Multiple Regression

Multiple Regression

As with the previous week’s Discussion, this Discussion assists in solidifying your understanding of statistical testing by engaging in some data analysis. This week you will once again work with a real, secondary dataset to construct a research question, estimate a multiple regression model, and interpret the results.

Whether in a scholarly or practitioner setting, good research and data analysis should have the benefit of peer feedback. For this Discussion, you will post your response to the hypothesis test, along with the results. Be sure and remember that the goal is to obtain constructive feedback to improve the research and its interpretation, so please view this as an opportunity to learn from one another.

To prepare for this Discussion:

• Review this week’s Learning Resources and media program related to multiple regression.
• Create a research question using the General Social Survey that can be answered by multiple regression.

BY DAY 3

Use SPSS to answer the research question. Post your response to the following:

1. What is your research question?
2. What is the null hypothesis for your question?
3. What research design would align with this question?
4. What dependent variable was used and how is it measured?
5. What independent variable is used and how is it measured?
6. What other variables were added to the multiple regression models as controls?
7. What is the justification for adding the variables?
8. If you found significance, what is the strength of the effect?
9. Explain your results for a lay audience, explain what the answer to your research question.

Be sure to support your Main Post and Response Post with reference to the week’s Learning Resources and other scholarly evidence in APA Style.

As with the previous week’s Discussion, this Discussion assists in solidifying your understanding of statistical testing by engaging in some data analysis. This week you will once again work with a real, secondary dataset to construct a research question, estimate a multiple regression model, and interpret the results.

Whether in a scholarly or practitioner setting, good research and data analysis should have the benefit of peer feedback. For this Discussion, you will post your response to the hypothesis test, along with the results. Be sure and remember that the goal is to obtain constructive feedback to improve the research and its interpretation, so please view this as an opportunity to learn from one another.

To prepare for this Discussion:

• Review this week’s Learning Resources and media program related to multiple regression.
• Create a research question using the General Social Survey that can be answered by multiple regression.

BY DAY 3

Use SPSS to answer the research question. Post your response to the following:

1. What is your research question?
2. What is the null hypothesis for your question?
3. What research design would align with this question?
4. What dependent variable was used and how is it measured?
5. What independent variable is used and how is it measured?
6. What other variables were added to the multiple regression models as controls?
7. What is the justification for adding the variables?
8. If you found significance, what is the strength of the effect?
9. Explain your results for a lay audience, explain what the answer to your research question.

Be sure to support your Main Post and Response Post with reference to the week’s Learning Resources and other scholarly evidence in APA Style.

READ BEFORE POSTING! 

For an overview of multiple regression, you might want to take a look at Dr.Kathyrne’s Practical Guide to multiple regression. I’ve attached it here and I’ll also try to remember to put it in the doc sharing area.

Remember that the type of multiple regression we are learning this week is linear multiple regression (sometimes also called OLS linear regression) — it is an extension of the simple linear regression that we learned last week. That means that it is designed for variables measured at the interval/ratio level.We can get away with one dichotomous predictor variable, but more than that and we get into trouble.

Just like last week, we can think of the hypotheses in two ways — as comparing beta values or as comparing R² values to 0.

Remember to keep “collinearity” (sometimes called multicollinearity) in mind when designing your study. Avoid collinearity — it creates big problems. We’ll learn more about collinearity next week.