Assignment 6: Bivariate and Multiple Regression
Using the class_survey dataset, you are going to address the following research
questions:
1. Do scores on the BSSS predict alcohol use?
State your null and research hypothesis.
Conduct regression analysis in SPSS to determine if scores on the BSSS Total
(BSSSTotal) predict alcohol use (Alcohol). Write up the findings from the SPSS
output in paragraph format. Report and interpret the r-squared value,
unstandardized and standardized regression coefficients (direction, change in Y
given X, significance – see below for examples).
Dependent variable = Alcohol
Independent variable/predictor = BSSSTotal
State your conclusion in terms of the null hypothesis.
2. Do scores on the BSSS predict alcohol use controlling for age, gender, ethnicity,
cigarette use, and callous-unemotional traits?
State your null and research hypothesis.
Based on prior research, we believe that age, gender, ethnicity (Latino vs. Non-
Latino), cigarette use, and CU traits (ICUTotal) may also be important in
predicting alcohol use. Conduct multiple regression in SPSS controlling for these
variables to determine if the association between the BSSS total and alcohol use
still holds.
Write up the findings from the SPSS output in a single paragraph (do not use bullet
points). Your paragraph should include: (1) interpretation of the r-squared value,
(2) which variables are and are not significant, (3) report unstandardized and
standardized regression coefficients for significant variables only (direction,
change in Y given X, significance – see below for examples), and (4) discuss the
relative effects for any significant predictors (i.e., compare the effects using the
standardized regression coefficients). Statistics should be reported in the proper
format (see examples provided below).
Dependent variable = Alcohol
Independent variables/predictors = BSSSTotal, Age, Gender, Ethnicity, Cigarettes,
and ICUTotal.
State your conclusion in terms of the null hypothesis.
Submit your hypotheses and write-ups – I do not need your spss output.
Reporting regression:
Make sure to interpret the R Square value as the percent of the variance explained in the
outcome by all variables included in the model and report the F-statistic when reporting
significance.
o Examples:
In a bivariate regression model, self-control predicted
23% of the variance in delinquency, R 2 = .23, F(2, 108) =
4.47, p = .04.
Age, gender, race, and self-control were included in the
model predicting delinquency. Together the predictors
explained 33% of the variance in delinquency, R 2 =
.33, F(4, 108) = 5.67, p = .03.
When interpreting the unstandardized regression coefficients make sure to provide the
proper interpretation:
o Example: Self-control had a significant negative impact on
delinquency (B = -.673, p = .002). Specifically, for every one unit
increase on the self-control scale, there was a .673 decrease in
self-reported delinquency.
Make sure to interpret the standardized regression coefficients in terms of standard
deviation units and to report the standardized regression coefficients when discussing the
relative magnitude of their effects on the outcome:
o Example: Although both were significant predictors of delinquency,
the effect of self-control (b = -.51, p = .002) was stronger than the
effect for age (b = .23, p = .03). That is, for every one standard
deviation increase in self-control there was a .51 standard deviation
unit decrease in delinquency compared to only a .23 standard
deviation increase associated with a one unit standard deviation
increase for age.