnover rates by selecting candidates with increased retention intentions. It has been determined that the number of children can lower the chance of quitting a job, while GPA has no effect on the intention to switch jobs. Moreover, the findings suggest that females are more loyal to the company than males, and employees with higher salaries are less likely to look for another job. While these two findings can find only limited use in predictive hiring, this information can be used by HR managers for targeted interventions.