# One might reasonably ask why we bother with correlations at all if causal relationships are what researchers really are seeking to address?

Discussion 2: Correlation does not Equal Causation

Huck (2012) instructs readers on p. 62 of Reading Statistics and Research (BELOW) to remember that correlation does not equal causation.

One might reasonably ask why we bother with correlations at all if causal relationships are what researchers really are seeking to address?

Reflect on the readings from this week and your experience with data and research. Then, in your own opinion, explain why or why not correlational research is (or is not) useful.

• Your initial post (approximately 200-250 words) should address each question in the discussion

Huck (2012) instructs readers on p. 62 of Reading Statistics and Research – BELOW

Correlation and Cause

It is important for you to know that a correlation coefficient does not speak to the issue of cause and effect. In other words, whether a particular variable has a causal impact on a different variable cannot be determined by measuring the two variables simultaneously and then correlating the two sets of data. Many recipients of research reports (and even a few researchers) make the mistake of thinking that a high correlation implies that one variable has a causal influence on the other vari- able. To prevent yourself from making this mistake, I suggest that you memorize this simple statement: correlation cause.

Competent researchers often collect data using strategies that allow them to address the issue of cause. Those strategies are typically complex and require a con- sideration of issues that cannot be discussed here. In time, however, I am confident that you will understand the extra demands that are placed on researchers who want to investigate the potential causal connections between variables. For now, all I can do is ask that you believe me when I say that bivariate correlational data alone can- not be used to establish a cause-and-effect situation.

Coefficient of Determination

To get a better feel for the strength of the relationship between two variables, many researchers square the value of the correlation coefficient. For example, if r turns out equal to .80, the researcher squares .80 and obtains .64. When r is squared like this, the resulting value is called the coefficient of determination. The coefficient of determination indicates the proportion of variability in one variable that is associated with (or explained by) variability in the other vari- able. The value of r2 lies somewhere between 0 and 1.00, and researchers usu- ally multiply by 100 so they can talk about the percentage of explained variability. In Excerpt 3.24, we see an example from a stress/eyewitness study where r2 has

EXCERPT 3.24 • r2 and Explained Variation

Pearson’s correlation coefficient between the change in heart rate (labyrinth mean heart rate–baseline mean heart rate) and state anxiety score showed a reliable association, r .76 [and] r2 .58. Change in heart rate accounted for 58% of the variance in state anxiety score. Source: Valentine, T., &Mesout, J. (2009). Eyewitness identification under stress in the London Dungeon. Applied Cognitive Psychology, 23(2), 151–16