Tests of correlation

Advantages of Nonparametric Tests

1. Used With All Scales

2. Easier to Compute

3. Make Fewer Assumptions

4. Need Not Involve Population Parameters

5. Results May Be as Exact as Parametric Procedures

Disadvantages of Nonparametric Tests

1. May Waste Information Parametric model more efficient if data Permit

2. Difficult to Compute by hand for Large Samples

3. Tables Not Widely Available

What are interval, ordinal and nominal data?

Are sign tets parametric or non parametric?

Sign tests are non- parametric

When are sign tests used?

Test of difference [ two separate conditions]

Repeated/matched pairs[ same participants doing both condition]

Level of measurement is ordinal [equal increments in unit]

How do you find the critical value of Sign tests?

To find the critical value for Sign test

• P < 0.05

• Check whether your hypothesis is directional [one tailed] or

• non-directional [ two tailed]

• Look at the number of participants

Are Chi squared tests parametric or non parametric?

When do you use Chi squared tests?

Test of difference/association [ two separate conditions]

Independent group design [ different participants in each condition]

Level of measurement is nominal [categories/groups]

How do you find the critical value or chi squared tests?

To find the critical value for Chi-squared test

1. P< 0.05

2. Check whether your hypothesis is directional [one tailed] or non-directional [ two tailed]

3. Work out the df [(number of rows-1) x (number of columns - 1)]

is Spearmans Ro parametric or non parametric?

Non parametric

When is Spearmans Ro used?

• Test of correlation [ 2 variables from same person]

• Repeated measure [ same participant is used for 2

• variables]

• Level of measurement is ordinal [can be ranked]

How do you find the critical value of Spearmans Ro?

To find the critical value under Spearman’s rho

• P < 0.05

• Check whether your hypothesis is directional [one

• tailed] or non-directional [ two tailed]

• Work out the N

How do you know results are significant?

Critical value with the calculated or obserevd value must be compared

How is the crtical value found?

• 1 in 20 or P<0.05

• Checkif hypothesis is directional or non directional

• Work out N

Pearsons R’ parametric or non parametri?

Parametric

When is Pearsons R’ used?

• To test for correlation

• For repeated measures

• When there are inteveal levels of measurement

How do you find the critical value of Pearsons R’?

• P < 0.05

• Check whether your hypothesis is one tailed or 2 tailed

• Work out the degrees of freedom [df = N - 2]

Mann Whitney

Is mann whitney non parametric or parametric?

Non parametric

When is Mann Whitney used used?

• Test of difference (Two seperate conditions)

• Independent group designs (different participants)

• Level of measurment is ordinal (can be ranked)

How do you find the critical of a mann whitney test?

• P < 0.05

• Check whether your hypothesis is one tailed or 2 tailed

• Work out the N_A and N_B

Wilcoxon

Is Wilcoxon parametric or non parametric?

Non parametric

When is the Wilcoxon used?

• Test of difference (Two seperate conditions)

• Repeated/matched pairs measure (same participants in ech condition)

• Level of measurment is ordinal (Can be ranked)

How do you find out the critical value of Wilcoxon

• P < 0.05

• Check whether your hypothesis is one tailed or 2 tailed

• Work out N

Test of difference

Is unrelated T- tests parametric or non parametric parametric

When do you use unrelated test if difference?

1. Test of difference [two separate conditions]

2. Independent group design[ different participants in each condition]

3. Level of measurement is interval [equal increments in unit]

How do you found the critical value for unrelated T-Tests

find the critical value for Unrelated T-test

1. P < 0.05

2. Check whether your hypothesis is directional [one tailed] or non-directional [two tailed)

3. Work out the df=[Na +NB)-2

is Related Test t parametric or non parametric?

Parametric

When do you use related T tests

Related T-test[parametric]

1. Test of difference [ two separate conditions]

2. Repeated/matched pairs[ same participants doing both condition]

3. Level of measurement is interval [equal increments in unit]

How do you find the critical value of related T tests ?

To find the critical value for Related T-test

P < 0.05

Check whether your hypothesis is directional [one tailed] or

non-directional [ two tailed]

Work out the df=N-1