T tests

Question 1

Independent T-test:
Question type
DV
IV
Design

Answer 1

Difference
continuous
categorical
Between subjects

Question 2

Independent T-test:

Non-parametric version?

Answer 2

Mann-Whitney

Question 3

Falsifying the null using the results?

Answer 3

Calculate 95% CI. If it contains predicted pop. mean = less than 5% chance = null isn’t correct

Question 4

Assumptions?
under 50 ps?
over 50 ps?

Answer 4

Histogram - normally distributed
Shapiro Wilk - normally distributed 
Spread of scores (variance/SD) is relatively equal in both groups= Levene's test for equality of variances:
p>0.05= variances equal 
0<0.05 = variance not equal
(Kolmogorov  smirnov)

Question 5

Degrees of freedom

Answer 5

no. of data that you can freely vary and still get the same descriptive stats in your sample
(n-1)

For independent = have 2 groups SO,
we have to do (n-2)

For paired = have one group so,
we do (n-1)

Question 6

Errors in hypothesis testing?

Answer 6

Type 1 error = p<0.05 - conclude there is a relationship/difference wrongly
(wrongly reject the null)
Type 2 error = wrongly accept the null

Question 7

Paired T-test
Question type
DV
IV
Design

Answer 7

differences
continuous
categorical
within subjects

Question 8

Non-parametric for paired t test?

Answer 8

Wilcoxon