How Good Is Survey Design in Medical Libraries? A Systematic Review of User Surveys
Introduction: Medical libraries very often base the decisions they make about library services on information gathered from user surveys. Is the quality of information obtained in this way sufficient to enable evidence-based practice? Aim: To determine what aspects of user survey design and presentation obtain the best response rates and therefore high external validity. Also to provide guidance for medical librarians who may wish to carry out user surveys. Methods: Library and information studies databases and Medline were searched to identify studies that reported the results of library user surveys that measured user perceptions of an existing library service or potential service. Studies that evaluated information skills training or clinical librarianship interventions were excluded as they have been looked in separate systematic reviews. Also studies that reported the results of LibQUAL or SERVQUAL were excluded. Results: 54 studies were included. The quality of the majority of the surveys was not clear as the reporting of the methodology of the user surveys was poor. However, it was determined that, as demonstrated in previous research, paper format surveys reported higher response rates than online-only surveys. It was not possible to extract any relevant data from the identified studies to draw any conclusions relating to presentation of the survey instrument. Conclusions: Unless survey methodology is reported in detail it is not possible to judge the quality of the evidence surveys contain. Good survey design is key to obtaining a good response rate and a good response rate means the results can be used for evidence-based practice. A Reporting Survey results Guideline (Resurge) is recommended to help improve the reporting quality of medical library survey research.