You will need
- - a series of distributions of the dependent and independent variable.
- paper, pencil;
- - computer and software to work with spreadsheets.
Select the two variables between which you suspect there is a connection, usually taking values that change over time. Note that one of the variables must be independent, it will act as the reason. The second must change synchronously with it will decrease, increase or vary randomly.
Measure the value of a dependent variable for each independent testimony. Record your results in a table, in two rows or two columns. To detect the presence of when you need at least 30 readings, but for more accurate results take care of not less than 100 points.
Construct a coordinate plane, with the y-axis put the value of the dependent variable, and the x – axis is independent. Sign axis and specify the units of each indicator.
Note on the chart the points of the correlation field. On the x-axis, find the first value of the independent variable and on the y – axis the corresponding value for the dependent. Construct the perpendiculars to these projections and find the first point. Mark it, circle with a soft pencil or pen. Likewise, build all the other points.
The resulting set of points is called the correlation field. Analyze the resulting graph, draw conclusions about the presence of a strong or weak cause-and-effect relationships, or lack thereof.
Note the random deviations from the schedule. If in General, there are traces of linear or other relationship, but the whole "picture" spoil one or two points, find themselves aloof from the General population, they can be considered random errors and not to be considered when interpreting the chart.
If you need to build and analyze the field correlation for large amounts of data, use programs designed to work with spreadsheets, such as Excel, or purchase a special program.
Advice 2 : How to calculate correlation
A correlation is called the mutual dependence of two random variables (often two groups of values) at which a change in one of them leads to a change in another. The correlation coefficient shows how likely the second value while changing the values of the first, i.e. the degree of its dependence. The easiest way to calculate this value use the appropriate function, built-in table editor Microsoft Office Excel.
You will need
- Spreadsheet editor of Microsoft Office Excel.
Start Excel and open the document that contains the data groups, the correlation coefficient between which you want to calculate. If such a document has not yet been created, enter the data into an empty table - the table editor creates it automatically when the program starts. Each of the groups of values, the correlation between which you are interested in, enter in a separate column. It doesn't have to be adjacent column, you are free to arrange the table in the most convenient way to add additional columns with explanations to the data, column headers, summary cells with total or average values, etc. Can even be positioned in a vertical data (columns) and horizontal (rows) direction. The only requirement that must be met - the data cells of each group should be placed one after another, to thus create a contiguous array.
Navigate to the cell that contains the correlation value data of two arrays, and click on the Excel menu tab "Formulas". In the group of commands "Library functions" and click the very last icon - "Other functions". Will reveal a drop-down list where you should go to "Statistics" and select the function correl. This will open the wizard window function with a shape designed to fill. The same window can be opened without the tab "Formulas" just by clicking on the icon insert function, placed to the left of the formula bar.
Specify the first group of correlated data in the field "Pattern1" masters of formulae. To enter a range of cells manually enter the address first and last cells, separated by a colon (no spaces). Another option - just select the desired range with the mouse, and the desired entry in this field form Excel places on their own. The same operation should be repeated and with the second group of data in the field "Pattern2".
Press the OK button. The table editor will calculate and display the correlation value in the cell with the formula. If necessary, you can save this document for future use (Ctrl + S).