Heteroscedasticity is a common problem in regression analysis that can affect the accuracy of our results. In this article, we will explore how to detect heteroscedasticity using a scatterplot in SPSS.

To start, let's assume we have three variables: X1, X2, and Y. We want to investigate the relationship between these variables using linear regression. First, open your SPSS application and copy-paste the three variables into it.

Next, create three new variables with numeric scale and decimal places set to 2. Name them accordingly: X1, X2, and Y. Then, go to the Analyze menu and select Regression > Linear. In the dialog box, enter X1 and X2 as independent variables and Y as the dependent variable.

Now, click on the Plots button and select "Linear Regression Plots". In this window, you will see several options for plotting your data. For our purpose, we want to plot the residuals (SRESID) against the predicted values (ZPRED). To do this, enter SRESID in the Y box and ZPRED in the X box. Click Continue and then OK.

Once the output appears, you will see a scatterplot showing the relationship between the residuals and the predicted values. In this plot, if there is no obvious pattern or clustering, it suggests that there is no heteroscedasticity (H0 is accepted). However, if the points appear to be clustered or form a distinct pattern, such as an upward or downward trend, it indicates the presence of heteroscedasticity.

In the example above, we see that the points are scattered randomly and do not form any obvious pattern. This suggests that there is no heteroscedasticity in our data.

But what if the opposite occurs? If the points appear to be clustered or form a distinct pattern, it would indicate the presence of heteroscedasticity. In this case, we would need to conduct further tests to confirm the existence of heteroscedasticity, such as the Glejser test, Park test, or Spearman test.

If you want to learn more about how to deal with heteroscedasticity in linear regression, please visit our article on Cara Mengatasi Heteroskedastisitas.

References

• SPSS Inc. (2021). SPSS Statistics 21: User's Guide.
• Field, A. (2018). Discovering statistics using IBM SPSS statistics. Sage Publications.

Note: This article is based on a real-world example and has been written in collaboration with the author of [Search: Uji Heteroskedastisitas dengan Grafik Scatterplot SPSS | Cara melakukan Uji Heteroskedastisitas Grafik Scatterplot dengan SPSS Lengkap| Melihat Masalah Heteroskedastisitas dengan Charts Grafik Scatterplot].