23. Choosing a statistical test (A2) Flashcards

1
Q

What is a statistical test?

A

Used in psychology to determine whether a significant difference or correlation exists, and consequently whether the no hypothesis should be rejected or retained.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the eight statistical tests in psychology?

A
The Chi Squared test
The Sign test
Mann-Whitney test
Wilcoxon test
Spearman's Rho
Pearson's r
Related t-test
Unrelated t-test
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Which three factors determine which statistical test should be used?

A
  1. Whether the data is looking for a difference or correlation
  2. Whether the experimental design is related or unrelated
  3. Whether the data is nominal, ordinal or interval
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is the mnemonic to remember the 8 statistical tests in the order they are in the table?

A

Simon (sign) Cowell (Chi) Wants (Wilcoxon) More (Mann-Whitney) Singers (Spearman’s) Receiving (Related -t) Unanimous (Unrelated-t) Praise (Pearson’s)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are the categories on the table showing the different statistical tests?

A

Testing for; difference or association
Design; Related or Unrelated
Data; Nominal, Ordinal, Interval

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Which statistical test can be used for both tests of difference and association/correlation?

A

Chi-Squared test (takes up two columns on the top row)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is nominal data?

A

Data represented in the form of categories - discrete data bc an item can only appear in one category

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is ordinal data?

A

Data that is ordered in some way but without regular intervals - often data from a sliding numerical scale - subjective so not totally reliable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is interval data?

A

Data from numerical scales that include units of equal, precisely defined size - more detailed and precise than ordinal which in turn is ‘better’ than nominal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

When should the Chi-Squared test be used?

A

Difference or association; both
Design; unrelated
Data; nominal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

When should the Sign test be used?

A

Difference or association; difference
Design; related
Data; nominal or better

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

When should the Mann-Whitney test be used?

A

Difference or association; difference
Design; unrelated
Data; ordinal or better

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

When should the Wilcoxon test be used?

A

Difference or association; difference
Design; related
Data; ordinal or better

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

When should the Spearman’s rho test be used?

A

Difference or association; association
Design; n/a
Data; ordinal or better

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

When should the Pearson’s r test be used?

A

Difference or association; association
Design; n/a
Data; interval

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

When should the related-t test be used?

A

Difference or association; difference
Design; related
Data; interval

17
Q

When should the unrelated-t test be used?

A

Difference or association; difference
Design; unrelated
Data; interval

18
Q

What are all of the test on the bottom row of the table known as?

A

Parametric tests

19
Q

What are parametric tests?

A

Tests that are more powerful and robust than the others, they can detect significance in data sets that others cannot

20
Q

What are the criteria for data to be tested using a parametric test?

A
  1. Must be interval level
  2. Data must be drawn from a population that wld expect to show a normal distribution of the variable being measured
  3. There shld be homogeneity of variance - the set of scores in each condition shld have a similar dispersion.
21
Q

What happens if you have data that wld usually require a test from the bottom row but it isn’t suitable for a parametric test?

A

You shift up one row in the table and use the ordinal test directly above the original test