5 Tips to Improve Data Integrity in Online Surveys

August 27, 2018

When it comes to market research nothing is more important than data integrity. In the days where phone surveys were the most common form of market research, this was less of an issue. Phone conversations added an additional layer of data quality because the interviewer managed and directed the conversation.

 

The data collection methodology with online surveys is more passive. There is no interviewer who manages the conversation and the market research firm is at the mercy of how much or how little a respondent is willing to type. This creates additional time for the market research analyst to review information on the back-end to understand the validity of responses.

 

It is important to note to use multiple data quality checks when reviewing your data. No single tip below will address all issues. All need to be taken into account when running your analysis.

 

 

Following data integrity protocols is very important in your online survey. It ensures the responses covered in your final analysis and report are valid and trustworthy.

 

 

Tip 1: Drop Cookies

The most advanced online survey platforms allow your survey team to drop a cookie on a respondent computer, tablet, or phone. This cookie prevents the survey taker from completing the questionnaire a second time. If the cookie is dropped, the respondent will click the survey link and a page will typically show stating: "You have already completed this survey".

 

Smart survey takers know how to avoid this by cleaning their cache and cookies but will they spend the time to do it? A fair percentage of respondents likely don't even know how to do this so this cookie-dropping is an extra layer of data integrity protection for your online survey.

 

 

Tip 2: Check IP Addresses

Similar to advanced cookies, survey platforms can also track IP address from the location a response is submitted from. All IP addresses are tied back to a physical address so it is easy to find multiple submissions from a single IP address. An example IP address typically looks something like this: 216.3.128.12

 

Be careful with B2B market research where organizations share the same IP address. If your survey collects data from multiple employees at one organization, cases may be valid. It is important to review all of these data integrity topics when reviewing your responses and not just a single factor.

 

 

Tip 3: If You Collect Contacts, Check for Duplicates

A simple check with your data analysis is to review duplicate cases. Contact information is often collected at the end of a survey to enter a sweepstakes or to participate in follow-up market research. It is important to review these names, emails, or phone numbers for duplicates. Identify these in your final review and remove these duplicates.

 

This can easily be tackled through a duplication formula or duplication tool in Excel. The program allows your team to easily remove duplicate rows from a data file.

 

 

Tip 4: Check for Speeders

Another excellent way to check for data integrity is to review speeders. Survey platforms often log time to complete for the respondent. This is the exact time it took to complete the survey from start to finish. This is an example of paradata in market research.

 

If you know your survey takes 7 minutes to complete and you have several responses where a respondent completed it 2 minutes, this is a red flag for data integrity. If the time to complete is borderline, review some of the other factors and the open-ended comments to see how much depth and detail was offered.

 

It is to easy to simply remove all cases submitted in under 2 minutes. Perhaps this respondent skipped some questions but offered some excellent partial responses you want to keep. Or maybe the respondent disqualified. Use time to complete or length of interview (LOI) as a guide to help you identify issues with data integrity.

 

 

Tip 5: Drop a Red Herring Question

A red herring question is an another suggestion for your quality checks. Unlike IP address, time to complete, and cookies, this is a purposeful question placed in the survey to identify bad responses. These questions help identify and eliminate speeders by the market research company.

 

An example of a red herring question is asking a respondent to select option C, choice A, or selecting the rating of 7 for a specific question. Those who do not can be eliminated from the study because they were not reading the question close enough to answer correctly. This can be a quick filter check or run in your reporting.

 

 

Contact Our Market Research Firm

Drive Research is a market research firm located in Syracuse, NY. Our team specializes in both qualitative and quantitative market research projects with clients across the country. This includes online surveys, focus groups, and in-depth interviews (IDIs).

 

Contact us below.

 

① Email us at info@driveresearch.com

Contact us on our website

③ Call us at 315-303-2040