Linear Regression 1

Week 14
Interpreting Linear Regression Output in SPSS
The best predictor of future behavior is past behavior.

Computing Linear Regression with SPSS

EXAMPLE: Job Satisfaction.sav
• JS: Job satisfaction score • TOI: Turnover intention score
• BO: Burnout score • YearsinMH: Years in mental health field
• CF: Compassion fatigue score • CS: Compassion satisfaction score
Simple Linear Regression
Dataset: Job Satisfaction.sav
Graphs ! Chart Builder
In the Gallery, click on Scatter/Dot: drag Simple Scatter (top row, left) into the canvas
Simple Scatterplot – Burnout = X Axis; Job Satisfaction = Y Axis; OK
Double-click on the scatterplot to open the Chart Editor; click on Add fit line at total;
close the Chart Editor
Analyze ! Regression ! Linear…
Move the variable Job Satisfaction into the Dependent: window;
Move the variable Burnout into the Independent(s): window;
Method should be set to Enter;
Statistics – check Confidence intervals; Descriptives; Durbin-Watson; and Casewise
diagnostics; Continue
Plots – Move ZPRED to X: & Move ZRESID to Y: & Click on Histogram and Normal
probability plot; Continue; OK

© 2019 Todd Daniel, Ph.D., Research by Design, LLC

 Linear Regression 2

Descriptive Statistics
Mean Std. Deviation N
Job Satisfaction 132.17 29.319 200
Burnout 48.9096 9.02511 200

Correlations
Job Satisfaction Burnout
Pearson Correlation Job Satisfaction 1.000 -.650
Burnout -.650 1.000
Sig. (1-tailed) Job Satisfaction . .000
Burnout .000 .
N Job Satisfaction 200 200
Burnout 200 200
Variables Entered/Removed
Variables
Model Variables Entered Removed Method
b
1 Burnout . Enter
a. Dependent Variable: Job Satisfaction
b. All requested variables entered.

Model Summary
Adjusted R Std. Error of the
Model R R Square Square Estimate Durbin-Watson
a
1 .650 .423 .420 22.334 2.073
a. Predictors: (Constant), Burnout
b. Dependent Variable: Job Satisfaction

ANOVA
Model Sum of Squares df Mean Square F Sig.
b
1 Regression 72296.564 1 72296.564 144.940 .000
Residual 98762.991 198 498.803
Total 171059.555 199
a. Dependent Variable: Job Satisfaction
b. Predictors: (Constant), Burnout
Coefficients
Standardized
Unstandardized Coefficients Coefficients 95.0% Confidence Interval for B
Model B Std. Error Beta t Sig. Lower Bound Upper Bound
1 (Constant) 235.459 8.724 26.990 .000 218.255 252.663
Burnout -2.112 .175 -.650 -12.039 .000 -2.458 -1.766
a. Dependent Variable: Job Satisfaction

Residuals Statistics
Minimum Maximum Mean Std. Deviation N
Predicted Value 84.81 170.58 132.17 19.060 200
Residual -57.890 51.305 .000 22.278 200
Std. Predicted Value -2.484 2.015 .000 1.000 200
Std. Residual -2.592 2.297 .000 .997 200
a. Dependent Variable: Job Satisfaction

© 2019 Todd Daniel, Ph.D., Research by Design, LLC

 Linear Regression 3

Histogram of dependent variable P-P Plot of residual errors Scatterplot of predicted and residual errors

APA Write-up for Bivariate Linear Regression:

A linear regression was conducted to examine whether burnout could predict level of job
satisfaction. A scatterplot showed that the relationship between burnout and job satisfaction was negative
and linear and did not reveal any bivariate outliers. An analysis of standard residuals showed that the data
contained no outliers (Std. Residual Min. = -2.59, Std. Residual Max. = 2.30). Independence of residual
errors was confirmed with a Durbin-Watson test (d = 2.073). Residual plots showed homoscedasticity and
normality of the residuals.
Burnout statistically significantly predicted job satisfaction, F (1, 198) = 144.94, p < .001,
accounting for 42.3% of the variability in job satisfaction with adjusted R2 = .42%. This is a moderately
strong relationship (Cohen, 1988). The correlation between burnout and job satisfaction was statistically
significant, r(198) = .65, p < .001. The regression equation for predicting the job satisfaction from burnout
was ŷ = 235.50 – 2.11x (burnout). The confidence interval for the slope to predict job satisfaction from
burnout was 95% CI [-1.766, -2.458] with a B = -2.11; thus, for each one unit of increase of burnout, job
satisfaction decreases by about 1.8 to 2.5 points. Predictions were made for mean job satisfaction among
mental health workers with burnout scores of 40, 49, and 58 (M ± 1 SD). For burnout scores of 40, mean
job satisfaction was predicted as 151.1; for burnout scores of 49, mean job satisfaction was predicted as
132.11; and burnout scores of 58, mean job satisfaction was predicted as 113.12 points.

*APA TIP: Do not write the correlation as 0.65 because the r value cannot be greater than ±1

References
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence
Earlbaum Associates.
Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York, NY: McGraw-Hill.
Spector, P. E. (1997). Job satisfaction: Application, assessment, causes, and consequences. Thousand
Oaks, CA: Sage.
Stamm, B. H. (2010). The concise ProQOL manual (2nd ed.). Pocatello, ID: ProQOL.org.
All IBM® SPSS® Statistics software screen reprints courtesy of International Business Machines Corporation, © International Business Machines Corporation.

© 2019 Todd Daniel, Ph.D., Research by Design, LLC