Science Buddies: "Ask an Expert"

I am doing an experiment on sleep deprivation in teens and how it effects blood pressure and recall ability. Need some help on the experimental design. The teacher said we should have at least 5 subjects in the experimental group. Reading your info. on design groups, I would need a very large group to get some valid statistical results. However, it is difficult to have sleepovers with over 20 teens at your home so you can control what they eat and drink, etc. (space) Any advice?

I was thinking of using the same group and have three to four nights of testing. The first would be getting the recommended 9-10 hours of sleep and then testing their memory and blood presure in the morning. Is there a better test than the Mini Mental Stae Exam to use? The second time would be decreasing the time to six hours of sleep deprivation and the third would be the same. Can I test the same group and use the first night as the control?

Or, would it be a better study to have different groups of teens tested, which I could also do?

Thank you.
Sleepless

Hi Katerina!
Sounds like a really interesting project!! There are many open possibilities that you might want to consider, the one you mentioned seems like a good one. It is true that the larger the sample size, the more varied results, the less bias, and therefore the more accurate the results. Perhaps you could try different groups of teens and focus on some specific characteristics (age, gender, etc.) since your question seems to be extremely broad. Narrowing it down might help you design a more controlled experiment without introducing too many variables. Hope this helps, good luck!

Here is a link that explains variables to keep in mind:
https://www.sciencebuddies.org/science- ... bles.shtml

For their eating and drinking, I suggest telling them to avoid foods that contain caffeine or sugar because those may keep them up or affect them in some other way. Any triggering foods should be avoided. Tell me if you can think of any others! Having 20 teens over would not be very practical (and probably chaotic), but you don't really need that many for a general idea. If you were to be extremely accurate, it would be a good idea.

I suggest having a group of teens (at least five since that's your teachers requirements, but remember that in the real world, more subjects would increase the accuracy) the same age, gender, and with generally the same characteristics (possibly think about weight, race, height, sleeping habits, etc.) It's a good idea to ask your teacher about this as well.

Make sure to test their time in the morning at the same time on your clock.

I did not find any other official memory test. The Mini Mental State Exam is fine.

When you said "decreasing the time to six hours of sleep deprivation", did you mean that instead of 9-10 hours, they would get 3-4 hours of sleep? I hope not, because that's a bit extreme (although it would probably show a more significant difference). Also, it would be more accurate to just use 9 and 3 hours or 10 and 4 hours if that's what you meant.

Oftentimes, teachers will not require you to do an experiment twice (like you said with repeating the 2nd test with 6 hours), but in the real world, it is important for scientists to repeat their experiments (3 times recommended) to make their results more reliable.

Keep the first night as the control, and test the same group because it is a controlled variable.

If you are going to have different groups of teens tested, make sure that you are testing them separately (organize the data into different graphs, charts, data tables, etc.), and when you are testing them, make sure their general characteristics are the same (age, sex, etc.) You would also need to change your hypothesis/conclusion a bit because you would need to list out the differences between the groups then compare them (example- Group 1: 15-year-old girls, Group 2: 13-year-old girls, etc.) With this example, a good hypothesis would be: "If the amount of hours of sleep a girl gets decreases from 9 hours to 6 hours, then a 15-year-old's blood pressure would increase more than a 13-year-old girl's." You'll actually need two separate experiments since you are testing two separate things. For the second hypothesis, replace "blood pressure" with "memory impairment". If you want to do less work, stick with one. Ask your teacher about how many you need.

I hope this helped! Good luck.

Shelly Han