In a one-way repeated measures ANOVA design,each subject is exposed to two or more different conditions, or measured on the continuous scale onthree or more occasions. It can also be used to compare respondents’ responses to two or more questions or items. These questions, hiwever, must be meausred using the same scale.( Likert scale) Example of research question: Is there a change in confidence scores over the three time periods? What you need: One group of participants measured on the same scale on three different occasions or under three different conditions, or each person measured on three different questions or items ( using the same scale). This involves teo variables: ● one indepenedet variable ( categorical) ( e.g. Time 1/Time 2/Time 3) ● one dependent variable ( numerical) (e.g scores on the confidence in coping with statistics test). The scores on the test for each time point will appear in the data file in different columns. What it does: This technique will tell you if there is a significant difference somewhere among the three sets of scores Assumptions: Same as ANOVA

Non-parametric alternative: Friedman Test

Example

A group of students were invited to participate in an intervention designed to increase their confidence in their ability to do statistics. The confidence levels wer measyred before intervention (Time 1), after the intervention (Time 2) and again three months later (Time 3) Data: file experim4ED.sav

Variables:

● Confidence scores at Time 1( Confid 1): Total scores on the confidence in coping with statistics test administered prior to the program. Score range from 10 to 40. ● Confidence scores at Time 2( Confid 2): Total scores on the confidence in coping with statistics test administered after the program. ● Confidence scores at Time3( Confid 3): Total scores on the confidence in coping with statistics test administered three months later

PROCEDURE FOR ONE-WAY REPEATED MEASURE ANOVA:

1. Analyze – General Linear Model- Repeated Measure

2. In the Within Subject Factor Name box, type in a name that represents your independent variable ( e.g. Time or condition) 3. In the Number of Levels box, type the number of levels or groups ( time periods) involved ( in this example, it is 3) 4. Click on the Add

5. Click on the Define button

6. Select the three variables that represent your repeated measures variable ( e.g. confid1, confide.2, and confide 3) and move into Within Subjects Variable box 7. Click on Options box

8. Tick Descriptive Statistics and Estimates of effect size boxes in the area labelled Display. If you wish to request Post-hoc tests, select your independent variable name( e.g. Time) in the Factor and Factor Interactions section and move it into the Display Means for box. Tick Compare main effects. In the Confidence interval adjustment section, click on the down arrow and choose the second option B 9. Click on Continue and then OK

RESULTS

MULTIVARIATE TESTS

Wilk’s lambda p = 0.0000 < 0.05 indicates that there is a statistically significant effect for time Effect size: Partial Eta Squared = 0.749. Using the guideline given by Cohen( .01 = small; .06 = moderate; .14 = large effect) this result suggest very large effect size. PAIRWISE COMPARISION

There is a difference among the groups

Within-Subjects Factors|

Measure:MEASURE_1|

Time| Dependent Variable|

1| fost1|

2| fost2|

3| fost3|

Descriptive Statistics|

| Mean| Std. Deviation| N|

fear of stats time1| 40.17| 5.160| 30|

fear of stats time2| 37.50| 5.151| 30|

fear of stats time3| 35.23| 6.015| 30|

Multivariate Testsb|

Effect| Value| F| Hypothesis df| Error df| Sig.| Partial Eta Squared| Time| Pillai's Trace| .635| 24.356a| 2.000| 28.000| .000| .635| | Wilks' Lambda| .365| 24.356a| 2.000| 28.000| .000| .635| | Hotelling's Trace| 1.740| 24.356a| 2.000| 28.000| .000| .635| | Roy's Largest Root...