# Class notes ancova

## 10 important questions on Class notes ancova

### How do we call a continuous variable that has influence on the dependent variable, but is not part of the experiment, so not manipulated?

### What do you do when you use ancova?

### How does a covariate increase explained variance?

- Higher grades + faster learning
- Never study anything twice
- 100% sure, 100% understanding

### When using standard contrasts in spss, what does simple, first mean and when should you use it?

- Everything is compared to the first group
- use it when the baseline group, placebo, is the lowest valued group in spss

### With an ancova contrast table, what do you look at?

### How do you test the assumption of equal scores?

- Perform an independent sample t-test (two groups), or a one way anova (multiple groups).
- when performing the test, put the covariate as dependent variable, and the groups as independent variable.
- if the test is not significant, the assumption is not violated.

### How do you test the assumption of equal regression coefficients?

- Perform the ancova
- click on model, and check custom.
- put the dependent variable in, and the covariate
- and the interaction between the two: medicine * covariate
- choose main effects in the middle
- check the table test of between-subject effects.
- if medicine * covariate is non-significant, the assumption is not violated.

### How do you interpret the results of an ancova?

- Go to the table test of between subject effects.
- check source covariate: if significant, then the covariate predicts the scores on the dependent variable.
- so the model is adjusted for the covariate.
- then check source independent variable: if significant, the adjust independent variable predicts the scores on the dependent variable.

### How to perform an ancova in spss?

- Analyze -> General Linear Models -> Univariate
- Dependent variable: afhankelijke variabele

Fixed factor(s): onafhankelijke variabele

Covariate(s): de covariaat (vb leeftijd of sekse) - Model -> Full factorial (alleen wanneer je de assumptie homogenity of regression slopes wilt checken kies je voor custom)
- Als je een hypothese hebt, kies dan bij contrasts voor simple en dan voor first of last (afhankelijk van welke groepen je met elkaar wilt vergelijken) en op change
- Plots -> onafhankelijke variabele -> horizontal axis -> add
- Descriptive statistics
- Homogeneity tests (Levene’s test)
- Estimates of effect size (partial eta squared, oftewel de geschatte effectgrootte van de onafhankelijke variabele)

- Descriptive statistics
- Options -> vink aan:

### When do you use custom model in spss?

The question on the page originate from the summary of the following study material:

- A unique study and practice tool
- Never study anything twice again
- Get the grades you hope for
- 100% sure, 100% understanding