301 Test 3

Independent samples t-test

To compare means of two separate samples

Independent samples t-test is a ______ subjects design

Between

Independent samples t-test formula for t

Pros of between subjects design

• Prevents carryover effects

• Prevents burnout (shorter duration)

Cons of between subject design

• larger samples needed for sufficient power

• More resources

• Individual differences

To compare an experimental and a control group, you would use a(n) ________.

Independent samples t-test

Degrees of freedom

The number of scores in a sample that are free to vary once a mean is known

How do df effect the distribution of t?

Large df: normal distribution

Small df: flatter curve (requires a larger OV, punishes small samples)

Effect size

Measure of the mean difference in terms of standard deviations between the raw scores

_____ is a small effect

0-0.2

_____ is a medium effect

0.2-0.5

_____ is a large effect

0.5+

Independent samples t-test effect size formula

How is t reported in APA?

t(df) = OV, p ___ .05

Independent samples t-test df formula

Statistical power

Probability that a test will correctly reject a false null

How can statistical power be increased?

• increase effect size

• Increase sample size

• Increase alpha level

• Use a one-tailed test

Dependent samples t-test

To compare means from testing the same sample more than once

Dependent samples t-test aka

Repeated measures design

Within subjects design

Within subjects design pros

• need fewer participants to get the same df/power

• Each participant acts as their own control

Within subjects design cons

• order/learning effects

• May increase potential for alternative explanations

Dependent samples formula for t

(T/F): dependent and independent sample t-tests use different t-tables to determine the critical value.

False, it is the same t-table

ANOVA

Analysis of variance

Based on sources of variation rather than mean differences

Factor

The independent variable that designates the groups to be compared in ANOVA

Level

Different groups within a factor

What is the goal of the ANOVA formula?

Divide out differences due to chance in order to isolate the treatment effect

Between groups variance reflects differences due to ________

1. Treatment effects

2. Chance, error, etc.

Within groups variance reflects differences due to ______

Chance, error, etc.

The larger the F value, the ______ the p value

Smaller

Sum of squares between

Sum of squares within formula

Variance ______ groups / variance _____ groups

Between / within

Omnibus test

Identifies that there is a difference between groups, but not which groups

Which test is an omnibus test?

F-test, ANOVA

ANOVA numerator df formula

k - 1

ANOVA denominator df formula

N - k

What is k?

The number of levels

N vs n

Total sample size vs group sample size

Carryover effects

If a dependent sample is given the treatment before tested for control, than the treatment effects may carryover into the control measurement.

Maturation effects

With a dependent sample, natural life development/maturation may confound treatment effects if they are measured with considerable time between

Order/learning effects

Participants in a dependent sample may learn from the first testing, enhancing performance on the second testing