PRC Data Quality & Validation 1 — Questions and Answers
Question 1: Which data quality check is specifically designed to detect respondents who provide the same answer to every question in a survey?
- Speeder detection
- Straight-lining detection (Correct answer)
- Duplicate respondent screening
- Open-end verbatim review
Correct answer: Straight-lining detection
Straight-lining detection identifies respondents who select the same scale point for all items in a grid or series, indicating inattentive or fraudulent responding.
Question 2: A research firm receives 1,200 completed surveys but 80 are flagged because respondents finished a 15-minute questionnaire in under 3 minutes. This quality check is called:
- Trap question analysis
- Speeder flagging (Correct answer)
- Break-off analysis
- Quota balancing
Correct answer: Speeder flagging
Speeder flagging identifies respondents who completed the survey far too quickly to have read and considered the questions thoughtfully.
Question 3: In the context of data validation, an 'attention check' or 'trap question' is best described as:
- A question designed to confuse respondents into revealing their true opinions
- An item with a pre-specified correct answer used to verify that respondents are reading carefully (Correct answer)
- A filter question that routes respondents to different survey paths
- A reliability measure comparing responses across two survey halves
Correct answer: An item with a pre-specified correct answer used to verify that respondents are reading carefully
Attention check questions have a known correct or expected response (e.g., 'Please select Strongly Agree for this item') and are used to identify inattentive respondents.
Question 4: Which of the following is the primary purpose of back-checking (or back-translation verification) in data collection?
- Ensuring interviewers actually conducted interviews as reported (Correct answer)
- Converting open-ended responses into coded categories
- Verifying that translated survey instruments retain their original meaning
- Checking that data entry operators transferred responses accurately
Correct answer: Ensuring interviewers actually conducted interviews as reported
Back-checking involves re-contacting a sample of respondents to verify that interviewers conducted interviews properly and did not fabricate data.
Question 5: A dataset contains a respondent who reported their age as 250 years. This is an example of which type of data quality issue?
- Item non-response
- Out-of-range value (Correct answer)
- Logical inconsistency
- Response bias
Correct answer: Out-of-range value
An out-of-range value occurs when a data point falls outside the logically or empirically possible range for that variable (e.g., age cannot be 250).
Question 6: When a respondent states they are 22 years old but later indicates they have 30 years of work experience, the data file contains a:
- Out-of-range error
- Duplicate record
- Logical inconsistency (Correct answer)
- Speeder flag
Correct answer: Logical inconsistency
A logical inconsistency exists when two or more responses contradict each other in a way that is impossible or highly implausible (e.g., working more years than the respondent has been alive).
Question 7: Which data quality metric measures the percentage of survey questions that a respondent left blank?
- Item non-response rate (Correct answer)
- Break-off rate
- Completion rate
- Incidence rate
Correct answer: Item non-response rate
The item non-response rate calculates how often individual questions were skipped or left unanswered within an otherwise completed survey.
PRC Data Quality & Validation 1
Which data quality check is specifically designed to detect respondents who provide the same answer to every question in a survey?