An investigation into low mail survey response rates of information technology users in health care organizations

Document Type


Date of Original Version



Surveys conducted by mail can result in a very low response rate raising the possibility of non-response bias that would question the validity of the findings. In this study, we compare the demographic profiles between the respondents and non-respondents, and conduct wave analysis as an alternative way to estimate the non-response bias among the information technology (IT) users in the health care industry. Our findings indicate that there is no significant difference among the respondents when considering the duration (early versus late) of response rate. Likewise, there was no significant difference when the available demographic characteristics (facility type and user's position) of those who responded were compared to the same characteristics of the sample. We were able to reach similar conclusion by using either of the proposed methods; analysis of demographic profiles of respondents versus non-respondents, and wave analysis. Thus our findings support the validity of the results from low response mail survey studies for IT users. © 2003 Elsevier Ireland Ltd. All rights reserved.

Publication Title, e.g., Journal

International Journal of Medical Informatics