Skip to main content
Glite Dictionary
📊 Variables highly related 📏 On one line
collinearity

collinearity

1 9.5
Two or more variables are so closely linked that their separate effects cannot be seen
  • noun
  • /ˌkɒlɪˈnɪərɪti/
  • Specialized
: colinealidad
  • We did not include educational level as a covariate because of its collinearity with income.
  • collinearity diagnostics
  • collinearity problem
  • collinearity among the independent variables

So, so that you could sort of, you know, that wouldn't induce this sort of perfect linear, you know, perfect linear collinearity problem.

Examples

  • Results of the collinearity diagnostics indicated that collinearity was not a problem for the set of independent variables used in the analyses.

    Academic text (2008)

  • Mean-centered age was used to reduce its collinearity with other independent measures.

    Academic text (2006)

  • Our motivation was to eliminate the collinearity problem, which can be particularly acute in a small sample.

    Academic text (2000)

  • We identified collinearity between two control variables used in tests of hypotheses.

    Academic text (2000)

  • The values for all predictor variables showed that collinearity was not a concern in the regression analyses.

    Academic text (1999)

  • However, in more commonly used regression models, collinearity, or correlations between risk factors used as predictors, may lead to misleading conclusions.

    Academic text (2005)

  • All variance inflation factors were within acceptable limits, indicating collinearity was not a problem (Cohen, Cohen, West, & Aiken, 2003).

    Academic text (2013)

  • Although HNR was not a significant contributor to the regression model, the variance inflation factor was 12.801, suggesting collinearity.

    Academic text (2013)

  • Menard (2002) suggested a protocol for logistic regression diagnostics that includes testing for collinearity, nonadditivity, and outliers.

    Academic text (2008)

  • In our analysis, we found that the collinearity between the predictor variables skewed the results significantly.

Synonyms

multicollinearity
vscollinearity
  • Specialized
1 9.9

When two or more things used to predict a result are very similar, you cannot tell their separate effects

is the broader statistical situation of highly correlated predictors beyond regression contexts

How Collinear

multicollinearity
  • Specialized
1 9.9
collinearity
  • Specialized
1 9.5
correlation
  • Specialized
14 4.9
independence
441 3.0

Surface Forms

collinearity singular
collinearities plural

Morphology

collinearity = collinear (semi-transparent) = col + line + ar + ity

Formed from the adjective 'collinear' + '-ity'; the statistical sense is a specialized, metaphorical extension of the geometric idea, so the derivation is not fully obvious to all learners.

Etymology

Collinearity comes from co- meaning 'together' and linear from line meaning 'a straight line', so it literally means 'together on a line'. In statistics, this is a helpful image: when two or more variables are 'together on a line' they move in the same way, so it is hard to tell their separate effects.