ABSTRACT
OBJECTIVE: To examine whether the method of measuring patient satisfaction influences the results. STUDY DESIGN: All families with NICU patients discharged home alive to parent(s) were surveyed at 2 to 42 days post discharge with two parallel surveys (mailed and phone) from August 2010 to December 2011. The responses to the same five questions on each survey were recorded and subjected to statistical analysis via paired Student's t and Pearson correlation coefficients to see whether the two surveys yielded the same population attitudes. RESULT: Eight hundred and thirteen families were included in the study. Seven hundred and sixty three (93.8%) completed the phone survey and 237 (29.2%) completed the mail in survey. Three of the five questions yielded significantly different answers between mailed and phone responses. In addition, no significant linear correlation between mail and survey could be found for the other two questions. CONCLUSION: As no linear correlation could be found between two of the five questions, a process constant mathematical value could not be identified. This indicates that mail survey and phone survey respondents have different attitudes that cannot be adjusted because of methodological measurement effects.
