Speeder Threshold and Survey Data Quality Calculator
Set speeder thresholds from survey completion times, see how many completes get flagged, and work out how much extra sample to field to cover quality removals.
inquiry@globalvoxpopuli.com · globalvoxpopuli.com
Why remove speeders?
Speeders finish a survey much faster than is possible while reading the questions. Their answers are mostly noise, and in online panels they can include bots and professional survey-takers. Removing them, along with straight-liners and people who fail attention checks, is a standard part of online data quality control.
How to use the calculator
- Paste the completion times of your completes (in minutes or seconds), or enter just the median.
- Choose a speeder rule. Flagging completes under 30–50% of the median is common practice; 40% is a balanced default.
- Enter the removal rates you expect from your other checks.
- Read how many completes are flagged and how many you need to field to end with your target of usable completes.
Choosing a speeder threshold
Use the median, not the mean: a few people who leave the survey open for hours drag the mean up. A stricter rule (50% of median) removes more genuine fast readers; a looser one (30%) keeps more noise. The table in the results shows how many completes each rule would flag in your data, so you can pick deliberately. Check the flagged cases' open-ends before removing them: if those are thoughtful, the rule is too strict.
Planning for quality removals
If you need 1,000 usable completes and expect 5% speeders, 3% straight-liners, 4% attention-check failures and 2% bad open-ends, you need to field about 1,155. Build this into the quote and quota targets from the start rather than topping up at the end of fieldwork.
Frequently asked questions
What counts as a speeder in a survey?
Usually anyone completing in less than a third to a half of the median time. Some studies use a fixed minimum based on the time needed to read the questions.
What is straight-lining?
Giving the same answer to every statement in a grid, regardless of content. Flag it when a respondent straight-lines several long grids, especially grids with reversed statements.
What removal rate is normal?
For well-managed online panels, total removals of 5–15% are common. Rates above 20% point to a sample source or questionnaire problem.
Need the respondents, not just the numbers?
Global Vox Populi runs quantitative and qualitative fieldwork in 170+ countries through its own proprietary panels of consumers, B2B and IT decision-makers, physicians, nurses, patients, caregivers and payers. ISO 9001, ISO 20252 and ISO/IEC 27001 certified and HIPAA compliant, following ESOMAR and Insights Association guidelines.
Check feasibilityLink to this tool
Free to use in proposals, courses and articles. If it helped, a link back is appreciated:
<a href="https://globalvoxpopuli.com/tools/speeder-data-quality-calculator/">Speeder & Data Quality Calculator</a> by Global Vox Populi