Scribd
Upload
18 views4 pages

wk7 MLR Output Practice

The document summarizes a multiple regression analysis that was conducted to predict alcohol consumption (ALQ) based on gender (SEX), socioeconomic group (SHTSEG), mental health (GHQ), and smoking status (SM1). The regression model was able to better predict alcohol consumption than chance alone, accounting for 17.5% of the variance in a moderate strength model. Current smoking status (SM1) was the strongest predictor of alcohol consumption. Males tended to consume more alcohol than females, and people with poorer mental health scored higher on the GHQ tended to drink more alcohol as well. However, socioeconomic group was not a significant predictor of alcohol consumption in this model.

Uploaded by

Andikaa
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
18 views4 pages

wk7 MLR Output Practice

The document summarizes a multiple regression analysis that was conducted to predict alcohol consumption (ALQ) based on gender (SEX), socioeconomic group (SHTSEG), mental health (GHQ), and smoking status (SM1). The regression model was able to better predict alcohol consumption than chance alone, accounting for 17.5% of the variance in a moderate strength model. Current smoking status (SM1) was the strongest predictor of alcohol consumption. Males tended to consume more alcohol than females, and people with poorer mental health scored higher on the GHQ tended to drink more alcohol as well. However, socioeconomic group was not a significant predictor of alcohol consumption in this model.

Uploaded by

Andikaa
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Online Practical:

Interpreting Multiple Regression Output


This exercise is designed to give you practice in extracting
and understanding the key information from the SPSS output
of an example of Multiple Linear Regression. Read the
background information, look at the SPSS output, and try to
answer the twelve questions that follow. Answers will be
provided within the Week 7 folder from Friday morning.
Background Information
An extensive survey of the United Kingdom investigated
factors predicting health related behaviours (sample ‘n’ was
in the region of six thousand respondents, but the analysed
numbers were lower due to missing values on some variables
for some respondents).
The results of one of the multiple regression analyses are
given below. The aim for this analysis was to predict alcohol
consumption (ALQ where higher values indicate greater
consumption) by using the following variables: gender (SEX
where 1= male 2=female); socioeconomic group (SHTSEG
where scoring polarity for low values indicated
professional/managerial versus high values indicated
unskilled manual); General Health Questionnaire (GHQ where
higher scores indicate worse mental health); and current
smoking status (SM1) (current smoker=1, ex-smoker= 2,
never smoked= 3).
Multiple Regression Output
Variables Entered/Removedb

Variables Variables
Model Entered Removed Method

1 sm1, agyrs, . Enter


ghqscore,
shtseg, SEXa

a. All requested variables entered.


b. Dependent Variable: alq

Model Summary

Adjusted R Std. Error of the


Model R R Square Square Estimate

1 .418a .175 .174 14.618

a. Predictors: (Constant), sm1, agyrs, ghqscore, shtseg, SEX

ANOVAb

Model Sum of Squares df Mean Square F Sig.

1 Regression 196840.334 5 39368.067 184.222 .000a

Residual 930661.469 4355 213.700

Total 1127501.804 4360

a. Predictors: (Constant), sm1, agyrs, ghqscore, shtseg, SEX

b. Dependent Variable: alq

Coefficientsa

Standardized
Unstandardized Coefficients Coefficients

Model B Std. Error Beta t Sig.

1 (Constant) 30.673 1.579 19.430 .000

SEX -11.862 .454 -.369 -26.144 .000

shtseg -.138 .178 -.011 -.774 .439

agyrs -.127 .014 -.127 -9.124 .000

ghqscore .045 .022 .028 2.028 .043

sm1 1.685 .177 .134 9.549 .000

a. Dependent Variable: alq


1. Does the model provide a better prediction of alcohol
consumption than chance alone? Please answer Yes or
No.

2. State the supporting reason (i.e. statistic) for the answer


given to the previous question.

3. What percentage of the variance in alcohol consumption


does the model account for?

4. Would the model predicting alcohol consumption be


described as: Weak, Moderate, or Strong?

5. Which variable is the strongest predictor of alcohol


consumption?

6. For the previous question, give an account of your


reasoning for choosing this variable as the strongest
predictor of alcohol consumption.
7. Do males tend to consume more alcohol? Please
answer Yes or No.

8. For the previous question above, give an account of


your reasoning.

9. Do people with poorer mental health (i.e. higher GHQ


scores) tend to drink more alcohol? Please answer Yes
or No.

10. For the previous question above, give an account of


your reasoning.

11. Is socioeconomic group a significant predictor of


alcohol consumption in this model? Please answer Yes
or No.

12. For the previous question, give an account of your


reasoning.

You might also like