Article directory
Continued from the previous article:
【Questionnaire analysis】Operation of moderation effect test①
2.2 When the moderator variable is a categorical variable
We will use the moderator variable in this model as an example to explain.
This moderator variable is tourism goal orientation (learning vs performance).
According to this model, in the actual data collection process, we have divided the data into four groups, according to the two-factor between-group design of travel frequency (higher vs. lower) X goal orientation (learning orientation vs. performance orientation).
(1) Determining the division of groups
Independent-samples t-tests were used to detect differences between moderator variables and independent variables to ensure that participants could effectively discriminate between travel frequency and travel-goal learning-oriented stimulus materials.
(2) Check the adjustment effect value
The steps here are the same as the interaction
. Select analysis in SPSS, then select general linear model, univariate, and then enter variables in turn.
As shown in the figure below, put the independent variable and the adjusted variable in the left box into the horizontal axis and a separate line respectively, and add them once to form two lines of the independent variable, the adjusted variable, and the adjusted variable .
As shown in the figure below, in the options, select the last interactive option, import it into the box on the right, and check the corresponding option. After the setup is complete, analysis can be performed.
(3) Interpretation of results
Look up the table in the data results, the red box is the independent variable-adjustment variable-interaction. But we only need to pay attention to the significance level corresponding to the interaction line, if it is less than 0.05, there is a significant moderating effect. Data from this table can be used for reporting in papers.
You also need to pay attention to the univariate test table given by SPSS. Locate the table with the moderating variable placed in the first column. It can be observed that the significance of the independent variable on the dependent variable is different under different groups of moderator variables. Therefore, when reporting data, we need to find out the mean value of independent variables under different groups according to the estimated value table given by SPSS to form a report form.
We can also use Excel to replicate the means of the four dependent variables and form a graph to show the moderating effect.