SPSS Test
When to use which test of significance – comparing means
One sample z test:
Assessing a mean against a certain value, standard deviation in population known.
Ratio variable.
One sample t test:
Assessing a mean against a certain value, standard deviation in population unknown.
Ratio variable.
Two sample t test:
Assessing whether there is a significant difference between means in two groups in the
population.
Ratio variable – DV.
Ordinal/nominal variable – grouping variable (2 groups).
ANOVA (F-test):
Assessing whether there is a significant difference between means in more than two groups
in the population.
Ratio variable – DV.
Ordinal/nominal variable – grouping variable (more than 2 groups).
Regression analysis
Regression analysis is the most commonly used multivariate technique to test causal models
in the social sciences.
As a general rule:
All variables in regression analysis must be measured on interval/ratio level.
One exception: dichotomous variables.
All regression analyses try to predict a Dependent Variable (DV), using one or more
Independent Variables (IVs).
Some central questions addressed by regression analysis:
How well can we predict a DV using the IV(s)?
What effect does the IV have on the DV?
How sure are we that our findings can be generalized to the population? (i.e. testing
significance).
, Week 1
Measures of Central Tendency
Calculate mean, median and, mode:
1. Analyse.
2. Descriptive Statistics.
3. Frequencies.
4. Add variables to the box.
5. Statistics.
6. Select Mean, Median, and Mode.
7. Make sure ‘’Display frequency tables’’ is selected.
8. Press OK.
Week 2
Measures of Variability
1. Analyse.
2. Descriptive Statistics.
3. Frequencies.
4. Add the variable of interest.
5. Statistics.
6. Click on all the information you want to know (Quartiles, Mean, Std. deviation, Range).
7. Continue.
8. Uncheck the box ‘’Display frequency tables’’. If you don’t uncheck it, you get too much
information you do not need.
9. Click OK.
Scatterplots
Bivariate analysis, so we need two variables for the scatterplot. The scatterplot tells us if there is a
relationship between two variables. It illustrates the relation between them. They can only be
plotted with interval or ratio variables.
1. Graphs.
2. Legacy Dialogs.
3. Scatter/Dot.
4. Define.
5. Put the independent variable on the X Axis.
When to use which test of significance – comparing means
One sample z test:
Assessing a mean against a certain value, standard deviation in population known.
Ratio variable.
One sample t test:
Assessing a mean against a certain value, standard deviation in population unknown.
Ratio variable.
Two sample t test:
Assessing whether there is a significant difference between means in two groups in the
population.
Ratio variable – DV.
Ordinal/nominal variable – grouping variable (2 groups).
ANOVA (F-test):
Assessing whether there is a significant difference between means in more than two groups
in the population.
Ratio variable – DV.
Ordinal/nominal variable – grouping variable (more than 2 groups).
Regression analysis
Regression analysis is the most commonly used multivariate technique to test causal models
in the social sciences.
As a general rule:
All variables in regression analysis must be measured on interval/ratio level.
One exception: dichotomous variables.
All regression analyses try to predict a Dependent Variable (DV), using one or more
Independent Variables (IVs).
Some central questions addressed by regression analysis:
How well can we predict a DV using the IV(s)?
What effect does the IV have on the DV?
How sure are we that our findings can be generalized to the population? (i.e. testing
significance).
, Week 1
Measures of Central Tendency
Calculate mean, median and, mode:
1. Analyse.
2. Descriptive Statistics.
3. Frequencies.
4. Add variables to the box.
5. Statistics.
6. Select Mean, Median, and Mode.
7. Make sure ‘’Display frequency tables’’ is selected.
8. Press OK.
Week 2
Measures of Variability
1. Analyse.
2. Descriptive Statistics.
3. Frequencies.
4. Add the variable of interest.
5. Statistics.
6. Click on all the information you want to know (Quartiles, Mean, Std. deviation, Range).
7. Continue.
8. Uncheck the box ‘’Display frequency tables’’. If you don’t uncheck it, you get too much
information you do not need.
9. Click OK.
Scatterplots
Bivariate analysis, so we need two variables for the scatterplot. The scatterplot tells us if there is a
relationship between two variables. It illustrates the relation between them. They can only be
plotted with interval or ratio variables.
1. Graphs.
2. Legacy Dialogs.
3. Scatter/Dot.
4. Define.
5. Put the independent variable on the X Axis.
