• To avoid bias in analysis, interpretation and generalisability of findings that can occur from missing data when using self-administered questionnaires
• To develop strategies to prevent missing data in self-administered questionnaires
• To understand the methodological and statistical considerations underlying the use of prevention strategies in quantitative studies
Background Self-administered questionnaires are efficient and low-cost ways of collecting data with wide cohorts. Nonetheless, their use in studies can result in a high occurrence of missing data, which can affect the statistical power, representativeness and generalisability of the findings. Imputation methods have been considered efficient statistical techniques for managing missing data. However, they have also been associated with limits, such as the risk of under-estimation of the effect, lower statistical power and decrease of correlation among variables. Recent studies have highlighted the importance of using prevention strategies to avoid missing data before the data are analysed.
Aim To identify strategies for preventing the occurrence of missing data and to discuss their effects, as well as their methodological and statistical considerations.
Discussion The article discusses prevention strategies related to the administration format and follow-up and reminders. Strategies such as the use of electronic tablets, email and telephone reminders are associated with lower rates of missing data in self-administered questionnaires. However, methodological and statistical limits, including the absence of a comparison group and statistical validation of the reported results, limits the capacity to establish robust consensus.
Conclusion Prevention strategies represent relevant and feasible avenues for handling missing data in a wide range of clinical, nursing and epidemiological research. More projects based on robust design are needed to ensure accurate and reliable data are collected from patients, families, communities and clinicians.
Implications for practice It is important for clinicians and nurses to understand the phenomenon of missing data and the strategies available to prevent missing data, to collect data representing the patients’ and families’ perspectives and experiences.
Nurse Researcher. 30, 3, 9-18. doi: 10.7748/nr.2022.e1835Correspondence
This article has been subject to external double-blind peer review and checked for plagiarism using automated softwareConflict of interest
Li-Anne Audet holds doctoral scholarships from the Fonds de recherche du Québec – santé, as well as from the Ordre des infirmières et infirmiers du Québec and the Réseau de recherche en interventions en sciences infirmières du Québec. Michèle Desmarais holds doctoral scholarships from the Ordre des infirmières et infirmiers du Québec, the International Association for Human Caring, and the Réseau de recherche en interventions en sciences infirmières du Québec No funding sources were involved in the studyPermission
To reuse this article or for information about reprints and permissions, please contact firstname.lastname@example.orgWrite for us
For information about writing for RCNi journals, contact email@example.com
For author guidelines, go to rcni.com/write-for-nurse-researcher
Alternatively, you can purchase access to this article for the next seven days. Buy now