Simple linear regression
Description: Standard (least-squares) simple regression analysis is carried out to test the predictive relationship between a single independent (predictor) variable and a single dependent measure. The dependent measure is measured (or at least assumed continuous). In most applications, the independent variable is continuous. However, it is also possible for your independent variable to be a binary variable. The model assumes (a) a linear relationship between the independent and dependent variables, (b) the absence of outliers and influential cases, (c) the residuals for the model are randomly and independently sampled from a population distribution that is normal and exhibits constant variance across levels of the independent variable.
For those of you who are interested, here's a fun fact! When the independent variable is binary, the test of the regression slope is effectively a test of the difference in means on the dependent variable. Thus, the test of the regression slope is the same test of the difference in means as the independent samples t-test. If you dummy code your IV (0 & 1), then the intercept of the model is equal to the mean for the group coded 0. Adding the intercept and the slope will provide you with the mean of the group coded 1.
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Introduction to simple linear regression in SPSS (June 2020): Powerpoint, SPSS data , video demo
-This presentation provides:
Basic overview of simple linear regression concepts
Hypotheses and notation
Demonstration of how to run simple linear regression in SPSS
Review of model assumptions
Demonstration of strategies for evaluation of assumptions, including scatterplots, residuals plots, and influence statistics. It also provides a demo of evaluation of non-linearity using hierarchical linear regression using a series of polynomial regressions.
Simple linear regression assumptions / A deeper dive (July 2020): Powerpoint , SPSS file
-This presentation provides a deeper overview of strategies for assessing assumptions. It also includes a demonstration of how to obtain heteroskedasticity-consistent standard errors.
Computing slope, intercept, and R-square in simple linear regression (July, 2019): video, Excel spreadsheet for demo
-This is provided for anyone who wants to know a bit more about 'where the numbers come from'.
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Below are several older videos and Powerpoints. Quite frankly, they are not as of high quality as the above presentations. However, I'm leaving these up in case you would like more practice or review:
Introductory concepts and demonstration in SPSS (video), Powerpoint, SPSS data
Interpreting unstandardized and standardized regression coefficients, SPSS data file
More on interpreting simple linear regression output in SPSS, SPSS data
Understanding simple regression: Notation, computations, and interpretations: Powerpoint, SPSS data
Evaluating simple linear regression assumptions, outlier detection, and identification or potential influential cases using SPSS: video, Powerpoint, SPSS data