Mother-Infant Studies: Subject Refusals and Sampling Bias

Publisher Website
Google Scholar
Abstract
Do sampling biases introduced in studies requiring a large amount of time make those samples less representative than the samples in studies requiring less subject involvement? This question was explored in the context of mother-infant interaction studies by eomparing maternal attitudes of two groups of mothers: 20 mothers who participated with their infants in 35 hours of observation and testing, and 17 mothers who refused to participate in that study, but did spend 30 minutes completing a questionnaire. These two groups of mothers differ widely in their willingness to be involved in a study. All subjects completed the Pregnancy Research Inventory at the end of the eighth month of pregnancy. There were no significant differences between the two groups on any scale. The importance of developing procedures to assess important biases introduced into studies requiring substantial commitments of time is discussed.