[QUOTE=Mosphaitus;13831]Hi all, ive been looking at a quetion that says

"What is the correlation coefficient, what information does it provide about the way two variables are related. What are its advantages and disadvantages?"

Here is my answer, although i need to know what else to put, and more advantages to it, and disadvantages if there are any

The question carries 12 marks, so they want a fairly long explanation of it.

-------------

1. The correlation coefficient is a numerical way to quantify the relationship between two variables, e.g. X and Y and it is denoted by the symbol R. The corelation coefficient is always between -1 and 1, thus -1 < R < 1.

2. Larger correlation coefficients, such as 0.8 would suggest a stronger relationship between the variables, whilst figures like 0.3 would suggest weaker ones.

3. However, the correlation coefficient does not imply causality, that is it may show that two variables are strongly correlated , however it doesnt mean that they are responsibile for each other.

4. Advantages of the correlation coefficient are that it is easy to work out and its easy to interpret ( Need more here! )

_____________________________________________________________

1(a). I would state "up front" that the Pearson correlation (r) is an index of the "strength of linearassociation between two variables..."

1(b).rcan be +-1, that is, Y is a linear tranformation of X i.e. Y = a + bX. Desribe graphically what the data would look like ifr=1 (line),r=0 (circle), 0<r<1 (ellipse).

2. I think you should write: "Larger correlation coefficients, such as+-0.8 would suggest a stronger relationship between the variables, whilst figures like+-0.3. Students (and others) often get confused and incorrectly interpret a negative correlation as weaker than a correlation of zero.

3. What does the following imply: "however it doesnt mean that they are responsibile for each other". I think you could do a better job here.

4. In terms of the advantages of the correlation coefficient, explain the advantages thatrhas when juxtaposed to the Covariance between two variables.

--Dragan