March 7, 2012
Respondent Validation
By Susan Frede, VP, Research
It’s important to take another look at respondent validation because marketing research clients are becoming increasingly strategic in their focus. A strategic focus requires more than data from discrete studies. It requires knowledge synthesis whereby data and findings from multiple studies and sources are combined. When taking this approach, using different survey measurement and quality practices can lead to differences in data that are hard to reconcile and synthesize. In addition, quality practices can impact consumer diversity. Therefore, it is important to employ data hygiene practices that meet clients’ need for high quality data, while maintaining the diversity of respondents.
At the recent CASRO Online Research Conference, I presented with Nallan Suresh from SurveyMonkey on TrueSample®, which Lightspeed Research has implemented as the quality standard on its US panel. Our research focuses on respondent validation. Current practices compare a survey-taker’s name and postal address to a database. When these respondents aren’t found in the database they are considered non-valid and are excluded from taking surveys.
Past TrueSample research on research has shown there are attitudinal differences between validated and non-validated respondents, which indicates non-validated respondents can impact data. However, validation rates are lower for key demographic groups, which leads to capacity constraints and possible data bias.
Our research explores a secondary validation process. The respondents who are found to be non-valid via the name/address validation check are taken through a secondary validation check using email address and a different database source (such as a social media account). If they pass, they are considered valid. If they fail, they remain non-valid.
This process successfully increased the validation rate for the key demographic groups while maintaining the differentiation between the validated and non-validated respondents. The secondary validation process provides our clients a consistent quality standard without dramatic losses to capacity.
Category:Data Quality, Panel Quality, Research on Research
Posted on March 7, 2012
Back to Home >>
Leave a Reply