Exercise on the t-test (intermediate)

Scenario:
You are investigating whether a motivation workshop significantly increases employees’ motivation. To do this, motivation is measured before and after the workshop.

Dataset: Mittel_Motivationsvergleich.csv

Download data

Variables:

person: Participant ID
motivation_vorher: Scale score (1–10) before the workshop
motivation_nachher: Scale score (1–10) after the workshop

Solution in R

1. Import the data

daten <- read.csv("Mittel_Motivationsvergleich.csv")

2. Check the assumptions

a) Calculate and test the differences

daten$diff <- daten$motivation_nachher - daten$motivation_vorher
shapiro.test(daten$diff)

b) Visualize (optional)

boxplot(daten$motivation_vorher, daten$motivation_nachher, names = c("Before", "After"))
qqnorm(daten$diff); qqline(daten$diff)

3. Paired-samples t-test

t.test(daten$motivation_vorher, daten$motivation_nachher, paired = TRUE)

4. Alternative if the normality assumption is violated

wilcox.test(daten$motivation_vorher, daten$motivation_nachher, paired = TRUE)

Solution in SPSS

1. Import the file

Open SPSS
File → Open → Mittel_Motivationsvergleich.csv

2. Check the prerequisites

Menu: Analyze → Descriptive Statistics → Explore
Target variable: diff (only after calculation)
Create the difference: Transform → Compute Variable → motivation_nachher - motivation_vorher

→ The Shapiro–Wilk test and histogram provide information about the distribution

3. Paired-samples t-test

Menu: Analyze → Compare Means → Paired-Samples T Test
Pairs: motivation_vorher and motivation_nachher
Click OK

4. Alternative for non-normal distributions

Test: Wilcoxon signed-rank test

Menu: Analyze → Nonparametric Tests → Two Related Samples

Test variables: motivation_vorher, motivation_nachher

Alles klar?

Ich hoffe, der Beitrag war für dich soweit verständlich. Wenn du weitere Fragen hast, nutze bitte hier die Möglichkeit, eine Frage an mich zu stellen!