**3. Methodology**

The data for this study was drawn in 2017 from a convenience sample of all frontline federal employees working as airport baggage and security screeners for the Transportation Security Administration (TSA) in the state of Oregon. The survey population were employed at the same occupational classification level and served on the front line of the agency. As frontline employees, their work required direct contact with citizens during their daily work. To stay abreast of the latest rules and regulations that governed their work, all employees were required to undergo monthly recertification training sessions. Agency officials integrated the survey instrument

into one of these required training sessions. The employees were provided with a workspace and instructions on how to complete the survey. The survey instructed the employees that their participation in the study was completely voluntary; their individual answers would be kept confidential; they could refuse to answer any question that made them uncomfortable; and they could end the survey at any time with no penalty or loss. Five hundred and fifty-seven (N = 557) useable surveys were collected with a response rate of 97%. The study sample was representative of the study population in terms of age, gender, and full-time status.

Several major variables were collected in this study. See **Table 1** for a description of the variables and coding strategies. PSM was collected using the Kim [78] 12-item revision of the Perry [79] 24-item PSM scale, which had good internal validity (Cronbach's α = 0.843). Job, organizational, environmental stressors were all collected using multi-survey questions and summed. Work-related stress is assumed to increase as the experience that the respondents have in the job, organizational, and environmental stressors increases. These scales were found to have good internal/composite validity (Cronbach's α = 0.740, 0.874, 0.822, respectively). For the sake of parsimony, turnover intentions and job satisfaction were collected using single-item survey questions. While multi-item scales are preferred in some cases, empirical evidence suggest that single-item survey questions can produce reliable and robust data [80].



*Does Stress Type Matter? Clarifying the Relationships between Public Service Motivation… DOI: http://dx.doi.org/10.5772/intechopen.112707*

#### **Table 1.**

*Description of study variables (N = 557).*

As shown in **Figure 1**, this study investigated the relationships between PSM, three work-related stressors, job satisfaction, and turnover intentions. The analysis of this study was conducted in three stages. First, a bivariate correlation analysis was conducted to confirm the relationships among the study variables. Second, a structural equation modeling (SEM) in AMOS was used to explore the relationships among the study variables. Third, bootstrap analysis (2000 samples) was used to test the significances of potential mediation effects. Statistical significance was set at 0.05, two-tailed. All regression weights are standardized maximum likelihood estimates, unless otherwise noted.
