Statistics & Probability for Data Science — Quiz | Data Science using Python

Question 1

What is the difference between descriptive and inferential statistics?

A There is none

B Descriptive summarises the data you have; inferential generalises from a sample to a population

C Descriptive is harder

D Inferential only uses charts

Question 2

Why use ddof=1 when computing a sample standard deviation?

A It runs faster

B It corrects bias when estimating a population from a sample (n−1)

C It rounds the result

D It is required for the mean

Question 3

Which SciPy method gives the probability that a value is below a threshold?

A ppf

B cdf

C pmf

D rvs

Question 4

The Central Limit Theorem says that as sample size grows, the distribution of the sample mean becomes…

A more skewed

B approximately normal regardless of the population shape

C uniform

D identical to the population

Question 5

The standard error of the mean scales as…

A σ × n

B σ / √n

C σ²

D 1 / n

Question 6

What does a 95% confidence interval actually mean?

A 95% chance the true value is in this interval

B If we repeated sampling, ~95% of such intervals would contain the true value

C The data is 95% accurate

D 95% of the data lies in it

Question 7

What is a key advantage of the bootstrap?

A It needs no data

B It builds confidence intervals by resampling, with few distribution assumptions

C It is always exact

D It removes outliers

Question 8

In hypothesis testing, what is the p-value?

A The probability the null is true

B How surprising the data is if the null hypothesis were true

C The effect size

D The sample size

Question 9

A result has p = 0.001 but a tiny effect size. What is the honest conclusion?

A It is a huge, important effect

B It is statistically significant but may be practically meaningless

C The test is wrong

D The null must be true

Question 10

Why should you fix the significance threshold and test duration before an A/B test?

A It is faster

B To avoid 'peeking' which inflates false-positive rates

C It is required by statsmodels

D To use less data

Question 11

Which test compares two conversion rates in an A/B test?

A A correlation

B A proportions z-test

C A histogram

D A pivot table

Question 12

What question does a Bayesian A/B analysis answer directly?

A How surprising is the data under the null?

B Given the data, what is the probability B is better than A?

C What is the sample mean?

D How many outliers are there?