such as serving on their local county Farm Bureau boards. Leadership is a function of
these variables and this regression analysis will determine which variables are the
greatest predictors of why individuals chose to participate or not participate in leadership
roles in their local county Farm Bureau boards.
There are several procedures available for selecting the independent variables in
the multiple regression equation. Stepwise selection is the most commonly used method
(Ary et al., 1996). A type of stepwise selection, a backward selection, is a method of
multiple regression in which each time a predictor is added to the equation, a removal test
is made of the least useful predictor. The regression equation is constantly being
reviewed to see whether any redundant predictors can be removed. The first step in this
model is placing all the predictors in the model and calculating the contribution of each.
If a predictor meets the criteria for removal, if it is not making a statistically significant
contribution, then it is removed from the model and then the model is recalculated for the
remaining predictors (Field, 2000).
The first variable considered is the one with the largest positive or negative
correlation with the criterion, this correlation is the Pearson correlation and is the second
table (after the descriptive statistics table) given in a multiple regression analysis using
SPSS 12.0 for Windows (Ary et al., 1996). Agresti and Finlay (1997) state that the
larger the absolute value of r, the Pearson correlation, the stronger the degree of linear
association between the independent and dependent variable.
In a multiple regression analysis, there are several other variables that need to be
considered when deciding if a dependent variable may be used to predict an independent
variable. The value j or Beta value is the second variable that is examined. It is reported