p-values and Significance
Instructional objectives for the lesson
1. The student will be able to interepet statistical significance therough the use of p values is in research.
2. The student will be able to differentiate between statistical and clinical significance.
p values
When you review research articles that conduct quantitative data analyses, one of the terms you will commonly see is a p-value. The p-value determines the significance of the results from a statistical standpoint. If we have p-value that yields statistical significance, there is a high probability something has impacted something else. Research often tests hypotheses to test the validity of a claim that is made about a population. Typically this claim states there is no statistical significance between two or more groups meaning some factor has not impacted these groups. This is referred to as the null hypothesis.
When p-value is calculated, depending on the value, we may reject the null hypothesis and state there is statistical significance. This results in what we call the alternative hypothesis.
The p-value is a number between 0 and 1 and interpreted in the following way:
Significance
Statistical significance is a term that tells how sure you are that a difference or relationship exists. However, a significant difference or relationship does not strong your relationship is. It does not say a relationship is strong, weak, or moderate. After finding a significant relationship, you also need to evaluate its strength. A lot of this can depend on your sample size. Calculating strength is not part of this module, but an important consideration when deciding to act on a statistically significant finding.
Clinical (practical) significance is a term that really looks to evaluate whether a treatment has a noticeable impact. It is not beholden to a statistical test like a p-value, but you do need to evaluate from a statistical standpoint the effect size and the number needed to treat to achieve a benefit.
Additional Links
Below are some links that will give you additional information on p-values and significance
http://www.statsdirect.com/help/default.htm#basics/p_values.htm
http://www.dummies.com/how-to/content/what-a-pvalue-tells-you-about-statistical-data.html
https://stat.duke.edu/~berger/p-values.html
Discussion Question
You and your classmates will be sent an email for access to a Google Doc to discuss this question.
Remember your responses should be respectful of other classmates feedback in this discussion.
In a clinical trial, which do you think is more important statistical or clinical significance? I will assign ½ of you to take statistical and the ½ clinical. You will post your responses on the Google Doc and respond to two classmates.
Assignment
By completing and submitting this assignment you are certifying it is your work. Please refer to the student's code of conduct as it discusses the ethics and responsibilities of earning an education at Youngstown State University.
Copy and paste the assignments below into a Word document and complete the assignments.
Once you have completed both assignments, please submit to Silvia Stefan at sastefan@ysu.edu
p-values.
Indicate whether the values below are significant (S) or not significant (N) to right of the value
Statistical and Clinical Significance
In narrative form, answer the following questions: