Hi iPod,
Apologies that no one has answered your post in the last few days. If you are still working on this, I have a few comments for you...
--This sounds like a psychology experiment rather than a physical sciences experiment. Don't be surprised if a moderator moves your post to a different forum, where experts with more relevant expertise will see your question:
https://www.sciencebuddies.org/science- ... m.php?f=25
--I recommend rethinking the definitions of your variables and controls. An example of a control might be having all of your subjects complete the same task, rather than different ones. A variable is anything that might conceivably affect the outcome of your measurements. Generally, an experiment will have one independent variable (IV), one dependent variable (DV), and everything else you can think of should be a control (kept constant during each measurement). The IV is what you change, and the DV is what you measure. You are determining the effect of the IV on the DV. Each person is a trial of your experiment, rather than a variable. More info here:
https://www.sciencebuddies.org/science- ... ndex.shtml
https://www.sciencebuddies.org/science- ... bles.shtml
--How are you measuring stress? How are you measuring performance? The method that you use to measure the IV and DV will have inherent sources of error -- we can't help you figure out what those are without knowing the method. For example, if you asked people to rate their stress on a scale of 1 to 10, you could reduce the error in that rating by giving them objective criteria to base their own ratings on. You could reduce error in the performance measurement by making it very clear-cut, such as time required to solve a puzzle; error in this sort of measurement would come from things like your accuracy with the stopwatch, if they were using a pen and the ink stopped flowing, etc.
--You might consider measuring each person's performance under two different levels of stress and compare the change in performance to change in stress level. (If you used something like a puzzle to measure performance, you would want to use two different puzzles since people could remember the solutions.) You could either measure their performance on two different days, or attempt to create stress somehow. You'll want input from experts in your field on working with human subjects and what sorts of conditions have been used to simulate stress before, but if I'm not mistaken, a simple example could be something like playing loud music. This kind of technique can help compensate for the fact that your trials are not identical -- each person is different. There could be unwanted correlations that effect your results; for example, if you test both adults and children, what if adults do better on the performance test but also tend to be more stressed? For this reason, it's also a good idea to keep your test population as homogeneous as possible. But, you also have limited resources and time and have to balance similarity of subjects against having a large sample size.
Best wishes,
Amanda
