CSM Data Analysis & Decision Making 2 — Questions and Answers
Question 1: A software manager notices that defect rates spike every Friday afternoon. Which analytical technique best identifies this pattern?
- Root cause analysis using fishbone diagrams
- Time-series analysis with control charts (Correct answer)
- Pareto analysis of defect categories
- Scatter plot correlation of team size vs. defects
Correct answer: Time-series analysis with control charts
Time-series analysis with control charts reveals temporal patterns and trends in process data, making Friday spikes visible.
Question 2: When evaluating a build-vs-buy decision for a software component, which quantitative method best accounts for uncertainty in future maintenance costs?
- Net Present Value (NPV) with sensitivity analysis (Correct answer)
- Simple payback period calculation
- Return on Investment (ROI) at fixed cost assumptions
- Earned Value Management (EVM) baseline comparison
Correct answer: Net Present Value (NPV) with sensitivity analysis
NPV with sensitivity analysis accounts for the time value of money and tests how outcomes change under different cost assumptions.
Question 3: A project dashboard shows Schedule Performance Index (SPI) of 0.85 and Cost Performance Index (CPI) of 1.10. What is the most accurate interpretation?
- The project is ahead of schedule and over budget
- The project is behind schedule but under budget (Correct answer)
- The project is on schedule and on budget
- The project is behind schedule and over budget
Correct answer: The project is behind schedule but under budget
SPI below 1.0 means behind schedule; CPI above 1.0 means under budget — both relative to the performance measurement baseline.
Question 4: Which statistical concept describes the tendency of extreme measurements to move closer to the average on subsequent measurements, potentially misleading managers?
- Selection bias
- Confirmation bias
- Regression to the mean (Correct answer)
- Hawthorne effect
Correct answer: Regression to the mean
Regression to the mean causes extreme values to drift back toward the average, which can make an intervention appear effective when it was not.
Question 5: A software manager must choose between two risk mitigation strategies. Strategy A has a 70% chance of saving $200K and Strategy B has a 40% chance of saving $400K. Based purely on expected value, which is preferred?
- Strategy A ($140K expected value)
- Strategy B ($160K expected value) (Correct answer)
- Both are equal in expected value
- Neither can be evaluated without standard deviation data
Correct answer: Strategy B ($160K expected value)
Strategy B's expected value is 0.40 × $400K = $160K, which exceeds Strategy A's 0.70 × $200K = $140K.
Question 6: When using a balanced scorecard for software management decisions, which four perspectives are typically evaluated?
- Cost, Schedule, Scope, and Quality
- Financial, Customer, Internal Processes, and Learning & Growth (Correct answer)
- People, Process, Technology, and Governance
- Inputs, Outputs, Outcomes, and Impacts
Correct answer: Financial, Customer, Internal Processes, and Learning & Growth
Kaplan and Norton's balanced scorecard framework evaluates Financial, Customer, Internal Processes, and Learning & Growth perspectives.
Question 7: A manager wants to determine if two software teams have statistically different defect rates. Which test is most appropriate when comparing two independent proportions?
- Paired t-test
- Chi-square test of independence (Correct answer)
- One-sample z-test
- ANOVA F-test
Correct answer: Chi-square test of independence
The chi-square test of independence is used to compare proportions across two independent categorical groups.
A software manager notices that defect rates spike every Friday afternoon.
Which analytical technique best identifies this pattern?