megavideolinks
Joined: 19 Nov 2011
Posts: 273
|
 Posted: Sun Nov 20, 2011 2:28 pm Post subject: Logistic Regression Analysi |
 |
|
Regression is a procedure that uses one or more independent variables to predict the
values of a dependent variable and to reveal each independent variable’s association with the
variable being predicted. In the two analyses presented in table 2, computer use and Internet use
(the dependent variables) were analyzed using several independent variables that have previously
been found to be associated with the use of information technologies.
Logistic regression is a form of regression used when the dependent variable is
dichotomous (that is, when it can take only two different values, such as “computer user” or
“computer non-user”). In logistic regression, the equation predicts the natural log of the odds
(the “log odds”) of an event occurring, such as the sampled individual being a computer user.
3
The form of the equation is as follows:
Log[P/(1-P)] = B0 + B1X1 + … + BpXp
In this equation, the value B0 is a constant. The X values are the observed values of
independent variables such as age or income, and the corresponding B values are parameters
indicating the effect of a one-unit change in X on the log odds of the event. The B parameters
indicate the association between the independent variable and the dependent variable when all
the other independent variables are statistically controlled.
Dichotomous independent variables and the “reference category.” Most of the
independent variables in the logistic regression equations are treated as dichotomous. For
example, the six categories of race/ethnicity are included in the regressions as five variables:
Black, non-Hispanic; Hispanic; Asian or Pacific Islander, non-Hispanic, American Indian, Aleut,
or Eskimo, non-Hispanic; and more than one race, non-Hispanic. White, non-Hispanic is the
“reference category.” (The largest group is usually used as the reference category for categorical
___________________
essiac tea powder
promo electro ménager |
|
