22- Choosing an Inferential Statistical Test

Question 1

why do we use statistical tests in psychology?

Answer 1

statistics allow us to interpret and organise amounts of data

Question 2

what are the ‘Three Ds’ that must be decided to choose your statistical test?

Answer 2

• Difference or Association
• Design
• Data

Question 3

what does difference mean?

Answer 3

• when the aim of a study is to test for a significant difference

Question 4

what are the two types of design?

Answer 4

• related
• unrelated

Question 5

what criteria would mean it is a related design?

Answer 5

it would mean the study was either:
- a matched pairs design
- or a repeated measures design

Question 6

what criteria would the study need to be for it to be an unrelated study?

Answer 6

independent groups

Question 7

what are the the three levels of Data?

Answer 7

1. nominal
2. ordinal
3. interval

Question 8

what is nominal data?

Answer 8

data that is organised into categories

Question 9

what is ordinal data?

Answer 9

when items are ordered on a scale but it may not always be universal

Question 10

what is meant by universal data?

Answer 10

data everyone would agree with
i.e. 2+2=4

Question 11

draw the NOI-URAC grid

Question 12

define parametric tests

Answer 12

statistical tests which make assumptions about a population based on a sample size

Question 13

give two examples of parametric tests

Answer 13

• related T-Test
• Pearsons R test

Question 14

what is the advantage of parametric tests over non-parametric tests

Answer 14

parametric tests are more powerful and robust than non-parametric tests

they are able to test the significance within some data when non-parametric tests cannot

Question 15

what are the three criteria that must be met before you use a parametric test

Answer 15

1. Data must be interval, as parametric tests use actual scores not ranked data
2. data should be expected to show a normal distribution
3. Homogeneity of variance, meaning a dispersion of data