Creates Basic Logistic Regression Writing Paper

By /

Creates Basic Logistic Regression Writing Paper

Section A

In a 2- to 3-paragraph post, construct a persuasive argument for the value (conditional probability, odds, odds ratio, etc.) that, intuitively, makes the most sense for you to report as a result to your audience. Be sure to provide a specific rationale for your choice.

Section B

n this assignment, you use Microsoft Excel to construct a specialized tool that creates basic logistic regression models given a crosstab/contingency table. As if that were not useful enough, this Excel tool is not specialized—you can use it given any crosstab/contingency tables you encounter in research. In the field of statistical research, this is just about as exciting as you can get!

The Assignment

Using one of the datasets provided, select two variables that allow you to construct a 2×2 contingency table. Use SPSS to run the initial crosstab table, using any two variables that you think are appropriate. Then, use Excel to construct a table in which you report:

  • Conditional probabilities
  • Conditional odds
  • Logits
  • Odds ratios
  • Relative risk
  • Slope

Be sure to apply the template from the Osborne text. Note that page 42 has a completed example that should help you determine these values. Be sure to use formulas and cell references in Excel so that the spreadsheet you create can be used as a tool for calculating similar values for other datasets.

Once you have created the tool, write a 1- to 2-paragraph summary in APA format interpreting your results. Submit both your Excel file and your summary to complete this assignment.

Resource

Osborne, J. W. (2015). Best practices in logistic regression. Thousand Oaks, CA: SAGE Publications.

Best Practices in Logistic Regression, 1st Edition by Osborne, J. Copyright 2015 by Sage College. Reprinted by permission of Sage College via the Copyright Clearance Center.

These chapters provide simple examples that demonstrate important aspects of logistic regression, including how logistic regression is distinct from ordinary least squares (OLS) regression and how it is much more effective than OLS regression in predicting dichotomous variables.

Chapter 1, “A Conceptual Introduction to Bivariate Logistic Regression” (PDF)