X Variable

An independent variable in regression analysis is termed an X variable because it is usually depicted graphically along the horizontal axis (“x-axis”) of a scatterplot. It is sometimes termed a right-side variable, in that it appears on the righthand side of the regression equation. The X variable represents the presumed cause in a cause-and-effect relationship, whereas the Y variable denotes the effect.

Take a standard regression equation of the form Y = a + bX. The X on the right-hand side of the equation represents an independent variable. Consider an analysis of the effect of population density on crime rates in cities. The population density would be the X variable, which might be found to affect the crime rate, the Dependent Variable.

Similarly, in a difference of means test, the X variable is the dichotomous variable for which the mean of the dependent variable is calculated. Inlooking to see whether men and women differ significantly in their vote turnout, for example, gender is the X variable. The turnout rates of men and women would be calculated, and then a difference-of-means test would be used to check whether that difference is statistically significant.

Generalizing to cross-tabulations, percentages generally are taken within categories of the X variable, with comparisons made across categories of the X variable in order to gauge the effect of the independent variable on the dependent variable. In comparing the rates of church attendance of different religious denominations, for example, denomination would be the X variable. The proportion of Catholics who go to church regularly might be compared to the proportion of Protestants and to the proportion of adherents of other religions who go to religious services regularly.