1
$\begingroup$

How do I find the correlation coefficient of two variables with unequal number of data using Excel? And do I use the mean as data to calculate?

$\endgroup$
2
  • 3
    $\begingroup$ to get correlation, your data will be paired. If the observation for one of the variables in a pair is missing, you'd normally exclude the other member from the calculation of the correlation. $\endgroup$
    – Glen_b
    Mar 3 '13 at 5:11
  • 1
    $\begingroup$ @Glen_b's comment indicates you will need to provide more details about your data, user21489, in order for this question to be answerable. $\endgroup$
    – whuber
    Mar 3 '13 at 18:49
2
$\begingroup$

A correlation coefficient requires that for every x-value there be a y-value ... otherwise, how should I know how much value of y to attribute to x? For example, if you're trying to determine how much height correlates with age, you need a collection of heights and ages. Knowing that a person is 5'5" tall doesn't help at all if you don't know how old that person is.

Do you mean that you have several y values for each of your x values? Like

5-year olds: 36", 37" 38"

6-year olds: 37", 39", 40" ... ?

If so, make a row for each y value, and include its corresponding x-value:

Age | Height

5 | 36

5 | 37

5 | 38

6 | 37

6 | 39

6 | 40

And then you can use CORREL()

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.