CPI Data Management & Analysis 3 — Questions and Answers
Question 1: A principal investigator receives a data request from a co-investigator at another institution. What must be in place before sharing identifiable research data?
- A verbal agreement between the investigators
- A signed Data Use Agreement (DUA) approved by both institutions (Correct answer)
- IRB approval from only the requesting institution
- A co-authorship agreement on future publications
Correct answer: A signed Data Use Agreement (DUA) approved by both institutions
A Data Use Agreement defines the terms under which identifiable data may be shared and is required by HIPAA and most institutional policies.
Question 2: Which type of validity refers to the extent to which a measurement instrument actually measures what it claims to measure?
- External validity
- Construct validity (Correct answer)
- Statistical conclusion validity
- Internal validity
Correct answer: Construct validity
Construct validity assesses whether a measurement tool accurately captures the theoretical construct it is intended to measure.
Question 3: A researcher uses a p-value threshold of 0.05 across 20 hypothesis tests simultaneously. What statistical issue arises?
- Underpowering of individual tests
- Inflated family-wise type I error rate (Correct answer)
- Increased risk of type II errors
- Violation of the normality assumption
Correct answer: Inflated family-wise type I error rate
Conducting multiple tests without adjustment inflates the probability of finding at least one false positive (type I error).
Question 4: According to GCP guidelines, how long must investigators retain essential clinical trial documents after study completion?
- 1 year
- At least 2 years after the last marketing approval or after a formal notification that no further marketing is pursued (Correct answer)
- 6 months
- 5 years from the date of publication
Correct answer: At least 2 years after the last marketing approval or after a formal notification that no further marketing is pursued
ICH E6 GCP requires retention for at least 2 years after the last marketing approval or study termination notification.
Question 5: What is the purpose of a range check in clinical data validation?
- To confirm participant demographics match population norms
- To flag values that fall outside predefined clinically plausible limits (Correct answer)
- To verify that all required fields have been completed
- To ensure randomization was performed correctly
Correct answer: To flag values that fall outside predefined clinically plausible limits
Range checks identify data entry errors by flagging values outside expected physiological or logical boundaries.
Question 6: In epidemiological research, what does 'attributable risk' measure?
- The ratio of disease risk in exposed vs. unexposed groups
- The absolute difference in disease rates between exposed and unexposed groups (Correct answer)
- The probability that an exposed individual will develop disease
- The proportion of cases in the population due to the exposure
Correct answer: The absolute difference in disease rates between exposed and unexposed groups
Attributable risk (risk difference) quantifies the excess risk in the exposed group compared to the unexposed group.
Question 7: A researcher wants to assess inter-rater reliability for a continuous measurement. Which statistic is most appropriate?
- Cohen's kappa
- Intraclass correlation coefficient (ICC) (Correct answer)
- Cronbach's alpha
- Phi coefficient
Correct answer: Intraclass correlation coefficient (ICC)
The ICC measures the degree of agreement and consistency between raters for continuous data, accounting for systematic biases.
CPI Data Management & Analysis 3
A principal investigator receives a data request from a co-investigator at another institution.
What must be in place before sharing identifiable research data?