Chapter 6: Analyzing Data From Independentgroups: Continuous And Ordinal Measures

Front 1

Used for comparing two (or more) groups which are independent of one another – meaning there are different people in each group.

Back 1

Independent Group Designs

Front 2

people are not assigned to conditions, as they already belong to different groups.

E.g., Males and Females, Normal and Abnormal etc.

Back 2

Quasi-experimental designs

Front 3

three different kinds of dependent variables can be used in Independent group designs

Back 3

1. Continuous outcomes,

2. Ordinal outcomes, and

3. Categorical outcomes.

Front 4

Advantages of IGD

Back 4

1. No practice effects

2. No sensitization

3. No carry-over effects.

Front 5

Disadvantages of IGD

Back 5

1. Need more people,

2. difficulty of matching controlled group with experimental group

Front 6

Statistical Tests for Independent Groups: CONTINUOUS DATA

Back 6

The Independent Groups t-test

Front 7

Usually the most powerful and is the test most likely to spot significant differences in your data (IGD)

Back 7

IGD T-TEST

Front 8

IGD T-test is also known as

Back 8

between-subjects t-test

or

the two-samplest-test

Front 9

The Mann-Whitney Test is also known as ____________ and equivalent to:____________

Back 9

Wilcoxon-Mann Whitney Test;

Wilcoxon rank-sum test

Front 10

Statistical Tests for Independent Groups for: NOMINAL/CATEGORICAL

Back 10

The Chi-Square Test

Front 11

Give the common assumptions about the Data

Back 11

1. Measured in continuous scale

2. The data is normally distributed

3. The SD of the two groups are equal

Front 12

FACT OR MYTH: Sample size must be above some value, such as 6, for the t-test to be valid

Back 12

MYTH

Front 13

Statistical Tests for Independent Groups: ORDINAL DATA

Back 13

The Mann-Whitney Test

Front 14

FACT OR MYTH: Sample sizes must be balanced, or similar

Back 14

MYTH

Front 15

distribution of two (or more) groups can be shown for the same amount of space.

Back 15

Box and whisker plot

Front 16

The equal standard deviations assumption

Back 16

Homogeneity of Variance

Front 17

Spread of Data

Back 17

Variance

Front 18

If you have (approximately) equal sample sizes in your two groups, use the __________ t-test.

Back 18

pooled variance

Front 19

If you don’t have (approximately) equal sample sizes in the two groups, use the __________ t-test

Back 19

unpooled variance

Front 20

If Levene's test gives a statistically significant result

Back 20

unpooled variance

Front 21

If Levene's test gives a not statistically significant result

Back 21

pooled variance

Front 22

The VARIANCES are DIFFERENT

Back 22

unpooled variance

Front 23

Pooled or unpooled:

VARIANCES are the SAME ( or almost)

Back 23

pooled variance

Front 24

Pooled or unpooled:

The VARIANCES are DIFFERENT

Back 24

unpooled variance

Front 25

Unpooled Variance T-test is also known as

Back 25

WELCH'S TEST

Front 26

measure of how far apart the means of the two samples are, in SD unit (gaano kalayo ang means)

Back 26

Cohen’s d

Front 27

The larger the _______, the larger the difference

Back 27

standard score

Front 28

Compares two unrelated groups when t-test isn't available

Back 28

The Mann-Whitney U Test

Front 29

Test for not norminally distributed and ordinal data

Back 29

The Mann-Whitney U Test

Front 30

In Mann Whitney U test, we only compare the ___________

Back 30

ranks