CQIA Data Collection & Analysis Methods 5 — Questions and Answers
Question 1: What is a key limitation of using correlation analysis to establish a quality improvement action?
- Correlation can only be used with attribute data
- Correlation does not prove causation between variables (Correct answer)
- Correlation requires a minimum of 1,000 data points
- Correlation charts cannot be updated over time
Correct answer: Correlation does not prove causation between variables
Even a strong correlation coefficient does not establish that one variable causes changes in the other; additional investigation is needed.
Question 2: A hospital tracks patient wait times and plots them on a run chart. The chart shows a pattern alternating above and below the centerline every other point. This pattern is MOST likely caused by:
- A stable in-control process
- An alternating special cause such as two alternating staff shifts (Correct answer)
- Random common cause variation
- Incorrect control limit calculation
Correct answer: An alternating special cause such as two alternating staff shifts
A consistent zigzag alternating pattern suggests a systematic special cause like alternating between two different process conditions.
Question 3: Which of the following BEST describes the purpose of a data collection plan?
- To determine which statistical test to use after data is collected
- To define what data to collect, how, when, and by whom before collection begins (Correct answer)
- To summarize results after the improvement project is complete
- To select which team members will analyze the data
Correct answer: To define what data to collect, how, when, and by whom before collection begins
A data collection plan is developed before data gathering to ensure consistent, relevant, and complete data capture.
Question 4: In a process capability study, a Cpk value of 0.85 indicates:
- The process is highly capable and centered within specifications
- The process is not fully capable and may produce out-of-specification output (Correct answer)
- The process variation equals exactly the specification width
- Cpk below 1.0 means the process is perfectly centered
Correct answer: The process is not fully capable and may produce out-of-specification output
A Cpk below 1.0 indicates the process is not fully capable, meaning some output will fall outside specification limits.
Question 5: A quality improvement team wants to verify whether a change in supplier material caused an increase in defect rates. Which data analysis approach is MOST appropriate?
- Draw a new fishbone diagram with the supplier as a branch
- Compare defect rate control charts before and after the supplier change (Correct answer)
- Conduct a customer survey about product satisfaction
- Create an affinity diagram grouping supplier issues
Correct answer: Compare defect rate control charts before and after the supplier change
Overlaying or comparing before/after control charts allows the team to detect whether a shift in defect rate coincides with the supplier change.
Question 6: Which of the following BEST explains why sample size matters in data collection for quality studies?
- Larger samples always guarantee the process is in control
- Adequate sample size ensures statistical conclusions have sufficient confidence and power (Correct answer)
- Smaller samples are always more accurate because they reduce measurement error
- Sample size only affects the type of chart used, not the validity of results
Correct answer: Adequate sample size ensures statistical conclusions have sufficient confidence and power
An insufficient sample size can miss real process signals, while adequate sample size ensures conclusions are statistically reliable.
Question 7: A team uses a box plot to display inspection data. What does the length of the box in a box plot represent?
- The total range from minimum to maximum value
- The interquartile range containing the middle 50% of data (Correct answer)
- The standard deviation of the dataset
- The distance between the mean and the median
Correct answer: The interquartile range containing the middle 50% of data
The box in a box plot spans from the 25th to 75th percentile, representing the interquartile range (IQR) of the data.
CQIA Data Collection & Analysis Methods 5
What is a key limitation of using correlation analysis to establish a quality improvement action?