How to Find the Correlation Coefficient when Doing Statistics
User-Submitted Article
The correlation coefficient is the strength of association between two variables, x and y. Suppose you are looking at a scatter diagram; you can find the average of the graph, but not every number will be near the average. This article will show you how to find how strong the association is between the plotted numbers, and the average.
Difficulty: Moderate
Instructions
Things You'll Need:
- Two lists of numbers
- A calculator
- Paper
- Writing utensil
- 1Gather two lists of numbers, either given to you, or extracted from a diagram. Make sure these two lists are labeled 'x' and 'y'. If you are gathering information from a diagram, the horizontal axis should be 'x', and the vertical will be 'y'. When you gather this information, make sure all the 'x'-values are in increasing order in your list (their subsequent 'y'-values may or may not be in order; this doesn't matter).
- 2Find both the average and standard deviation (SD) for each of these columns. In other words, you will want to have the average and SD for only 'x', and the average and SD for 'y'.
- 3Next, make three columns, and label the top of each one. The first should be 'x', the second should be 'y', and the last will be 'xy'.
- 4
Next, convert each 'x' and 'y' value into standard units. Taking your first 'x'-value, subtract the average from that number, and divide it by the SD of 'x'. Do the same for every value of 'x' and 'y', so in the end, you should have the same number of values in standard units as you originally had in normal units. - 5Once you have filled the 'x' and 'y' columns with their standard values, multiply each 'x' and 'y' pair, and put the product in the 'xy' column.
- 6When the 'xy' column is all filled out, find the average of all the values in this column. The answer you get will be your correlation coefficient.
Read more: How to Find the Correlation Coefficient when Doing Statistics | eHow.com http://www.ehow.com/how_4839395_correlation-coefficient-doing-statistics.html#ixzz1GHBsTPbH
0 comments:
Post a Comment