Statistical Concepts and Inference Flashcards
7 cards from real Data Science practice questions. Tap to flip, then mark Knew It or Still Learning — missed cards come back until you master them.
Read the first 7 Statistical Concepts and Inference flashcards as text
A researcher conducts 20 independent tests at alpha = 0.05 with all nulls true; roughly how many false positives are expected?
Answer: 1
With a 5% error rate per test, about 1 in 20 tests is expected to be a false positive.
The Bonferroni correction is used to:
Answer: Control the family-wise error rate across multiple comparisons
Bonferroni adjusts the significance threshold to limit the chance of any false positive across many tests.
Bootstrapping is primarily used to:
Answer: Estimate the sampling distribution of a statistic by resampling with replacement
Bootstrapping resamples the observed data with replacement to approximate the sampling distribution of a statistic.
Which condition is a key assumption of ordinary linear regression?
Answer: Residuals have constant variance (homoscedasticity)
Linear regression assumes residuals have roughly constant variance across fitted values.
A Type II error occurs when a test:
Answer: Fails to reject a false null hypothesis
A Type II error is a false negative: failing to detect a true effect.
Selection bias most directly threatens which property of a study?
Answer: The internal validity and representativeness of the sample
Selection bias produces a non-representative sample, undermining the validity and generalizability of conclusions.
What does a wide confidence interval typically indicate about an estimate?
Answer: Greater uncertainty about the parameter
A wider interval reflects more uncertainty and less precision in the estimate.