Stat4all can upload data from different types of files, including files from SPSS, SAS, MS Excel, STATA, and other statistical packages. (If you don't have a data set, you can use one of the example data sets to explore how Stat4all works.)
If you still need to prepare your data set before you can analyze it, you can use Stat4all to label categorical variables, standardize numerical variables, create composite variables, and perform various other operations on your data set.
Examine the distributions of your numeric and categorical variables with bar plots, density plots, and histograms; inspect the distribution of numeric variables with Q-Q plots; examine the relationship between variables using scatter plots.
When you compare groups or pre-/post-test measurements, Stat4all will automatically test parametric assumptions and choose the statistical tests accordingly. If you examine correlations between variables and assume a linear relationship, an outlier analysis graph will be included in the report. When you build simple and multiple regression models, a model diagnostic is performed if statistical significance is found.
For most analyses, you won't need to change the default analysis settings. However, you still have some control over the p-value used for statistical significance cut-off, test used to assess normality, and preference of parametric tests in large sample sizes.
No registration required.