RSS

Correlation is not Causation! Correlation is not Causation! Correlation is not Causation!

19 Feb

Right same old same old, firs of all let’s define the terms – correlation and causation. It will get more interesting I promise.

Correlation – it compares two sets of measurements (variables) and indicates the strength (positive, negative, no relationship) of a relationship. A classic example would be measure everyone’s height and weight, in general the taller you are, the heavier you are. So here an established relationship / correlation between the two variables. The purpose of this is to make a prediction about one variable based on what we know about another variable.

Causation – an act or an occurrence cause a change in another variable, such as smoking cause lung cancer.

I do not think correlation solely can show causation.

I have found some silly examples to show that correlation does not mean causation, ready ??

  • Ice-cream sales and shark attack frequency are strongly correlated

This example shows that there is a correlation between the two HOWEVER this is not because sharks will start to attack in response to ice-cream, it is due to the fact that the two variables exhibit a common response to the hot weather (which you will never get in the UK, oooopss no offense guys).

  • A correlation between shaving habits and stroke risk

This example shows that shaving less than once a day means increasing your risk for a stroke by 70% and a 30% increased risk of dying from any cause. To be frank, this concept is ridiculous – by shaving more , it could change your chances of getting a stroke? Hmmmmm … yet some British researchers have found a correlation between the two (amazed). In the article, the researchers think that the underlying cause for this relationshop has to do with hormones. Testosterone has already been used to link with baldness to higher risk for heart disease. (Source: http://news.bbc.co.uk/1/hi/health/2729457.stm)

  • (x) the shorter a girl’s skirt and/or the lower her top was, (y) the more free drinks she received

I found this a rather interesting topic. What do you guys think? Is this true? Positive relationship? By wearing your skirt higher up and your top lower down will increase the likelihood of people buying you drinks?

Undeniably, the stats behind these studies suggested that a correlation do appear to exist, but a high positive correlation does not equate to cause and effect. I have found Wayne C. Curtis to back me up on this, who stated that, ” a high degree of positive correlation … does not necessarily mean that there is a cause-and-effect relationship between two variables. A high correlation can occur by chance alone or because both variables are related to some other variable which has not been considered in the analysis.”

In psychology, researchers undertake the experiments by using a very controlled environment, to ensure that the only effect that is causing DV is the IV. To this end, the researcher could then conclude that certain variable caused the change in the other. However confounding/extraneous variable can still occur which means it can be difficult to show an effect.

To conclude, (let’s finish off with one more example ;P) education and income are positively correlated, here we do not know if one caused the other. It might be that having more education causes a person to earn a higher income. It might be that having a higher income allows a person to go to school more. It might also be some third variable. A correlation tells us that the two variables are related, but we cannot say anything about whether one caused the other. This method does not allow us to come to any conclusions about cause and effect.

Thank you 🙂

Advertisements
 
5 Comments

Posted by on February 19, 2012 in Uncategorized

 

5 responses to “Correlation is not Causation! Correlation is not Causation! Correlation is not Causation!

  1. secretdiaryofapsychstudent

    February 21, 2012 at 2:32 pm

    I enjoyed reading the silly examples of correlation thus meaning correlation is not causation. Whilst researching this topic I found that although it does not imply cause it is a necessity in detecting relationships between variables without manipulation or control. Correlation is an important measure in psychology and without it many relationships between variables would not have been discovered. Also assessing the independence of concepts, correlation is valuable to use to assess whether or not the ideas one is examining are distinct and separate. Finally we can make predictions using correlations-you made a good example with height and weight.

     
  2. serenapsychology

    February 22, 2012 at 8:02 pm

    I totally agree that correlation does not show causation, however without a relationship there can not be causation and i feel that correlation is a good starting point for researchers to do further studies on. I also find it a bit of a shame that within society, such as the media, correlation is often is misinterpreted as causation and i think this can cause many problems and worries within society. For many articles such as http://www.telegraph.co.uk/education/educationnews/8064435/Children-in-single-parent-families-worse-behaved.html , this article says that children in single parent families are the worst behaved and imply that the cause of this is because of them being raised in a single parent household, however this correlation could be down to many reasons like the type of area they live in or that the child’s behaviour is the cause of the marital breakup.

    Not only can you not assume that a correlation is causation but if you did you are unsure of which on is having the effect on the other. This is because their are no experimental controls in a correlation study to ensure which variable is causing a change in the other. For example many researchers have found that depression is caused by low levels of serotonin (Bernstein, 1990), however it could be that depression causes the levels of serotonin to decrease.

    Also as you have mentioned when it comes to causation there are many factors that researchers have to consider to ensure that the only effect on the DV is the IV.

     
  3. poeywycheung

    February 23, 2012 at 12:59 am

    Good Blog 😀 and with many (funny) examples!
    I agree that Correlation is not Causation.
    Although correlation shows a relationships between 2 variables/things, for example the water intake is strongly correlate with the health level. It is possible to be the cause and effect but its is not definite since there is no proof, not enough evidence for this. But correlation can always be a clue and hint for us to investigate more to see whether the relationship is direct/indirect or actually no relationship.

     

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

 
%d bloggers like this: