Regression techniques for analysis of variance with application to the reduction on serum cholesterol level

الملخص EN

In this study both of linear regression model and analysis of variance (ANOVA) for completely randomized design (CRD) with equal replications ni were used.

In fact, it can describe ANOVA method as a special case of the regression model and access to same results with dummy variables which are encoded as an indicator of (0, and 1) and then used directly for the solution of the linear system described by a linear regression.

To prove and validate this fact regression model and one-way analysis of variance (ANOVA) were applied to the field of biological experiment diets effect to the reduction on serum cholesterol level (in mg/100 ml) for 32 normal men to test whether the means of all diets are significantly different.

Concluding that multiple regression analysis is similar to analysis of variance according to the equivalency of the results achieved from the two studied models.

Since, the null hypothesis was accepted, there was insufficient evidence to show that the means are not equal.

It can be said that the diets studied has no an effect on the serum cholesterol level in normal males.