tests of correlation: spearman's and pearson's r

Question 1

when should you use spearman’s rho?

Answer 1

• looking for correlation between two sets of values
• when one or both variables are ordinal
• type of design is not an issue as investigation is correlational rather than experimental

Question 2

(SR) step 1: table of ranks

Answer 2

• rank each set of scores separately in each group / condition from lowest to highest
• if two or more scores share the same ranks, find the mean of their total ranks

Question 3

(SR) step 2: calculate the different

Answer 3

• find the difference between each pair of ranks
• square the difference
• add up the squared differences

Question 4

(SR) step 3: working out the value of rho

Answer 4

rho = 1- (6∑d²) / (N (N² -1))

Question 5

(SR) step 4: the calculated and critical values

Answer 5

if the calculated value of rho (ignoring the sign) is more than the critical value, the result is significant. we reject the null hypothesis

Question 6

when should you use pearson’s?

Answer 6

• looking for correlation between two sets of values
• interval data
• type of design is not an issue as investigation is correlational rather than experimental
• parametric test

Question 7

(SR) step 1: the table of data

Answer 7

• calculate the sum of the scores for x (∑x) and y (∑y)
• square each x value and each y value (∑x² and ∑y²)
• multiply x and y for each participant. add these values together = ∑(xy)

Question 8

(SR) step 2: working out the value of r

Answer 8

suuuper complicated equation sorry. maybe this wasn’t such a good idea

Question 9

(SR) step 3: calculated and critical values

Answer 9

• df = N - 2
• if calculated value of r is more than the critical value, the result is significant at the 0.05 level. we can reject the null hypothesis and accept the alternative hypothesis