Lean Six Sigma Black Belt Certification Measure Phase: Data Analysis 5 — Questions and Answers
Question 1: A Black Belt calculates a process sigma of 4.5 using long-term data. What is the equivalent short-term sigma level assuming the standard 1.5-sigma shift?
- 3.0
- 4.5
- 6.0 (Correct answer)
- 5.0
Correct answer: 6.0
Adding the 1.5-sigma shift to the long-term sigma of 4.5 gives a short-term sigma of 6.0, which represents the process's best-case capability.
Question 2: When should an EWMA (Exponentially Weighted Moving Average) chart be preferred over a standard Shewhart X-bar chart?
- When detecting large, sudden shifts quickly is the priority
- When the process produces small, sustained shifts that accumulate over time (Correct answer)
- When the subgroup size is greater than 10
- When data is attribute rather than continuous
Correct answer: When the process produces small, sustained shifts that accumulate over time
EWMA charts give more weight to recent observations and are more sensitive than Shewhart charts for detecting small, gradual process shifts.
Question 3: In hypothesis testing, what does 'power' (1 - beta) represent?
- The probability of a Type I error
- The probability of correctly rejecting a false null hypothesis (Correct answer)
- The confidence level of the test
- The p-value threshold chosen by the analyst
Correct answer: The probability of correctly rejecting a false null hypothesis
Power (1 - beta) is the probability that a statistical test correctly detects a real effect and rejects a false null hypothesis.
Question 4: A stem-and-leaf plot of cycle time data reveals two distinct peaks. What does this bimodal distribution most likely suggest?
- Random measurement error in the data
- Two different process streams or populations are mixed in the data (Correct answer)
- The data is normally distributed
- The process is in statistical control
Correct answer: Two different process streams or populations are mixed in the data
A bimodal distribution typically indicates that two distinct populations or process streams—such as two machines or shifts—are combined in the dataset.
Question 5: What is the primary risk of using a t-distribution instead of a Z-distribution when the sample size is large (n > 120)?
- The t-distribution produces much wider confidence intervals
- The results are virtually identical, so the risk is negligible (Correct answer)
- The t-distribution is invalid for large samples
- The t-distribution understates variance for large samples
Correct answer: The results are virtually identical, so the risk is negligible
For large sample sizes, the t-distribution converges to the Z-distribution, making the practical difference between conclusions negligible.
Question 6: Which graphical tool is most effective for detecting outliers and comparing distributions across multiple groups simultaneously?
- Histogram
- Run chart
- Box plot (box-and-whisker plot) (Correct answer)
- Scatter diagram
Correct answer: Box plot (box-and-whisker plot)
Box plots display the median, quartiles, and outliers for each group, making them ideal for comparing distributions and identifying outliers across multiple groups at once.
Question 7: In a Gauge R&R study, the 'reproducibility' component specifically measures variation due to:
- The gauge's inherent mechanical precision
- Differences between appraisers using the same gauge on the same parts (Correct answer)
- Differences between measurements taken by the same appraiser on the same part
- Environmental factors such as temperature and humidity
Correct answer: Differences between appraisers using the same gauge on the same parts
Reproducibility is the variation in measurements caused by different appraisers measuring the same part with the same gauge under the same conditions.
Lean Six Sigma Black Belt Certification Measure Phase: Data Analysis 5
A Black Belt calculates a process sigma of 4.5 using long-term data.
What is the equivalent short-term sigma level assuming the standard 1.5-sigma shift?