Correlation does not necessarily imply causation, as you know if you read scientific research. Two variables can be associated without having a causal relationship. However, just because a correlation has limited value as causal inference does not mean that correlation studies are not important to science. The idea that correlation does not necessarily imply causation has led many correlation studies to downgrade. However, used appropriately, correlation studies are important to science.
Why are correlation studies important? Stanovich (2007) points out the following:
âFirst, many scientific hypotheses are stated in terms of correlation or lack of correlation, so that such studies are directly relevant to these hypothesesâ¦â
âSecond, although correlation does not imply causation, causation does imply correlation. In other words, although a correlational study cannot definitively prove a causal hypothesis, it can exclude one.
Third, correlational studies are more useful than they appear, because some of the complex correlational models recently developed allow very limited causal inferences.
â¦ Some variables simply cannot be manipulated for ethical reasons (eg human malnutrition or physical disabilities). Other variables, such as birth order, sex, and age are inherently correlational as they cannot be manipulated and, therefore, scientific knowledge about them must be based on correlation evidence.
Once the correlation is known, it can be used to make predictions. When we know a score on one measure, we can make a more accurate prediction of another measure that is closely related to it. The stronger the relationship between / among the variables, the more accurate the prediction.
Where possible, evidence from correlation studies may lead to testing this evidence under controlled experimental conditions.
While it is true that correlation does not necessarily imply causation, causation implies correlation. Correlational studies are a stepping stone to the more powerful experimental method, and with the use of complex correlational designs (path analysis and cross-shift panel designs), allow very limited causal inferences.
There are two major problems when attempting to infer causation from a simple correlation:
- directionality problem – before concluding that a correlation between variable 1 and 2 is due to changes in 1 causing changes in 2, it is important to realize that the direction of causality can be, therefore, the reverse of 2 to 1
- third variable problem – correlation between variables can occur because the two variables are related to a third variable
Complex correlational statistics such as path analysis, multiple regression and partial correlation “allow the correlation between two variables to be recalculated after removing the influence of other variables, or” factored “or” partial “” (Stanovich , 2007, p. 77). Even when using complex correlational designs, it is important that researchers make limited causal claims.
Researchers who use a path analysis approach are always very careful not to formulate their models in terms of causal statements. Can you understand why? We hope you thought the internal validity of a path analysis is low because it is based on correlational data. The direction of cause and effect cannot be established with certainty, and âthird variablesâ can never be completely excluded. Nevertheless, causal models can be extremely useful in generating hypotheses for future research and in predicting potential causal sequences in cases where experimentation is not possible (Myers & Hansen, 2002, p.100).
Conditions necessary to deduce causality (Kenny, 1979):
Time priority: For 1 cause 2, 1 must precede 2. The cause must precede the effect.
Love relationship: The variables must be correlated. To determine the relationship between two variables, it is necessary to determine whether the relationship could occur due to chance. Lay observers are often not good judges of the presence of relationships, so statistical methods are used to measure and test the existence and strength of relationships.
Not false (false meaning meaning ânot authenticâ): âThe third and last condition of a causal relation is the non-false (Suppes, 1970). In order for a relationship between X and Y not to be wrong, there must not be a Z that causes both X and Y so that the relationship between X and Y disappears once Z is controlled â(Kenny , 1979. pp. 4-5).