Paired vs Unpaired

Paired analysis is appropriate in the following circumstances:

• When you do a before and after measurement in each subject.

• When you match subjects in pairs (i.e., age, etc.), then treat one of the subjects and not the other. Pairs must be made before data are collected.

• When you compare relatives (i.e., sibling studies).

• When you perform an experiment many times, each time with the experimental and control sample treated in parallel (including log-transformed ratio data, as described in Example 7).

Example 7. You have done Northern blots of gene expression in mutant animals vs. wild-type animals. Each of your blots shows wildly different signal magnitudes from other blots, due to technical reasons such as exposure time; having mutant and wild-type samples on every blot allows you to control for this variability by expressing data as a ratio of mutant to wild-type values.

Ratios are not normally distributed, but logs of ratios are; therefore, you log- t ransform your data. Ratios become differences when log- t ransformed [i.e., log(mutant/wild type) = log(mutant) - log(wild type)], so to analyze this type of data you can take the log of each data point and perform a paired t-test on the transformed data.

In judging whether distributions are normal in paired testing, it is the differences between the two members of each pair that must be approximately normally distributed rather than the two distributions themselves. If data from two sets to be compared do not meet one of the criteria above, they should be analyzed by unpaired testing [35].

Project Management Made Easy

Project Management Made Easy

What you need to know about… Project Management Made Easy! Project management consists of more than just a large building project and can encompass small projects as well. No matter what the size of your project, you need to have some sort of project management. How you manage your project has everything to do with its outcome.

Get My Free Ebook

Post a comment