One-way ANOVA, multiple regression, paired-sample T-test and logistic regression are all methods used for data analysis. Each one corresponds to data with specific characteristics so that the results of the analysis are as accurate as possible. Analysis using these methods can be conducted either manually on paper or using statistics software on a computer.
Multiple regression is one of the most common methods used for data analysis. The general purpose of this type of analysis is to examine the relationship between several independent variables and one dependent variable, known as the predictor. For the multiple regression test there are several assumptions required, one of which is that the residuals of the data are normally distributed.
Logistic regression is a similar model, however it is used to predict variables with a dichotomous outcome. This analysis can handle even non-linear relationships, as opposed to the other regression models. Nevertheless, the predicted variable needs to be binomial for this analysis to apply.
One-way ANOVA is used to test for significant differences between means. In order to check for such differences in means, the model compares variances. It can be applied between two or more samples. Typically, thought, ANOVA is used to test for differences among at least three samples, since the two-samples case is covered by the t-test model.
A paired-sample t-test is used to compare two population means in cases where the two samples are correlated. It is a method used commonly in "before-after" studies. To conduct this type of analysis, independent samples are needed and only matched pairs in the data can be used.