Hypothesis tests for correlation

Hypothesis tests for correlation Flashcards

Spec O1 · AQA A-Level Maths: Flashcards

Studying this course? Keep your progress in one place.

Flip each card. Be honest: the ones you mark Not yet come back first.

12 cards

Tap a card to flip it. Swipe right if you knew it, left if not.

See all 12 cards
What does the product moment correlation coefficient rr measure?
How close the points lie to a straight line, with −1≤r≤1-1\le r\le1.
What does rr close to 00 tell you?
There is little or no linear correlation (a non-linear relationship may still exist).
What symbol is used for the population correlation coefficient?
ρ\rho (rho).
What is the null hypothesis in a test for correlation?
H0:ρ=0H_0:\rho=0, no correlation.
What are the possible alternative hypotheses?
ρ>0\rho>0 or ρ<0\rho<0 (one-tailed), or ρ≠0\rho\ne0 (two-tailed).
When do you reject H0H_0 for H1:ρ>0H_1:\rho>0?
When r≥r\ge the critical value.
When do you reject H0H_0 for H1:ρ<0H_1:\rho<0?
When r≤−r\le- the critical value.
When do you reject H0H_0 in a two-tailed test?
When ∣r∣≥\lvert r\rvert\ge the two-tailed critical value.
How do you use a pp-value?
Reject H0H_0 if the pp-value is less than the significance level.
Why can a correlation test not prove causation?
It shows association only; another variable may explain both.
Why is predicting outside the data range unreliable?
The relationship may not continue beyond the data (extrapolation).
When is a one-tailed test justified?
Only if the direction of correlation is decided before seeing the data.

Flashcards