Does the sequence of data collection influence participants鈥?responses to closed and open-ended questions? A methodological study

详细信息	查看全文 | 推荐本文 |

• 作者：Christine L. Covell^a ; ^{ccovell@uottawa.ca} ; Souraya Sidani^b ; ^{ssidani@ryerson.ca} ; Judith A. Ritchie^c ; ^{judith.ritchie@muhc.mcgill.ca}
• 关键词：Interviews ; Mixed-methods ; Qualitative ; Quantitative ; Research design ; Survey
• 刊名：International Journal of Nursing Studies
• 出版年：2012
• 期刊代码：134_00207489
• 类别：med
• 出版时间：June, 2012
• 卷：49
• 期：6
• 页码：664-671
• 文件大小：203 K

摘要

Background

The sequence used for collecting quantitative and qualitative data in concurrent mixed-methods research may influence participants鈥?responses. Empirical evidence is needed to determine if the order of data collection in concurrent mixed methods research biases participants鈥?responses to closed and open-ended questions.

Objectives

To examine the influence of the quantitative-qualitative sequence on responses to closed and open-ended questions when assessing the same variables or aspects of a phenomenon simultaneously within the same study phase.

Design

A descriptive cross-sectional, concurrent mixed-methods design was used to collect quantitative (survey) and qualitative (interview) data. The setting was a large multi-site health care centre in Canada.

Methods

A convenience sample of 50 registered nurses was selected and participated in the study. Participants were randomly assigned to one of two sequences for data collection, quantitative-qualitative or qualitative-quantitative.

Analysis

Independent t-tests were performed to compare the two groups鈥?responses to the survey items. Directed content analysis was used to compare the participants鈥?responses to the interview questions.

Results

The sequence of data collection did not greatly affect the participants鈥?responses to the closed-ended questions (survey items) or the open-ended questions (interview questions).

Conclusions

The sequencing of data collection, when using both survey and semi-structured interviews, may not bias participants鈥?responses to closed or open-ended questions. Additional research is required to confirm these findings.