WEEK 12: Comparing means t-tests part A

Question 1

What is the purpose of a t-test?
Which measures of central tendency is looked at?

Answer 1

Determines if two means are significantly different

Question 2

Assumptions of t-tests
What scale should the data be on?
What distribution should the scores in each group roughly have?
What should the two groups show? define?

Answer 2

• Data should be of interval or ratio scale
• Scores in each group should be roughly normally distributed (test for normality is the Kolmogorov simirnov test pvalue less than 0.05 shows data is not normally distributed)
• The two groups should show homogeneity of variance i.e. the spread of scores (variance) within each group should roughly be equal (levines test: pvalue less than 0.05 - assumption violated)

Question 3

Independent samples t-test
When does a significant result occur? More… between…than?

Answer 3

A t-test calculates the amount of variability between two conditions (the difference between the mean score in each condition) and the amount of variability within the conditions

A significant result occurs when there is more variability between the conditions than within the conditions

Question 4

Independent samples t-test
What columns are evaluated to report a t-test
Structure? which part has to be italicized? When is the result significant? when should equal variance not be assumed? which row should be reported?
How to work out degrees of freedom? what does n stand for?

Answer 4

The columns to evaluate are: t, df, and the direction of the hypothesis
Structure: t(df)= t-value, p< xxx
The p has to be italised
For a significant result, the p value has to be less than 0.05 (look at the ‘sig’ column

If the value is less than 0.05 the equal variances should not be assumed so the bottom row should be reported

Equation for degrees of freedom: (n1 - 1) + (n2 - 1) n = participants