Archive | October, 2011

The Causation Correlation Confusion.

21 Oct

“Man is so intelligent that he feels impelled to invent theories to account for what happens in the world. Unfortunately, he is not quite intelligent enough, in most cases, to find correct explanations. So that when he acts on his theories, he behaves very often like a lunatic.” –Aldous Huxley

Causation is when one thing has caused the other outcome to happen. For example, saying that smoking causes lung cancer. Correlation shows a relationship between two events, for example there is a relationship between smoking and alcoholism. A correlation however does not mean that one has caused the other, it just shows a link. It does not mean that smoking causes alcoholism, or alcoholism causes smoking, but there is a link between the two. In research there could be loads of other variables, which can’t always be controlled, and so therefore although there is a correlation, it cannot easily be proven to be causation.

I think that people do generally use correlations to assume causations; I know that I for one do. The human mind likes to make relationships between information and jump to conclusions. Thankfully, in professional Scientific and Psychological research papers, this is not the case! But a huge culprit for assuming causations is the media! I think that by now, we all know that the media do not always tell the truth. They are there to interest people, sell stories and make money. Sadly the truth is not always as interesting as they like, people want facts! They want to know the cause of things, not a link between two things that could mean anything. So the media, especially tabloid papers, like to make mountains out of molehills when it comes to scientific and psychological discoveries. Even politicians have been known to assume causation to put across their beliefs to the public.

Some silly assumptions by the media are:

“Men who have beards are happier than men who do not.” This leads the public into thinking that beards have a magic that CAUSE the male owner of them to be happier. This has to be nonsense! If a man grew a beard, he wouldn’t necessarily be happier overnight! The correlation between the two may have been due to the fact that maybe more self confident and laid back men grew beards.

“People who own red cars are twice as likely to have an accident as people who own blue cars.” This doesn’t mean that red cars are more dangerous. There could be many reasons as to why red cars have more accidents. The owners of red cars may pick the colour red due to them being aggressive. The colour red may also be more popular with bold, ‘I want to stand out’ people, who are more likely to drive aggressively.

sooooo… to avoid all this silly confusion: ALWAYS LOOK AT THE ORIGINAL PAPER!

An interesting website on this correlation vs causation topic is:


“What role does the null hypothesis REALLY play in the scientific process?”

13 Oct

3rd blog already, how time flies! This week I was pretty stuck for ideas on what to write my blog on, so I resorted to looking at the list of possible blog topics. I have been puzzled over the need for a null hypothesis since writing my ‘Is Psychology a Science’ assignment, so I have decided to debate it.

In science, research must follow the rule of the null hypothesis. This is the assumption that all assertions are false, unless they have been proven, with evidence, to be correct. So in other words, everything that hasn’t been backed up with evidence is considered untrue. The null hypothesis is the opposite of what the experimenter predicts, that there is no relationship between the two variables. The Alternate hypothesis is the null hypothesis’s rival. The alternate hypothesis would assume that the assertion was true.

For example:

Hypothesis– The plate is hot due to it being in the sun.

Null hypothesis– The plate being hot is not due to it being in the sun.

Alternate hypothesis– The plate is hot due to it being in the sun.

But why do scientists actually use the null hypothesis? They say that the only way to support a hypothesis is to disprove a null hypothesis! Rather than trying to prove that the alternate hypothesis is correct, you must find the null hypothesis to be wrong! This means that we would have to look for other reasons that the plate could be hot, rather than it being in the sun. Doesn’t that sound all very complicated and unnecessary really?  Why cant we just find evidence for a hypothesis being correct?

I think that the null hypothesis is mainly used to act as a safety guard. It helps to reduce bias in experiments. Researchers will be inclined to be bias towards the results that they would like in their experiments. But if they were to set out to prove or disprove a null hypothesis, would they be less biased? It helps to cancel out other variables and ensures that the researcher has covered all possible variables in their experiment.

I’m still finding this all a little confusing, so please criticise my ideas as to why the null hypothesis is important!


Can a theory ever be truly proven?

6 Oct

A theory can only be proven by evidence and research that has happened in the past. But what is there to say that something new and different, which has never happened before, will happen in the future? Does this mean that no theories can ever be 100% proven? For example, how do we know that the sun is going to rise tomorrow for certain, apart from the fact that is has risen every morning for as long as can be recorded. Does this mean that all theories are, are educated guesses and not ever facts? Can data ever be truly reliable?

Where can we, as Psychologists, decide when a theory has been proven? Prove in the Collins English dictionary is defined as; “establish the validity of”. But to what extent can validity be established? In Psychology particularly, every single person in different, so results from experiments might indicate one thing but different people/ cases may be completely different. To make an experiment fair, the researcher would surely have to find a group of participants with identical characteristics, backgrounds and personalities; this would be impossible. Even if it were possible, the outcome of the research would only be specific to this group of people, and may not apply to anyone else.

So this has led me to wonder if, in any sciences, the word proven can ever be used truthfully. I don’t think it is ever possible for enough research to be done to ensure that results are 100% valid, especially in Psychology because every persons mind is so different and complex. I believe that if enough research is done to get repetitive matching results, it can be relied upon, but it will never be truly proven. I think we have to use our own initiative and judgement when coming to decide whether a theory is correct in Psychology, but even then, we never know what is round the corner!