Premium
How do we design and report a high‐quality survey?
Author(s) -
Pagano Monica B.,
Dunbar Nancy M.,
Stanworth Simon J.
Publication year - 2020
Publication title -
transfusion
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.045
H-Index - 132
eISSN - 1537-2995
pISSN - 0041-1132
DOI - 10.1111/trf.15861
Subject(s) - generalizability theory , quality (philosophy) , reliability (semiconductor) , relevance (law) , research design , data collection , survey research , psychology , medicine , computer science , applied psychology , political science , statistics , developmental psychology , philosophy , power (physics) , physics , mathematics , epistemology , quantum mechanics , law
Every day, new surveys are planned, distributed, or reported by health care professionals. Surveys are an inexpensive and convenient research tool used with increasing frequency as an approach to gather and collate information on attitudes and behaviors for a specific topic. However, surveys can squander the valuable time of respondents who may derive little, if any, benefit from participation. Similar to any other research methodology, a careful design is needed to avoid introducing bias and to obtain meaningful information. A recent study evaluating the quality of surveys addressing clinical topics in transfusion medicine (TM) identified common deficiencies in the quality and design, including the failure to report validity and reliability, to address nonresponse error, to report funding and ethics/consent considerations, and to discuss the generalizability of results. Instructions to authors for reporting survey results are lacking in most journals. Inadequate survey design, analysis, and reporting can prevent accurate data collection and compromise the interpretation of the results, which is of critical relevance considering the high citation rates for some of these surveys. Further, survey results might be used to inform policies when no higher level of evidence is available. In this article, the authors seek to provide practical recommendations for designing high‐quality surveys based on personal experience and published literature and to address frequently missing key elements in survey‐based studies related to clinical TM.