The Correlation Says So, Ash Chambers

The general understanding of what information Correlational Research can provide is varied and sometimes out right awful. Not everyone understands it. A correlational study investigates the relationship between two (or more) variables–two factors that can change. Researchers measure these variables and can use math to determine if they change in a related manner. If they vary closely enough, it is determined that they have a significant relationship and that one variable can predict the other. However, this does not in any way convey that the two variables affect one another. This is because causation (a cause and effect relationship) cannot be inferred from a correlational study. It has predictive value only.

A recent experience with this concept was a rather funny picture I stumbled upon. Many people online that work/study outside the realm of math/science often mistakenly assume that correlation means causation. This picture takes an assortment of random correlations that do covary, but are in no way related to one another. It’s basically poking fun at the general misunderstanding of correlational research. To take a peek at the picture, click here.

This picture well illustrates the confusion that the general public experiences when evaluating correlational research. While some correlational studies are good start-points for later experimentation, some have little scientific value. But because it has the ‘math’ and ‘sound’ of what people consider science, they buy into it. The picture is pointing out various incidents that are clearly unrelated yet correlate well. Meaning that you could get just as strong a correlation from unrelated variables as you could get from interrelated variables. Correlational studies cannot account for all factors that go into events–it can only report the amount included in the study. A big problem of this is that it ignores confounding variables. Not to mention that it is also impossible to determine the direction of a relationship, even if it seems apparent that one exists. The author of this picture sought to convey this flaw by using many incomparable and laughable variables. But even though this is humor, he/she is hitting on the biggest danger of correlational research. Although it can help identify possible relationships and can support ideas, it can in no way be relied on as definitive evidence or suitable to infer causation. It is only a description, sometimes a predictive tool, but nothing more.

  1. Leave a comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: