The general process for dealing with a dataset is as follows.
- The first step is to understand the data:
- Check for missing values
- Check for normality
- Check for outliers
- Check for errors
- Depending on the data, the next step is to consider transformations
- Next is to find the distributions:
- Look at the variance and correlations
- It is also important to note if any of the items are reverse-coded
After getting a sense of the dataset and understanding it, the analysis follows. These are some of the techniques that were discussed in class:
- Covariance Structure: Principal Components and Fator Analysis
- To reduce the number of variables
- ANOVA, MANCOVA, Chi Square, Hierarchical Regression Analysis
- To understand differences among group means
- Moderators and Mediators
- To better understand relationships among variables
- Survival Analysis
- When the question is about success/failure
- Structural Equations
- For elaborate models
- Panel Data
- For longitudinal data