Anabolic steroids have an incredible impact on athletic performance. Even if just 20 athletes took a moderate dose of anabolic steroids for a few months, every single one of them would experience enhanced performance.
Blank Slate and anabolic steroids have that in common.
Like steroids, Blank Slate has an enormous impact on knowledge retention. In our research with patients with Alzheimer’s disease, every single patient who used Blank Slate for a month experienced memory improvement. In our research with medical school students, using Blank Slate resulted in a 67% boost in knowledge retention on average.
Things like anabolic steroids and Blank Slate are said to have a large effect on outcomes like athletic performance or knowledge retention. And the thing about large effects is, researchers don’t need to demonstrate them in thousands of people. Because anabolic steroids and Blank Slate are so effective at what they do, their impact is clear with even a small sample of people.
So, this post isn’t actually about steroids. It’s about how we at Blank Slate choose our sample sizes, which refers to the number of people we recruit for participation in our research studies.
How many people make up a small vs. a large sample?
There are no hard and fast cutoffs for what constitutes small or large samples. Instead, researchers rely on a statistical technique to estimate how many people they should recruit for a given study. But that doesn’t stop people from having incorrect intuitions about sample sizes: many consider fewer than 50 or even 100 participants to be small, and hundreds or thousands of participants to be large.
Small samples can accurately reveal large effects.
I’ll say it again: anabolic steroids have a massive and ubiquitous impact on athletic performance. A small sample of even 20–40 participants could provide a clear indication of the effect of anabolic steroids on muscle gain.
Large samples are needed for revealing small effects.
Not all effects are as pronounced as those of steroids. Let’s say you’re investigating the impact of multivitamins on athletic performance. Certainly some people with vitamin deficiencies might see a boost in performance from taking a multivitamin, but many people wouldn’t benefit from it at all. In this case, the multivitamin is having a small impact on athletic performance that would only be revealed if hundreds or even thousands of athletes were studied.
We don’t make the rules, statisticians do.
Long ago, statisticians invented a method of determining the right sample size for research studies: a power analysis. This statistical tool recommends a sample size based on how large of an effect the researcher expects to find and how certain the researcher wants to be that they truly found that effect. Researchers studying large effects (think: anabolic steroids) only need a small sample size. But researchers studying small effects (think: multivitamins) need a large sample size.
Sample sizes at Blank Slate
Like anabolic steroids, Blank Slate produces very large effects on knowledge retention. When an effect is that pronounced, it can be reliably detected in a small sample of participants. At Blank Slate, we follow the rules of wise, old statisticians and conduct power analyses to inform our samples sizes. These rules usually require us to recruit between 20 and 100 participants for each of our studies.
Understanding proper sample sizes is key to conducting meaningful and reliable experimental research. Small sample sizes are suitable for detecting large effects that stand out clearly, while large samples are necessary for detecting small effects that are more subtle. By selecting an appropriate sample size based on the size of the effect you’re expecting to find, researchers can ensure that their findings are valid and relevant to the real world.
- Andrews, M. A., Magee, C. D., Combest, T. M., Allard, R. J., Douglas, K. M. (2018). Physical effects of anabolic-androgenic steroids in healthy exercising adults: A systematic review and meta-analysis. Current Sports Medicine Reports, 17(7), 232–241. doi:10.1249/JSR.0000000000000500
- Marin, A., Smith, A. M., Decaro, R. E., Feinn, R., Wack, A., Hughes, G. I., Rivard, N., Umashankar, A., Turk, K. W., Budson, A. E. (2023, July). Promoting memory retention in older adults with and without Alzheimer’s Disease using the mobile app Blank Slate. Poster presented at the annual Alzheimer’s Association International Conference, Amsterdam, the Netherlands.
- McHugh, D., Feinn, R., McIlvenna, J., & Trevithick, M. (2021). A random controlled trial to examine the efficacy of blank slate: A novel spaced retrieval tool with real-time learning analytics. Education Sciences, 11(3), 90. https://doi.org/10.3390/educsci11030090