Paired samples t-tests

Question 1

What does a t-test tell us?

Answer 1

The t test tells us how likely it is that we will observe a difference - is this by chance?

Question 2

why would we use a t test?

Answer 2

when one of our variables has more than one condition.

Question 3

what kind of data design must we have to do a t-test?

Answer 3

repeated design

Question 4

how many conditions must our IV have to do a t test?

Answer 4

2

Question 5

How many participants do you need to do a t-test?

Answer 5

10

Question 6

What scale should our data be on to use a t-test?

Answer 6

interval/ratio

Question 7

The difference scores (change scores) for t tests must be…

Answer 7

Normally distributed :)

Question 8

T tests are interested in the __________ between participant’s T1 and their T2 scores.

Answer 8

differences

Question 9

When will the t value get bigger?

Answer 9

1. when the difference scores are larger, assuming they are also consistently in the same direction.
2. when there is less variation in difference scores.

Question 10

___ variation is more likely to be significant as there is a much smaller distribution, this score is an extreme.

Answer 10

less

Question 11

t (951) = -12.39, p< .001 (two tailed)

Anything missing?

Answer 11

Yes effect size!

Question 12

If you use a 1 tailed hypothesis what do you have to do to the p value?

Answer 12

We have to half it!

Question 13

what is a small effect size?

Answer 13

0.1

Question 14

What is a medium effect size?

Answer 14

0.3

Question 15

What is a large effect size?

Answer 15

> 0.5

Question 16

What is the formula for r (effect size)?

Answer 16

√ t2/ (t2 + df)

Question 17

t = -12.39
df= 951

Calculate the effect size.

Answer 17

√ t2/ (t2 + df)

√-12.39 squared/ (-12.39 squared + 951)

r = 0.37

Question 18

Larger T value =…

Answer 18

larger difference and smaller variation

Question 19

What do t-tests evaluate?

Answer 19

The difference in participants’ scores from one condition to the other.

Question 20

What 2 non-parametric tests would you use?

Answer 20

2 conditions: Wilcoxon Signed ranks

3 or more conditions: Friedman Test

Question 21

if you have 2 conditions what non-parametric test do you use?

Answer 21

Wilcoxon Signed Ranks

Question 22

If you have 3 conditions or more, what non-parametric test do you use?

Answer 22

Friedman Test

Question 23

What is the non-parametric equivalent to the t-test?

Answer 23

Wilcoxon signed Ranks

Question 24

what conditions does non-parametric data reflect?

Answer 24

Less than 10 participants
Ordinal data
Data skewed

Question 25

What does Wilcoxon Signed Ranks test?

Answer 25

whether data in 1 condition is consistently ranked more highly than in the other condition.

Question 26

How does ranking work if people have the same score?

Answer 26

If we have 3 score 1’s.
1 + 2 + 3 = 6

6/3 = 2

Question 27

The ___ value is always the smaller number of the 2 sums of ranks.

Answer 27

the T value :)

Question 28

How is a Wilcoxon signed ranks reported?

Answer 28

T = 2, N= 8, p= .023 (two tailed).

Question 29

What is the effect size in Wilcoxon Signed Ranks?

Answer 29

r = z / square-root-of-N

Question 30

What are Cohen’s guidelines for effect sizes

Answer 30

r = .10 (small), .30 (medium), .50 (large

Question 31

T = 2, N = 8, p = .023 (two-tailed), r = .80.

When would you not report effect size?

Answer 31

If p wasn’t significant.

Question 32

What is an extension of the Wilcoxon?

Answer 32

Friedman- for 3 or more conditions :)

Question 33

What is the standard form to report Friedmans?

Answer 33

χ2 (2) = 12.25, p = .002

Question 34

After we get our Freidman’s what do we have to do?

Answer 34

We have to conduct follow up Wilcoxon’s to find out where the sig differences lie!

Question 35

If we have 4 conditions, how many Wilcoxon follow up tests do we have to do?

Answer 35

A v B
A v C
A v D
B v C
B v D
C v D

6 tests!!!

Question 36

What do Wilcoxon follow up tests inflate?

Answer 36

They inflate the chances of making a type I errors.

Question 37

How do we resolve the increased chance of makimg a type I error in a follow up Wilcoxon test?

Answer 37

We do a bonferromi Correction!

Question 38

What is a Bonferroni Correction?

Answer 38

Divide the level of significance (0.05) by the number of tests you are conducting.

Question 39

I you have 3 Wilcoxon follow up tests- what is the significance value?

Answer 39

0.05/3 = 0.017

Question 40

What is the appropriate effect size to use for paired sample t-tests?

Answer 40

Cohen’s d