Statistics & Math for ML · Lesson 66 of 138
Inference Quiz
Test yourself.
Let's test what you know about inference. Pause before each answer.

- Five questions
- Pause and think
- Then listen
Question 1
How does the standard error change when you collect four times as much data? The answer is, it halves.
Quick check
4× the data?
Question 2
A p value is zero point zero three. With alpha at five percent, what do we do? The answer is, reject the null hypothesis.
Quick check
p = 0.03, α = 0.05?
Question 3
Which error is a false alarm, finding an effect that isn't there? The answer is a type one error.
Quick check
A false alarm?
Question 4
Which method builds a confidence interval by resampling your own data? The answer is the bootstrap.
Quick check
Resampling your data?
Question 5
A result is significant, but the effect size is tiny. Is it important? The answer is, probably not.
Quick check
Significant but tiny?
Summary
Excellent. Next, a new section, relationships between variables, starting with covariance.
Key takeaways- SE shrinks with √n
- p < α: reject
- Effect size matters
Next up: Covariance — moving together