Vocational Study and Public Service Motivation: Disentangling the Socializing Effects of Higher Education

ABSTRACT

Most studies of public service motivation investigate differences in motivation between public and private sector employees, but how these differences emerge and evolve in a pre-entry setting is still puzzling. Based on cross-sectional survey data with 3,521 Danish students enrolled in different vocational studies and at different stages (years) of their educational programs, this article investigates the socializing effects of higher education into different levels of public service motivation. The analysis demonstrates that students’ levels of public service motivation at different stages of their educational programs depend on the field of study: The level of public service motivation among students in vocational studies aimed at jobs with core public service delivery stays the same during education, whereas the level of public service motivation among students in other fields increases substantively with the length of their studies. This reveals that the association between education and public service motivation is perhaps not as uniform as previous studies have assumed.

Notes

^a These vocational studies are 5–6-year educational programs at Aarhus University, whereas the other vocational studies are 3–4-year educational programs at VIA University College Aarhus.

Note: No significant problems with multicollinearity were detected as all VIF are below 8 (tolerance indicators > 0.200) and the F-ratio change between Models 2.3 and 2.4 is significant (p < 0.000).

**p < 0.01; ***p < 0.001, standard errors in parentheses.

Note: Population data in parentheses (according to availability).

Note: PSM (no.) refers to the Perry (1996) items. All public service dimensions are reflective indexes (unweighed, theoretical range 0–100).

Different educational programs with similar content (e.g., building technicians and building engineers) have been collapsed into the same vocational study category. Architecture, journalism, psychology, and dentistry students were also surveyed but had to be excluded for the purpose of the present study as only first-year students from these educational programs participated in the survey. In addition, students who were still enrolled in the educational programs more than two years after they should have graduated (according to the standard duration time of the educational program) have also been excluded. Finally, the relatively low number of completes can also be subscribed to the length of the questionnaire (286 items).

Practical training/internships (approximately no more than eight months in total during the educational program) are a compulsory part of all the educational programs included in this study, except from law and theology.

The higher educational system in Denmark thus offer diplomas at two levels of increasing specialization, with the five- to six-year programs having the highest degree of specialization compared with the three- to four-year programs.

Root mean square error of approximation (RMSEA) score of 0.077, goodness-of-fit index (GFI) score of 0.967, and comparative fit index (CFI) and normed fit index (NFI) scores of 0.905 and 0.908, respectively (χ² = 1968.8, df = 84, p < 0.001). The generally accepted threshold value for a model with good fit is a GFI score > 0.95 and CFI and NFI scores > 0.90. The threshold value for the RMSEA index is somewhat more contested in the literature. Some scholars suggest values below 0.06 as demonstrating a good fit with the data (Hu and Bentler 1999); others argue that values between 0.05 and 0.08 still show a fair fit (Browne and Cudeck 1993; see also Jöreskog and Sörbom 1993; Schumacker and Lomax 2010).

The percentage of full-time employed working with core public service delivery (health, education, welfare/social services and security) in the Danish public sector (publicly owned organizations at the state, regional, or municipal level) is approximately 92% compared with 8% working with these services in the private sector (Statistics Denmark 2010). Hence, parents’ public/private sector employment is considered useful as a proxy for possible public service/non-public service socialization earlier in the students’ lives.

The students’ age has also been taken into consideration, but because of the variable “years of study,” age only measures at what time the students’ got enrolled (and the age range is very limited, with 85% of the students being below 30). A robustness test with age included in the analysis showed it is insignificant and does not change the main results; age is therefore left out.

Similar results have been found in additional analyses with each of the four PSM sub-dimensions as dependent variables. However, the interaction terms between years of study and core public service study were only significant (p < 0.05) for the compassion and self-sacrifice dimensions.

Absolute convergence (as indicated by the crossing lines) cannot be confirmed. A split file analysis between the students enrolled in the two different programs (five- to six-year programs compared with three- to four-year programs), however, verified the robustness of the results illustrated in Figure 1 for both types of programs.

Additional information

Notes on contributors

Anne Mette Kjeldsen

Anne Mette Kjeldsen ([email protected]) is an Assistant Professor at the Department of Political Science and Government, Aarhus University. Her primary research interests include public service motivation and management in the public and private sectors and especially in the areas of social services and health care.