Avoid Inferior Quality-Control Questions For Top-Notch Results
SSI Presents Best Quality-Control Questions for Reliable Research Data
SHELTON, Conn., Oct. 31, 2013 – On Oct. 31, the Australian, Market & Social Research Society’s Young Researchers Network will host a research event in Melbourne, Australia at 6 p.m. Ania Kubiak, Account Manager, will represent SSI and present solutions to ensure high-quality data for every project.
Respondents occasionally make mistakes when completing surveys, but when they fail poorly designed quality-control questions, it’s not their fault. Kubiak’s presentation will delve into some of the commonly-used tricks and highlight better options researchers can use to ensure better data quality.
One sub-standard quality-control questions used is “the blind.” The survey specifically asks respondents to “not answer a question,” or “type 3 here” as a quality control. Many honest respondents may respond to these trick questions by feeling hurt that their integrity is being questioned. SSI research shows that this technique screens out too many people who are otherwise giving good data. It’s counter-intuitive to not answer a question, or interrupt your concentration to give a meaningless answer.
Another inferior quality-control check is to hide a “nonsense” response in a list of reasonable responses. For example, asking respondents if they have done a number of activities, including “visiting Saturn.” This type of question may lead respondents to doubt the seriousness of the entire survey and thus lessen their commitment to it.
In another example, respondents are asked if they recognize a fake brand name which sounds similar to a real brand. This, along with the other trick questions usually results in exclusion of high-quality survey-takers.
No quality-control methods remove all the “bad” respondents and almost all the methods remove some of the “good” respondents. Poorly designed quality-control questions hurt feasibility and create a poor user experience. Researchers should keep in mind that any survey participant can become disengaged in the moment and fail a single quality-control question so participants flagged for removal should fail multiple quality-control measurements. Kubiak’s presentation shares SSI’s recommendations for quality-control best practices. SSI suggests using a combination of six checks to identify problem respondents – including conflicting statement checks, speed and other checks.
For more information on these quality-control methods or other survey safeguards ensuring data quality on your next research project, email SSI at firstname.lastname@example.org.
SSI is the premier global provider of sampling, data collection and data analytic solutions for survey research, reaching respondents in 78 countries via Internet, telephone, mobile/wireless and mixed-access offerings. Additionally, SSI staff operates from 26 offices in 18 countries, offering CATI, questionnaire design consultation, programming and hosting, online custom reporting and data processing. SSI’s 3,100 employees serve more than 3,000 clients worldwide. For additional information, please visit www.surveysampling.com.