spurious correlation


Also found in: Dictionary, Thesaurus, Wikipedia.

spurious correlation

[′spyu̇r·ē·əs ‚kä·rə′lā·shən]
(statistics)
The value of the coefficient of correlation when it is computed correctly but its relationship implications are nonsensical or unreasonable.

spurious correlation

see MULTIVARIATE ANALYSIS.
References in periodicals archive ?
Along the same lines, including the use of other substances, such as cigarettes, serves as a proxy for the discount rate or other unobserved characteristics, such as a general propensity for deviance, and thus also helps address the issue of a possible spurious correlation.
1], will be biased and may merely reflect a spurious correlation between the two outcomes.
A spurious correlation between ESP scores and extraversion could arise if subjects' responses on the extraversion scale were influenced by knowledge of their performance in the ESP task.
If this is the case, correlations between the cost advantage ratio and measured capital intensity are an inadequate test of the hypothesis of spurious correlations between the cost advantage ratio and price cost margins.
The problem of spurious correlations can be alleviated in two ways.
Both are spurious correlations until such time as some real statistics are shown to suggest otherwise.
The book's investigation of the discovery of such problems as spurious correlations between time series and the development of smoothing and decomposition techniques were fascinating to me.
First, filtering series that lack a seasonal component can induce spurious correlations in the adjusted series.