%0 Journal Article
%T Illegitimate work tasks: an investigation of psychometric properties of the Swedish version of the BITS instrument and its suitability in human versus 'non-human' service occupations.
%A StengÄrd J
%A Leineweber C
%A Berthelsen H
%J BMC Public Health
%V 24
%N 1
%D 2024 Jul 18
%M 39026291
%F 4.135
%R 10.1186/s12889-024-19393-x
%X BACKGROUND: Illegitimate tasks, i.e. working tasks that are perceived as unnecessary or unreasonable, are commonly measured by the Bern Illegitimate Tasks Scale (BITS). Despite a growing research attention paid to illegitimate tasks, the Swedish version of BITS needs yet to be properly evaluated. Moreover, previous multiorganizational studies in this field have taken for granted, without really testing it, that the instrument functions invariantly in different contexts; a prerequisite for meaningful comparisons.
METHODS: Two occupational groups that differs hugely-966 human services workers (teachers and registered nurses) and 750 non-'human service' workers (construction and IT-workers) were targeted utilizing questionnaires data collected 2018 within the Swedish Longitudinal Occupational Survey of Health (SLOSH) study. Psychometric properties, i.e., reliability, dimensionality, and measurement invariance between the occupations were examined using confirmatory factor analyses and structural equation models. Also, the associations between the two dimensions of illegitimate tasks and job satisfaction and emotional exhaustion, respectively, were tested.
RESULTS: Good reliability was supported and two- versus one-dimensionality showed better psychometric properties. Partial scalar measurement invariance was satisfactory supported with only 25% relaxed constraints of equal intercepts between groups. Examining the two subdimensions revealed different associations with emotional exhaustion, where the associations was more substantial for unreasonable tasks. Nevertheless, the findings point to potential improvements for future investigation.
CONCLUSIONS: This study shows that the Swedish version of BITS is working satisfactory and allows for meaningful comparisons of occupational groups. Furthermore, construct validity of the two dimensions was corroborated.