How to report your regression results in table form
A table that presents regression results should contain the following elements:
1. the number of observations
2. the constant
3. the adjusted R square
4. the unstandardized coefficients for each independent variable, with the level of statistical significance indicated by stars (*) and standard error in parentheses
5. the standardized coefficients (t value in parentheses)
Examine the following SPSS output for a regression concerning support for AOW cuts. There are five independent variables, two of which are statistically significant.
Model Summaryb |
---|
Model | R | R Square | Adjusted R Square | Std. Error of the Estimate |
---|
1 | .295a | .087 | .075 | .73484 |
---|
a. Predictors: (Constant), HOE VAAK GAAT U NAAR DE KERK, chancepension, income, LEEFTIJD, leftright |
---|
b. Dependent Variable: AOWcuts | |
Coefficientsa |
---|
Model | Unstandardized Coefficients | Standardized Coefficients | t | Sig. | 95% Confidence Interval for B | Collinearity Statistics |
---|
B | Std. Error | Beta | Lower Bound | Upper Bound | Tolerance | VIF |
---|
1 | (Constant) | 2.245 | .272 | | 8.259 | .000 | 1.711 | 2.780 | | |
---|
chancepension | -.033 | .022 | -.073 | -1.488 | .137 | -.076 | .011 | .962 | 1.039 |
---|
leftright | .104 | .035 | .153 | 3.005 | .003 | .036 | .172 | .896 | 1.116 |
---|
income | .038 | .051 | .037 | .748 | .455 | -.062 | .139 | .942 | 1.061 |
---|
LEEFTIJD | -.013 | .003 | -.210 | -4.202 | .000 | -.020 | -.007 | .925 | 1.081 |
---|
HOE VAAK GAAT U NAAR DE KERK | -.006 | .019 | -.015 | -.292 | .770 | -.043 | .032 | .892 | 1.121 |
---|
a. Dependent Variable: AOWcuts | | | | | | | | |
The results are presented in the following table. Notice that the dependent variable is part of the title of the table. The independent variables are listed in the first column, with statistical significance indicated with stars. At the bottom of the column, after the independent variables are: the constant, the R square and number of observations.
NB: not all researchers include the standard error and t values in their regression tables. We think it is a good idea to do this, because some readers will want to see these.
Tabel 1: OLS Regressie aangaande steun voor AOW verlaging
| B (standard error) | Beta (t value) |
kans pensioen | -.033 (.022) | -.073 (-1.488) |
links-rechts | .104** (.035) | .153 (3.005) |
inkomen | .038 (.051) | .037 (.748) |
leeftijd | -.013*** (.003) | -.210 (-4.202) |
frequentie kerkgang | -.006 (.019) | -.015 (-.292) |
| | |
constant | 2.245 | |
adjusted R2 | .075 | |
n=401 | | |
*=p<.05; **=p<.01; ***= p<.001
NB: in order to find out the number of observations, you need to request “descriptives” when you run the regression.
Add new contribution