Data Input

        
To enter your data, you must first save it as a comma separated (csv) or tab-separated (tsv)  file. You can prepare the data file in a spreadsheet application such as Microsoft Excel or Apple Numbers, and then export or save to the format you prefer.


In your spreadsheet application set up the data with 1 column for each group. For example, if you wanted to correlate data from four different groups with a maximum total of 30 variables in each group 1...group 4, you would enter the group names in the top row to give four columns, and then fill each column with the data for the respective groups, as such:

 
        





#
Group 1
Group 2
Group 3
Group 4






1
67
70
68
NA




2
56
68
52
71




.
.
.		
.
.
.		
.
.
.		
.
.
.		
.
.
.		




30
NA
87
52
69







Note: If you have any missing data or empty cells, enter NA as in the case of Group 4 #1 and Group 1 #30 above.


Once you have the data ready, you can then import it into the application and run the correlation. To do so, follow these steps:


    

  1. Under Choose file to upload click Browse.... This will allow you to locate the file on your computer and upload the data to the application. Once you have completed the upload, you will see the name of the file and Upload complete displayed on the screen. You can check your data has been properly imported by checking the Display data checkbox under the Submit button.
    2. 

  2. If all is correct, select the different options to determine how your data will be read and processed.
    3. 

  3. Choose the correlation method you want to use.

  4. If you wish to generate bootstapped confidence intervals (CIs) then choose that option and the size of the bootstrap.
    5. 

  5. Once you have selected and checked everything, click the Submit button.
    6. 

  6. Output from your analysis will be displayed under the Results Output tab.
    7. 



The Results Output displays the following:


  • pairs.panels for the data set. Details on this type of plot can be found here, but basically it shows the correlation coefficients along with scatterplots of the data including a regression line.
    • 

  • The set of data shows the correlation matrix, the lower and upper confidence intervals (CIs), and for convenience, p values denoted as Probability values. The p.adj value
    • 

  • The figures in the coefficient of determination R2) table show the percentage of variability in one variable that is shared by the other. If you have a R2 value of, for example, 51% for variable 1 and 2, then you can say that 51% of the variance is shared by variable 1 and 2. See this article for further explanation.
    • 

  • If you chose to calculate bootstrapped CIs, this output will also be displayed.
    • 



Under the Exploring Data tab, the application displays descriptive statistics and a density (violin) plot. The latter shows the data distribution along with the median and 25th & 75th percentile lines. Use this to help you decide which correlation method to select. You can also look at the skew and kurtosis figures in the descriptive statistics to help make your choice (note that the kurtosis value provided is an adjusted measure of excess kurtosis, details here).