Analysis of Human Reaction Times

[danv_MIT]

Hello,

I posted this in the Life, Earth, and Social Sciences category yesterday and it was recommended that I post this here too.

I'm conducting a science fair project on human reaction times- I want to determine how each of several factors affects reaction time. I have over 50 subjects who took my reaction time test (a computerized assessment, accurate to the millisecond). Each subject was measured 50 times on their latency to respond to a simple stimulus (box appears on-screen).

I recorded each subject's age and gender. In addition, I had them answer some questions to determine a numerical value representing their physical fitness, mental ability, fatigue, and stress levels. I also have the individual reaction times for each subject-- all 50 of them, not just the averages. (I knew I might need to compute standard deviation or something similar so I didn't discard the individual times.)

Now, after having collected the data, it seems I'm not so sure about how to draw conclusions from it. I want to know how each of the aforementioned factors affects the reaction times-- which affects it the most? Which has little effect on the reaction times?

I'm enrolled in an AP Statistics class, but we haven't studied ANOVAs in depth yet and I'm not sure whether ANOVA is the best choice for analyzing this data/answering these questions. Are there other superior methods of analyzing this kind of data? Any help would be appreciated.

Dan V.
MIT Hopeful

[APknowledgequeen]

Im doing a similar project except I'm seeing what stimuli helps increase a person's reaction time the best.

I would research each factor and how they tie into reaction time on the internet. I googled reaction time and fatigue for example and got several websites that helped extensively in my project and since yours is very similar to mine it might help you out as well.

As for you trying to conclude which effected it the least and the most, you said you were thinking about using ANOVA but didn't know how to use it. I just googled ANOVA and I got many websites. This is just one of them
http://www.physics.csbsju.edu/stats/anova.html
Hope it helps and good luck!

[aftab]

What is your hypothesis for this project? That will help you how you want to present all the data you gathered. Does it support your hypothesis?

You mentioned that you are trying to determine the affects of certain things on reaction time. Using your dependent, independent and controlled variables you may want to find the statistical correlation between reaction time and other variable(s).

[danv_MIT]

For my hypothesis I simply made an educated guess about which factor would have the most influence, backed up with sources of course

The problem is, I didn't directly control any variables. I just have many observations with values for each factor and then each reaction time.

I also stumbled upon multiple regression. This seems like a better alternative to ANOVA. I have visited some sites but can't find one with a very clear explanation of how to statistically determine which factor has the greatest influence? I want to prove that x affects reaction time more than y. How do I go about this?

I have found the correlation coefficients for each variable against reaction time. However,

Correlation does not imply causation.

So.... how do I establish causation when I didn't directly control variables?

How could I have controlled variables anyway? Human subjects can't just edit their physical fitness level or stress level like a text file on a computer or anything. I have to go by whatever their habits are.

[aftab]

Defining a hypothesis and variables is critical for any science project. The understanding of the scientific method is one of key aspects of doing projects. I believe you have the hypothesis, you just have to formalize it. The educated guess you started out with is what you need to formalize. E.g. 'Age, stress level and fatigue negatively affect human response time'.

True, that correlation does not imply causation; but you don't necessarily have to show or prove causation. If your hypothesis states something like above, then you are showing that x, y, z affect response time negatively. You are not stating x,y,z cause negative response time.

The coefficients of your regression analysis to some extent (without going too deep into regression) show the correlation of that variable. For example;

reaction_time = a x Age + b x Fatigue + c x Stress_level

The coefficients are a, b and c.
let say, a=15, b=5 and c=20 then reaction_time is showing higher correlation with stress_level (c=20), then Age (a=15) and then Fatigue (b=5).