Should I use MANOVA, ANOVA for Regression? I'm currently developing a research proposal and I'm looking at the variables values orientation and environment as determinants of pro-environmental behavior. Is it right to use ANOVA or MANOVA? All variables are continuous in nature. Further, I want to see each independent variables' relationship with the dependent variable and the interaction among them. 
 A: It sounds more like a multiple regression question.
All three of MANOVA, ANOVA, and regression fit into the heading of the general linear model. ANOVA is a special case of the linear model where you have categorical predictor variables. Since you have continuous predictor variables you would model them as 
DV~ X1 + X2 + X3
Hope this helps. 
A: The difference between ANOVA and MANOVA is merely the number of dependent variables fit. If there is one dependent variable then the procedure is ANOVA, if two or more dependent variables, then MANOVA is used.
1-way ANOVA assumes: Two or more independent samples measured on a continuous scale. 
Samples are from a population with a normal distribution and must have the same variance, also known as homogeneity of variance. 
2-way ANOVA assumes: A sample measured on a continuous scale, observed for two ordinal or nominal scale factors. 
Samples are from a population with a normal distribution and must have the same variance, also known as homogeneity of variance. 2-way ANOVA is performed to understand if there is an interaction between the two independent variables on the dependent variable.
MANOVA: The obvious difference between ANOVA and a "Multivariate Analysis of Variance" (MANOVA) is the “M”, which stands for multivariate. In basic terms, A MANOVA is an ANOVA with two or more continuous response variables. Like ANOVA, MANOVA has both a one-way flavor and a two-way flavor. The number of factor variables involved distinguish a one-way MANOVA from a two-way MANOVA. 
Further reading that explains the differences you asked about appears on this site.
With regards to your question as to relevance of parameters, some, but hardly all ANOVA software will calculate how significant leaving out each parameter would be:
 
Typically one tests for interaction terms between predictor variables. With respect to interaction terms, the procedure followed depends on the specific results of testing. For example, in SPSS software, a typical procedure might involve General Linear Models repeat measures, which model allows for testing of repeat tests, thus correlated data, provided that we correct for any sphericity. Mauchly's Test of Sphericity if significant would give us reason to degree of freedom correct the F-statistic of the repeat measures with the interaction test. This might be done with the "Tests of Within-Subjects Effects" and attention to the interaction probability of Greenhouse-Geisser testing. Since this is a multiple if...then structure, there is no reasonable substitute for working through a chain of results.
