Nội dung text SPSS - 3 - Regression, t-Test, ANOVA
SPSS manual guide for early researchers | Lido When we run a regression in SPSS, we are trying to answer three main questions: Q1: Is the model a good fit? (Model Fit): Does our model, as a whole, actually explain anything? We look at the R (R-square) value. ranges from 0 to 1. It tells us the percentage of variation in the dependent 2 R 2 variable (Y) that is explained by our independent variables (the X's). For example, an Adjusted R of 0.60 means that 60% of the change in our dependent variable can 2 be explained by the independent variables in our model. The other 40% is unexplained (it's the error term, e). Q2: Are the predictors significant? (Hypothesis Testing): Do our independent variables actually have a statistically significant effect on the dependent variable? We will look at p-values for the overall model and for each individual coefficient (the t-test). Q3: What is the relationship? (Coefficients): If an independent variable is significant, what is its effect? Is it positive or negative? How strong is that effect? We will look at the β coefficients to find out. 1.3 SPSS steps Analyze > Regression > Linear Move all IV factors to “Independent(s)” and DV factor to “Dependent” box
SPSS manual guide for early researchers | Lido