Below is a pair of hypotheses. They constitute a well -formed pair.

Research Hypothesis:
Upper elementary students who have ‘morning’ recess for at least 20 minutes will engage in more “on-task” behavior in the hour after recess than a comparable group of students who are in a class where recess is not part of the morning routine.

Null Hypothesis:
Upper elementary students who have ‘morning’ recess for at least 20 minutes will engage in the same amount or less “on-task” behavior in the hour after recess than a comparable group of students who are in a class where recess is not part of the morning routine.


Key Elements of Each Form of the Hypothesis:

The hypotheses shown above comprise a pair of well-formed hypotheses because they clearly identify the target population, the treatment, the outcome measure, the expected outcomes for both the research and null versions of the hypothesis and the set of expected outcomes cover the full range of possible outcomes. It is also important to note that the wording is essentially the same in the research and null versions except for the segment where the different outcomes are specified.

Upper elementary students
--> the target group

who have ‘morning’ recess for at least 20 minutes
--> the intervention which defines the “treatment group”

will engage in more
--> predicted level of performance for the research hypothesis

will engage in the same amount or less
--> predicted levels of performance for the null hypothesis

“on-task” behavior
--> the outcome measure

in the hour after recess
--> when the outcome measure is expected to be observed

than a comparable group of students who are in a class where recess is not part of the morning routine.
--> definition of the comparison (or control) group