Statistical Procedure of Linear Regression INTRODUCTION The purpose of this researc

Administrators and managers of organizations use statistics in three general ways. These three ways are as follows:

1. To describe events. Descriptive statistics describe the performance or activity of one group or class, without attempting to make generalizations about other groups or classes.1

2. To infer causes or future events. Inferential statistics permit the findings with respect to one set of relationships to be extended to other relationships and to generalize findings and conclusions on the basis of statistical inference.2

1R. I. Levin, Statistics for Management, 3rd ed. (Englewood Cliffs, New Jersey: PrenticeHall, Inc., 1984), 12.

3. To enhance the decision making process. Decision theory provides administrators and managers with knowledge about events and relationships which reduce the level of uncertainty in the data upon which decisions for an organization are based.3

The basic theoretical principle upon which inferential statistics are based is that of probability.4 Probability is "a number expressing the likelihood of occurren

. . .
for a dependent variable; (2) to control for confounding variables so as to better evaluate the contribution of other variables; and (3) to test explicit causal theories. Multiple regression is used as an inferential tool in hypothesis testing. In the testing of hypotheses, multiple regression provides the data by which researchers may either reject or not reject the null forms of the hypotheses being tested. If the null form of an hypothesis is rejected, then, by inference, the hypothesis may be accepted; and, conversely, if the null form of an hypothesis cannot be rejected, then, by inference, the hypothesis may be rejected. It is held that before a regression equation can be used for causal analysis, a theoretical basis for a causal relationship between the variables concerned must be established.11 __________ 11H. M. Blalock, Jr., Causal Inferences in Nonexperimental Research, 2nd ed. (New York: W. W. Norton & Company, Inc., 1982), 4344. 6 It has also been pointed out that the more causal variables that are included in a regression equation to be used in causal analysis, "the simpler our assumptions must be concerning the manner in which the variables are combined"
. . .

Some common words found in the essay are:
Economic Advisers, REGRESSION ILLUSTRATION, TESTING HYPOTHESES, Steinbrenner Bent, Irwin Inc, Company Inc, Book Company, PrenticeHall Inc, Weekly Wage, STATISTICS Administrators, regression analysis, weekly wage, unemployment rate, dependent variable, average weekly, average weekly wage, linear regression, 3rd ed, multiple regression, regression equation, independent variable, linear regression analysis, methods 3rd ed, analysis sampled data, 2nd ed york,
Approximate Word count = 2433
Approximate Pages = 10 (250 words per page)