In my statistics class, we reviewed simple and multiple linear regression analysis and prediction. We learned that linear regression analyses are exciting, flexible, and powerful methods of data analysis that can also function as theoretical frameworks for many clinical decisions.
Regression analysis works best when variables are highly correlated, and is a way of predicting an outcome or dependent variable from one or more predictor or independent variables. When one independent variable is used to predict one dependent variable, the method is called simple linear regression. When several predictor or independent variables are used to predict a single dependent variable, the method is called multiple or factorial regression. Regression is incredibly useful because it allows data to be used to make predictions. . Multiple regression with more than two predictor variables offers tremendous flexibility in modeling real-world theories studied by psychology professionals.
I need help coming up with two interesting questions (and answers) relating to the overviewed topics.
I am think two questions now.
1. When we do the simple regression analysis, we may find the the prediced value of dependent variable using the regression equation is always far from the actual data. In other words, the residual is always very large. We are told that the sample size of our data is too small. So the qustion is: I wonder how large the sample size must be in order to obtain better ...
The solution gives two hard questions that involve performing regression analysis. B