EDD611 Inquiry I Lab 4

Lab 4 covers multiple regression with emphasis on interpreting collinearity statistics.

Learning Objective: 

By the end of Lab 4, students will be able to:

  • Conduct a multiple regression analysis in SPSS
  • Assess and analyze regression output using collinearity statistics
  • Utilize Astin's I-E-O model to explain the relationships between variables
Expand or collapse content Getting Started

Before starting your exercises, ensure that your SPSS is open.

For instructions on how to download and/or open the SPSS software, please see the following guide: EDD611 Inquiry I Getting Started

Exercises

Expand or collapse content Multiple Regression

Step 1:

Click on Analyze, Regression, then Linear

SPSS Data Editor screen highlighting Analyze, Regression, and Linear options

Step 2:

Select the CSS Academic Self-Concept Score [ACADEMIC_SELFCONCEPT] variable

Highlight of a variable in the Linear Regression window

Step 3:

Click on the top arrow button to add it to the Dependent field

Arrow pointing to the arrow button

Step 4:

Select the Your sex: [SEX] variable

Highlight of a variable on the Linear Regression window

Step 5:

Click on the (second from top) arrow to add it to the Independent(s) field in Block 1 of 1

highlight of the arrow button

Step 6:

Repeat Steps 4 and 5 to add the following variables:

  • Race: White or Person of Color [RACE1]
  • First-generation college student [FIRSTGEN_TFS]
  • Academic Self-Concept at college entry [ACADEMIC_SELFCONCEPT_TFS]

Step 7:

Click on the Next button to move to Block 2

Arrow pointing to the Next button

The block will update to show 'Block 2 of 2'

Highlight of the 'Block 2 of 2' designation

Step 8:

Select the Discussed course content with students outside of class [ACT14] variable

Highlight of a selected variable

Step 9:

Click on the (second from top) arrow to add it to the Block 2 of 2 field

Arrow pointing to a arrow button

Step 10:

Repeat Steps 8 and 9 to add the following variables:

  • Hours per week studying/homework [HPW17]
  • Faculty provide intellectual challenge and stimulation [FACPRV16]

Step 11:

Click on the Statistics button

Arrow pointing to the Statistics button

Step 12:

Ensure the following checkboxes are checked:

  • Model fit
  • R squared change
  • Descriptives
  • Collinearity diagnostics
Highlight of the following options: Model fit,  R squared change,  Descriptives, Collinearity diagnostics

Step 13:

Click the Continue button 

Arrow pointing to Continue button

The results will populate in your Output window

Overview of the Output window with the Regression results
Expand or collapse content Collinearity Statistics Interpretation

To interpret the values, you will be analyzing any values below .40. A value of .40 or less indicates a potential multicollinearity problem. To remedy, this would require some extra analysis including removing it from the model and rerunning it. 

Step 1:

Review the Coefficients table, the Standardized Coefficients Beta column and the Sig. (P-Value) column to determine significance.

Highlight of the Coefficient and P-value (Sig.) columns

Step 2:

Review the Model Summary table, the R Square column to asses how well your factors your score.

Highlight of the Model Summary table within the  Output window

Example: 

The 7 variables explained 57.9% (R2=.579) of the total variance of the academic self-concept score.

Expand or collapse content Assignment

Utilize Astin's (202) I-E-O model to guide your analysis and write-up of the regression analysis. Connect the research question: What is the relationship with student background characteristics, college experiences, and academic self-concept? following the sample write-up provided below.

Alexander Astin's I-E-O- Model

Diagram of Astin’s I-E-O model showing Inputs, Environment, and Outputs. The model illustrates how inputs interact with the environment to influence outcomes

Astin, A. W. (2012). Assessment for excellence: The philosophy and practice of assessment and evaluation in higher education. Rowman & Littlefield Publishers.

Sample Write-up

A multiple regression analysis was conducted to examine the predictors of college students’ academic self-concept. Results show that [number] background characteristics variables of the study significantly affected survey respondents’ academic self-concept: Race (b =-.11, p<.001), first generation student (b =-.10, p<.001). This finding suggests _________________ . When it comes to college experiences, results demonstrate that [number] of variables of this study were significantly related to academic self-concept: hours per week studying or doing homework (b=.07, p<.05) and receiving intellectual challenge and stimulation from faculty (b =.14, p<.001). The results suggest that students who spend more hours in studying/homework and report more frequently receiving intellectual challenge from their faculty tend to report higher academic self-concept  compared to those who do not or do so less. The [number] variables explained 52% (R 2= .52) of the total variance of the academic self-concept score.

Format should be in narrative (paragraph) form, do not use bullet points, lists or tables in your analysis. 

Questions? Need Help?

Contact your professor: Dr. Newman at [email protected]