The Assessment of the Reliability and Validity of the Polish Version of the Adult Vaccine Hesitancy Scale (PL-aVHS) in the Context of Attitudes toward COVID-19 Vaccination.

Vaccine hesitancy has become a pivotal consideration in assessing society's readiness to accept recommended vaccination programs. The role of vaccination as a preventive measure during great epidemic challenges cannot be overestimated. On the other hand, the overwhelming flow of misinformation and attitudes resulting from denialism may have a profoundly harmful effect on the acceptance of preventive interventions. The adult Vaccine Hesitancy Scale (aVHS) is a result of efforts to develop a tool that will be relevant to the views about vaccination in the general adult population. It was derived from the Vaccine Hesitancy Scale (VHS), initially developed by researchers attempting to assess the opinions and attitudes of parents. This study's main aim was to determine the reliability and validity of the Polish version of aVHS (PL-aVHS). We have also analyzed whether the scale can feasibly predict the COVID-19 vaccination status of respondents. The analysis was performed on data originating from a computer-based web-interviewing (CAWI) survey of 2008 adult Internet users. It included the analysis of internal consistency, test-retest reliability, and hypotheses testing. Exploratory (EFA) and confirmatory factor analyses (CFA) were performed on the subsets generated by randomly splitting the initial survey data. We have found that the scale has excellent internal consistency (Cronbach α = 0.935), acceptable levels of inter-item bivariate correlations, and good test-retest reliability (interclass correlation coefficient, ICC = 0.843). The EFA revealed that the tool has a two-factor latent structure; however, similar loadings of item 10 to both factors spoke for its exclusion from the model. Two extracted factors were responsible for 68.90% of the variance after rotation based on the maximum likelihood method. The CFA showed that the best fit of the model to measurement data was obtained for the two-factor model after excluding item 10. All seven fit indexes calculated in the analysis suggested at least an acceptable fit. In conclusion, the assessment of the PL-aVHS revealed good reliability and validity of the instrument. Furthermore, we have obtained similar EFA results as reported for the English version of the tool. Finally, to our knowledge, it is one of a few tools available in Polish for the measurement of vaccine-related attitudes.