Lean Six Sigma Green Belt Certification DMAIC: Analyze Phase Techniques 5 — Questions and Answers
Question 1: A process improvement team wants to test whether the proportion of defective units differs between two production lines. Which test is most appropriate?
- Two-sample t-test
- Two-proportion z-test (Correct answer)
- One-way ANOVA
- Mann-Whitney U test
Correct answer: Two-proportion z-test
The two-proportion z-test compares proportions (attribute data) from two independent groups to determine if they are statistically different.
Question 2: When a team performs regression analysis and finds that adding a second X variable increases R² from 0.72 to 0.73, the BEST interpretation is:
- The second variable is highly significant and must be included
- The second variable adds minimal explanatory value and may not justify model complexity (Correct answer)
- The model is now significantly better and ready for the Improve phase
- R² should always be maximized regardless of the number of variables
Correct answer: The second variable adds minimal explanatory value and may not justify model complexity
A marginal R² increase suggests the second variable contributes little additional explanatory power while adding model complexity, so its inclusion should be questioned.
Question 3: In a Lean Six Sigma Analyze phase, what is the purpose of calculating takt time?
- To determine the upper control limit for a control chart
- To set the pace of production required to meet customer demand (Correct answer)
- To calculate process sigma level
- To identify the critical path in a project schedule
Correct answer: To set the pace of production required to meet customer demand
Takt time is calculated as available production time divided by customer demand, setting the required production pace to meet demand without overproduction.
Question 4: A team uses an affinity diagram during the Analyze phase. What is the PRIMARY output of this tool?
- A ranked Pareto chart of defect categories
- Ideas or data grouped into meaningful themes based on natural relationships (Correct answer)
- A regression equation relating X to Y
- A control plan for process monitoring
Correct answer: Ideas or data grouped into meaningful themes based on natural relationships
An affinity diagram organizes large amounts of qualitative data (ideas, observations) into related groups or themes through collaborative sorting.
Question 5: Which type of error occurs when a team concludes that a root cause IS significant when it actually has no real effect on the output?
- Type II error (Beta error)
- Type I error (Alpha error) (Correct answer)
- Sampling bias
- Measurement system error
Correct answer: Type I error (Alpha error)
A Type I error (false positive) occurs when the null hypothesis is incorrectly rejected, concluding a significant effect exists when it does not.
Question 6: A Green Belt discovers through data analysis that the process mean has shifted over time but process spread remains constant. Which graphical tool would BEST confirm this finding?
- Box plot comparing time periods
- I-MR chart showing a sustained shift in the individual values (Correct answer)
- Normal probability plot
- Multi-vari chart showing positional variation
Correct answer: I-MR chart showing a sustained shift in the individual values
An I-MR (Individuals and Moving Range) control chart reveals a sustained shift in the process mean while the Moving Range chart confirms spread is stable.
Question 7: A team identifies three potential root causes through analysis. Before moving to Improve, what is the MOST important validation step?
- Build a control plan for all three causes
- Conduct designed experiments or additional data analysis to confirm each cause's impact (Correct answer)
- Immediately pilot solutions for all three causes
- Update the project charter to list the root causes as confirmed
Correct answer: Conduct designed experiments or additional data analysis to confirm each cause's impact
Root causes must be statistically confirmed through data analysis or controlled experiments before moving to Improve, ensuring resources are directed at true causes.
A process improvement team wants to test whether the proportion of defective units differs between two production lines.
Which test is most appropriate?