To ensure our online samples are representative, for our omnibus survey and our other surveys, we quota sample by a range of demographics. We then rim weight the results by gender, age, region, race/ethnicity, Hispanicity, and educational attainment. For social policy research, we weight by political party, ideology, registration status, and evangelicalism.
When weighting, we only include in the analysis those respondents who completed all the weighting questions: typically this includes the entire questionnaire that was shown to them (based on answer validation, branching, and skip patterns).
As these are not probability-based samples, calculating the theoretical margin of sampling error is not applicable. However, as with probability surveys, it is important to keep in mind that results are estimates and typically vary within a narrow range around the actual value that would be calculated by completing a census of everyone in a population. One estimate of this precision is the credibility interval; for these surveys, the credibility interval is plus or minus 4 percentage points for questions answered by all respondents (the interval is larger for questions answered by fewer respondents). Again, as with probability surveys, on occasion the results from a particular question will be completely outside a typical interval of error.
Many types of survey errors can limit the ability to generalize to a population. Throughout the research process, Researchscape followed a Total Survey Quality approach designed to minimize error at each stage. Total Survey Quality, also known as Total Survey Error, recognizes that multiple sources of error can reduce the validity of survey research. Besides sampling error, five types of non-sampling error can occur: specification error, frame error, nonresponse error, measurement error, and processing error. At each step in the survey research process, the research team followed best practices and used quality controls to minimize the impact of these sources of error.