-
Asked by tomandclare to Audra, Fiona, Gavin, Justin, Steve on 16 Mar 2012.
Question: How have you made your investigation fair?
- Keywords:
-
Steve Faulkner answered on 13 Mar 2012:
To make my trials fair one of the big things I have to do is ensure that I use athletes who are used to performing the task that I use as a performance test. For example, if I asked a runner to ride a 60 minute cycling time trial, it is very likely that they would come out with very different time each time. However, as an experienced cyclist, and they should be able to repeat their performance again and again, within as close as a few seconds variation. Secondly, as the time of day has an effect on exercise performance, all trials are completed at the same time of day, again to minimize variation between trials. Finally, and this is a big one….it’s important to balance your experimental design. This means that the order of conditions is different for each participant if you’re doing trials that involve multiple tests or visits. For example, they may complete condition 3 first, 1 second and 2 last. This will help to reduce any practice/learning effects that occur as participants get more practice. Hope this helps!
Steve
-
Gavin Devereux answered on 16 Mar 2012:
Hello. sorry for the slight delay in getting back to you.
As Steve mentions, if we have human participants, we need to ensure they are aware of what’s being asked of them (otherwise you can get dodgy data).
Also, we typically use something called a control condition/group. This is where we do not ask people in the control group to do certain things. In the other group they do complete all of the experiment. That way, we can check if changes are actually caused by the experiment (if control people have no change, but the ‘experimental’ group do). This is one of (if not the) most important aspects of scientific testing when you want to find out cause and effect.
Oooh dear. Verging into some dull research theory, sorry 🙂
Comments
tomandclare commented on :
thank you