DESIGNING AND VALIDATING A PUBLIC HEALTH SURVEY: A MIXED METHODS PARADIGM

Abstract

Public health phenomena are usually complex and multifactorial, and public health surveys often have to take account of many aspects and factors. Designing and especially validating such a survey can therefore be a very challenging process. Asking appropriate questions, that will yield useful and data specific to the research questions, is of high importance. Equally important is to validate the survey instrument to ensure the trustworthiness of the obtained data. The aim of this presentation is to describe how a public health survey is designed, validated, and eventually implemented in a specific population.

Methods: For this purpose, we used a specific example of a mixed methods survey which was created and validated in 2016 to assess the knowledge and attitudes of public health professionals on the role of emotion in health and public health practice. A purposeful sample of 6 public health experts completed a survey instrument containing a new taxonomy of the emotions for use in the self-assessment of emotional status in a community and related questions to determine face validity and content validity. According to their feedback, the taxonomy and the survey were significantly revised. Then, a sample of approximately 100 participants completed the survey containing the face and content validated taxonomy, in order to test construct validity. Factor analysis was used for quantitative questions and theme analysis for qualitative questions. Further, a sample of approximately 300 participants was asked to complete the final instrument containing the validated taxonomy and related questions to determine reliability. Reliability was assessed calculating Cronbach’s alpha coefficient. Demographic data were also collected to investigate potential associations between these data and ‘emotional status’.

Results: An adequately validated and reliable survey instrument was produced to address specific research questions and to explore potential solutions to the problem under study. Specific steps will be described and discussed in this presentation on designing, validating and eventually implementing this survey.

Conclusions: With this presentation, the audience can become familiar with the process of designing and validating a survey instrument for public health purposes, and be able to note how a mixed methods approach can be applied to design a public health survey.