two-sample t-test

Question 1

When would i used a two sample t-test?

Answer 1

• two groups (e.g. treatment group vs. control group)
• groups are independent
• data is numerical
• ## comparing means

Question 2

What are the three types of two-sample t-tests?

Answer 2

Parametric = Student t-test and Welch t-test
Non-Parametric = Mann-Whitney U test

Question 3

Assumptions for two-sample t-test

Answer 3

• Continuous DV & categorical IV
• Data is independent
• Homogeneity of variance (variances between the groups are equal)
• Residuals in each group are approximately normally distributed

Question 4

How do you find t-value for student t-test?

Answer 4

The difference between the mean of group 1 and the mean of group 2
divided by
the square root of
(the pooled variance divided by the sample size of group 1) plus (the pooled variance divided by the sample size of group 2).

Question 5

What does Student t-test assume?

Answer 5

Assumes equal variances

Question 6

how do you calculate df?

Answer 6

df = n(1) + n(2) - 2

Question 7

How do you find pooled variance?

Answer 7

(The variance of group 1 multiplied by its sample size minus 1) plus (the variance of group 2 multiplied by its sample size minus 1),
all divided by
the total sample size of both groups minus 2.

Question 8

Formula for the Welch t-test

Answer 8

t = (The difference between the mean of group 1 and the mean of group 2) divided by square root (Sample variance divided by sample size - same for other group)

Question 9

How do u find the df for the Welch t-test?

Question 10

How would you report results for Levene’s Test for Equality of Variance in APA?

Answer 10

A Levene’s test of homogeneity of variances was used to compare the variances of the Sophomores and Junior . The test indicated that the variances of the two groups were homogeneous, 𝐹(1,150)=1.05, 𝑝=.307

Question 10

What test do we use to check Homogenity of variance

Answer 10

Leven’s Test for Equality of Variance

Question 11

What test do we use if any assumptions are violated?

Answer 11

non-parametric Mann-Whitney U test

Question 12

What do we need for a Mann-Whitney U test?

Answer 12

• data must be ordinal, interval or ratio
• Needs independent scores