Assisting children and youth with completing self-report instruments introduces bias: A mixed-method study that includes children and young people's views

Many children and youth struggle to complete surveys and questionnaires by themselves. They are often assisted when are asked to give their opinion. From discussions with youth from the client council at a residential treatment facility for youths in The Netherlands, the notion emerged that interaction factors may impact the results of surveys, especially when sensitive topics are addressed. Using a mixed methods design, we explored the question if and how survey results are influenced by the presence of an assistant during assessment. 120 children and youth that reside at one of the treatment facilities of Koraal, a Dutch multi-site care facility, completed a survey about the perceived quality of care at the facility. They were randomly assigned to one of three conditions: (a) unassisted, (b) assisted by their care worker, or (c) assisted by a research assistant. The resulting scores in each condition were compared quantitatively. In successive focus groups with children and youth, the results and possible explanations were discussed. Participants: in the Assisted by care worker condition exhibited significantly higher satisfaction scores than participants in each of the other two conditions. Results from the focus groups indicated that complex client-carer interaction dynamics contribute to these differences. Several explanatory mechanisms and implications for practice were suggested by the participants. These results suggest that bias may be introduced when children and youth need assistance to complete surveys or diagnostic measures. This requires careful consideration on the part of researchers working with these vulnerable participants.