We design our projects to minimize the errors inherent in surveys that are often ignored or undetected and can lead clients to make wrong and costly decisions. Click here to read more about the types of errors inherent in surveys and what we do to minimize these.
Survey and Sample Design Quality Standards
The saying “garbage in, garbage out” very much applies to the survey instrument and sample design aspects of a research project. When the survey protocols (e.g., the questionnaire, interviewer guide) or the sample plan are not properly designed, regardless of the analysis or complex reports issued, the data collected are likely to lead to biased results. We take extra measures to assure that these two aspects of the research process are well designed and tested.
- Survey Design: We design the survey instruments with an eye on minimizing cognitive, visual, sequential, order, and other effects that lead to measurement error. In developing surveys, we conduct cognitive interviews, pre-test the questionnaires and also map the survey questions to the overall and specific objectives of a research project to be sure that the data generated will yield answers to the research questions.
- Sample Design and Weighting: Creating optimal sample designs beyond the standard simple random sample requires practical statistical expertise as well as academic training. Our survey research statistician has many years of practical experience designing samples for surveys and recommending sample designs to minimize sampling, coverage and non-response errors. As sample designs depart from simple random sampling, survey design effects and data weighting schemes are considered in our sample plan and data analyses.