Statistics CERTIFIED-DATA-SCIENCE-PRACTITIONER Practice Questions (Page 5)

Questions 21–24

1. 21foundation · medium

   In hypothesis testing, what does the significance level (alpha) represent?

   • AThe probability that the observed result is due to chance
   • BThe probability of correctly rejecting the null hypothesis when it is false
   • CThe probability of failing to reject the null hypothesis when it is false
   • DThe maximum probability of rejecting the null hypothesis when it is actually true
   Reveal Solution Discussion Select an answer first
2. 22expert · hard · select all that apply

   A data scientist runs a hypothesis test comparing two models' accuracy and obtains a p-value of 0.04. Which of the following statements are valid interpretations or appropriate next steps? Select all that apply.

   • AThe null hypothesis can be rejected at the 0.05 significance level.
   • BThe null hypothesis cannot be rejected at the 0.01 significance level.
   • CThe p-value of 0.04 means there is a 4% chance the null hypothesis is true.
   • DThe result is practically significant because the p-value is below 0.05.
   • EThe data scientist should compute an effect size to assess the magnitude of the difference.
   Check answers Discussion Select an answer first
4. 23expert · hard

   A data scientist is comparing the mean scores of two independent groups using a t-test. The Shapiro-Wilk test for normality returns a p-value of 0.03 for Group 1 and 0.20 for Group 2. The sample sizes are 15 and 15. What is the most appropriate course of action?

   • ATransform Group 1's data and then run the t-test on the transformed data.
   • BUse the Mann-Whitney U test because Group 1 is not normally distributed.
   • CRun the t-test anyway because the Shapiro-Wilk test is too conservative.
   • DProceed with the t-test because the combined sample is large enough for the central limit theorem.
   Reveal Solution Discussion Select an answer first
5. 24expert · hard

   A data scientist is evaluating a new fraud-detection model. The null hypothesis is that the new model's false positive rate equals the current model's rate. The test yields a p-value of 0.049. The compliance team requires a significance level of 0.01 to avoid false alarms. The business team argues the model is better because the p-value is just below 0.05. What is the most defensible decision?

   • AChange the significance level to 0.05 after seeing the p-value to justify adoption.
   • BReject the null hypothesis at the 0.01 level and adopt the model.
   • CAdopt the new model because the p-value is below 0.05, the conventional threshold.
   • DFail to reject the null hypothesis at the 0.01 level, but report the p-value and effect size so stakeholders can make an informed decision.
   Reveal Solution Discussion Select an answer first