Social Research Glossary A B C D E F G H I J K L M N O P Q R S T U V W X Y Z Home
Citation reference: Harvey, L., 201217, Social Research Glossary, Quality Research International, http://www.qualityresearchinternational.com/socialresearch/ This is a dynamic glossary and the author would welcome any email suggestions for additions or amendments.

A fastpaced novel of conjecture and surprises 
_________________________________________________________________
Multicollinearity
Multicollinearity is a term that applies when the regression model is used to predict a dependent variable from two or more independent variables that are highly correlated among themselves.
For instance, if one wished to predict people's heights from their weights as given by two different weighing machines, the two weight measures would be very highly correlated and in fact only one, and not two, dimensions is really being measured. When serious multicollinearity exists this may play havoc with regression equations and with path coefficients. In extreme cases it is impossible to calculate a regression equation, and even if it is still possible, the regression weights of the equation are unstable from sample to sample. Path coefficients are apt, rather arbitrarily, to become zero.
See also
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z Home