Page 40 - IJPS-11-4

International Journal of
Population Studies Stress in early adulthood

reliability and a less adequate approximation of perceived those responding in 2017. This percentage is comparable
stress levels compared to the 10-item scale (Eskildsen to official data from Statistics Denmark, which shows that
et al., 2015). Consequently, there is a possibility of non- 76% of 25 – 34 years olds had a job in 2017 (DST, 2020).
differentiated misclassification. In addition, the study may In this study, we used multiple imputation analyses,
be susceptible to healthy worker bias, as individuals with which enabled us to analyze the complete sample,
psychosocial demanding work environments might be potentially enhancing statistical power and reducing
underrepresented. This bias could arise because younger bias (Sterne et al., 2009). Information on the participants
workers facing high work demands or experiencing elevated responding to the questionnaires (without imputed data)
stress levels may have already left the labor market, leading can be seen in appendix (Tables A1 and A2).
to potential underestimation of the associations examined
in the study. Moreover, there is a risk that individuals with In addition, a complete case analysis containing 846
the most demanding work environments or highest stress respondents was performed that substantiated the results
levels did not choose to participate and, therefore, were not of the study (results not shown).
represented in the study population. When using self-reported information, there is a risk
When using questionnaire data to measure stress and work of participants’ over- or under-reporting, and a risk that a
environment, there is a risk of introducing negative affectivity negative response to one question concerning well-being
(Watson & Clark, 1984), where high-stress could influence may negatively affect the answers to other questions related
the way the individual perceives the working environment. to well-being. This potential bias, called common method
This may result in more negative reports on all aspects of the bias (Podsakoff et al., 2003), has, to some extent, been
work environment for those with high stress. However, as the minimized using both questionnaire and registered data and
correlation of these measures is low, we do not consider it to using prospectively collected data. However, there could be a
significantly affect the validity of the measures. potential problem related to the collection of information on
In this study, we attempt to capture work experience by the two stress outcome measures and the information about
measuring “time since educational completion.” However, temporary employment and psychosocial work environment
there is a risk that short work experience (<2 years) could aspects because they were collected at the same time. However,
be due to long tertiary education, which is why the results as there was a low correlation between these measures, this
must be interpreted with caution. does not seem to have caused any substantial information bias.
The socioeconomic characteristics of the West Jutland The limitations presented are not considered to result
Cohort study, such as income and parental educational in serious bias concerning the observed associations.
level, were comparable to those of families of young However, caution is advised when interpreting causality.
individuals of the same age in Denmark (Glasscock et al., 5. Conclusion
2013). Comparison of the 2406 non-responders to the
1275 responders on socioeconomic measures shows that This study reveals a significant association between stress
more non-responders (17%) than responders (13%) had levels during adolescence and stress experienced in
low-educated parents and had grown up in low-income early adulthood. From a public health perspective, it is
families (39% vs. 31%), and more non-responders had important for both occupational health-care professionals
low educational level (17%) compared to the responders and educators at secondary and tertiary educational
(4%). This selection may have underestimated the true institutions to recognize that stress during adolescence can
association. It is important to note that the external validity increase the risk of stress in early adult working life. The
and generalizability of the current study are confined fact that those who experience stress during adolescence
to individuals residing in Denmark and other countries reported high levels of general perceived stress at the
sharing similar welfare systems. beginning of work life underscores the need for nuanced
The response rate was 83% at the first collection of and tailored support for young individuals entering the
questionnaires, declining slightly to 57% at the collection workforce. Moreover, this study underlines the importance
in 2017. A previous study investigating non-participation of employers prioritizing psychosocial aspects of the work
across questionnaire collection rounds in the same cohort environment, such as promoting work-life balance. These
found that the selection of socioeconomic measures had factors can significantly influence levels of work-related
no significant impact on the validity of the measured risk stress among young adults.
estimates (Winding et al., 2014). The present study only Acknowledgments
included those who had indicated being employed at age
28 in the 2017 questionnaire, which corresponds to 71% of None.

Volume 11 Issue 4 (2025) 34 https://doi.org/10.36922/ijps.0864