Weighting Strategy in the Quebec Survey on the Experience of Health Care

• Activity Number: 444
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Marie-Eve Tremblay*+ and Robert Courtemanche
• Companies: Institut de la statistique du Quebec and Institut de la statistique du Quebec
• Keywords: weighting ; nonresponse adjustment ; paradata ; proxy respondents

Abstract

The Québec Survey on the Experience of Health Care is a telephone survey conducted between October 2010 and December 2011 among 48,100 respondents aged 15 and over living in a non-institutional dwelling in the province of Quebec. This paper focuses on the survey weighting strategy, which incorporates the use of paradata in the nonresponse adjustment. The idea was to take into account the difficulty in contacting each person from the sample, because this information is linked to the probability of response and to some of our main survey indicators, such as health condition. To achieve this goal, the number of phone calls was included in the logistic regression models used to create the weighting classes. Another distinctive feature of the nonresponse adjustment strategy was the idea of treating the people who were unable to respond to the questionnaire due to illness or disability differently from other nonrespondents. This type of nonresponse occurred when no proxy respondent could answer for the selected person. For that reason, it was decided to use only the proxy respondents to account for these nonrespondents.