The mode is the message : using predata as exclusion restrictions to evaluate survey design / by Heng Chen, Geoffrey Dunbar and Q. Rallye Shen. : FB3-5/2017-43E-PDF

“This paper addresses the following question: Do individuals respond differently to the same question when presented with different survey modes (offline vs. online)? We find that for the Bank of Canada's 2013 Methods-of-Payment Survey, there are mode effects for recall questions and for questions eliciting subjective opinions, e.g., whether cash payment method is easy to use. We find no differences for factual questions where the respondent could easily verify the information requested, e.g., the respondent's cash balance. Overall, our empirical results suggest caution when aggregating survey responses for mixed-mode surveys. In addition, our paper contributes to the statistical methods by applying the predata (information available prior to the specific survey design) to correct for both non-response and sampling bias. Such predata are easy and cheap to obtain from the survey marketing company, and provide enough variation to identify both response choice and sampling strategy, while not confounding the answers of the questionnaire"--Non-technical summary, p. [1].

Lien permanent pour cette publication :
publications.gc.ca/pub?id=9.847606&sl=1

• Format MARC XML
• Format MARC HTML

Demander des formats alternatifs