Reporting Statistics in Research

A few days ago, I met with a researcher regarding their statistics. Results were presented to them in Amy fashion- all parties involved were pleased with the outcome. The other day, I discovered the same researcher had presented their work in a public forum – without p-values. What’s wrong with not including p-values? When you write-up a report that includes statistics, you should include p-values to substantiate the claims in your paper. Now I am not saying that the world revolves around p-values in statistics. But there is a difference between saying something is statistically significant versus¬† showing it’s true. Anyone can claim their work is statistically significant when it’s not. You want someone to take your findings seriously. Your point should be ‘I discovered something and here is the evidence’.

So how should you present p-values? If you declare something is statistically significant then add the p-value to your claim. For example, suppose we conduct statistical analysis and discover that a drug is statistically significant to a model focused on the outcome , disease. We have the statement:

“The Student T test showed¬†statistically significance¬†(p=.0301). ”

This is just an example. Of course there is more to a report than a p-value. I am just sharing information that I have come across in my work.

 

If you enjoyed this blog there are more to follow.-Amy

 

 

 

 

 

 

 

 

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.