Multivariate analysis refers to statistical techniques that look at a number of variables simultaneously, with a view to clarifying the relationships between them. Many of them try to build mathematical models that can be used for prediction. The general linear model is not a discrete statistical technique in itself, but rather a strategy for analysis. Its goal is to determine whether and how one or more independent variables relate to or affect one or more dependent variables, assuming that the relationships between them are linear. It is an umbrella concept, that in addition to simple linear regression and one-way analysis of variance (ANOVA), includes several multivariate techniques. Multiple linear regression models a situation where a single numerical dependent variable is to be predicted from multiple independent variables. Logistic regression is an analogous technique that is used when the outcome variable is dichotomous in nature. In this, the natural logarithm of odds ratios is modeled as a linear function of the explanatory variables. The log-linear technique models count type of data and can be used to analyze cross-tabulations where more than two variables are included. It can look at three or more variables at the same time to determine associations between them and also show just where these associations lie. Factor analysis and principal component analysis are a group of related techniques that seek to reduce a large number of predictor variables to a smaller number of factors or components, that are linearly related to the original variables. Analysis of covariance (ANCOVA) is an extension of ANOVA, in which an additional independent variable of interest for which data is available, the covariate, is brought into the analysis. It tries to examine whether a difference still persists after ‘controlling’ for the effect of the covariate that can impact the numerical dependent variable of interest. MANOVA and MANCOVA, are multivariate extensions of ANOVA and ANCOVA respectively, and are used when multiple numerical dependent variables have to be incorporated in the analysis.