TUTORIAL UJI ASUMSI KLASIK VARIABEL MODERASI DENGAN SPSS -TERBARU ❗❗❗

Summary

In this tutorial, the presenter demonstrates how to conduct classical assumption tests in multiple regression with a moderating variable. Key tests covered include normality, multicollinearity, and heteroscedasticity. The video explains the process of running these tests in SPSS and interpreting the results, emphasizing the importance of checking assumptions before proceeding with regression analysis. It also touches on handling non-normal data and provides an easy-to-understand guide for those working with moderated regression models. The video aims to clarify common questions from viewers about the inclusion of a moderating variable in assumption testing.

Takeaways

• 😀 The video demonstrates how to conduct assumption tests in multiple linear regression with moderation variables.
• 😀 Assumption tests are essential prerequisites for performing regression analysis.
• 😀 The script explains the process for testing assumptions in models with both independent and moderation variables.
• 😀 Key assumptions to be tested include normality, heteroskedasticity, multicollinearity, and autocorrelation.
• 😀 Autocorrelation is not tested in this case, as the data used is not time series data.
• 😀 To conduct assumption tests in SPSS, the variables must be entered correctly into the dataset.
• 😀 The process includes setting up the dataset with independent variables (X1, X2), a moderation variable, and a dependent variable (Y).
• 😀 The script emphasizes checking for multicollinearity by evaluating tolerance values and Variance Inflation Factor (VIF).
• 😀 Normality is tested using the Kolmogorov-Smirnov test, and the data is considered normally distributed if the significance value is greater than 0.05.
• 😀 Heteroskedasticity is tested using the Glejser method, and values greater than 0.05 indicate that heteroskedasticity is not present.
• 😀 The video aims to clarify the process of assumption testing with moderation variables in regression models, and invites viewers to subscribe for more tutorials.

Q & A

What is the main topic of the video?

The video explains how to perform classical assumption testing using moderation variables in multiple regression analysis.

What are the classical assumptions in multiple regression?

The classical assumptions include normality, heteroskedasticity, multicollinearity, and autocorrelation.

How does the presence of a moderation variable affect assumption testing?

The presence of a moderation variable does not change the basic process for assumption testing in multiple regression, but it is an additional factor to consider when analyzing relationships between the independent and dependent variables.

What is the role of the moderation variable in the model?

The moderation variable either strengthens or weakens the relationship between the independent variables (X1, X2) and the dependent variable (Y).

What tool is used to perform the assumption tests in the video?

The video uses SPSS to perform the assumption tests.

How is multicollinearity tested in SPSS?

Multicollinearity is tested by looking at the tolerance and Variance Inflation Factor (VIF) values. A tolerance value greater than 0.1 and a VIF less than 10 indicate no multicollinearity.

What does it mean if the Kolmogorov-Smirnov test for normality is significant?

If the Kolmogorov-Smirnov test results in a p-value smaller than 0.05, it suggests that the data does not follow a normal distribution.

What is heteroskedasticity and how is it tested?

Heteroskedasticity refers to the situation where the variance of residuals is not constant across observations. It is tested using the Glejser test, where a significant p-value (less than 0.05) indicates the presence of heteroskedasticity.

What happens if the assumption tests fail?

If the assumption tests fail, such as if the data does not meet the normality or heteroskedasticity assumptions, it may be necessary to transform the data or use alternative regression methods.

Why is the autocorrelation test not used in this video?

The autocorrelation test is not used in this video because the data is not time series data, which is required for this specific test.