Data was collected from this survey-thread: t/174509
The percentage of left-handed people in the general populace is somewhere around 10% (between 7% and 15%, depending on the study). However, in the thread it is 19% and that with a sample size of 407 (lower if leaving out the votes for "Both"). I strongly believe that this is not a coincidence, that there is a reason for the disproportional amount of left-handed players.
Testing statistical significance
math
Objections and conclusion
Objection 1: The survey is inaccurate
Objection 2: You left out the votes for "both"
Objection 3: It is just a coincidence
Conclusion: Taking all of this into consideration, we'd expect the real k (number of left-handed people in the thread) to be even bigger than the results of the survey suggest, which further underlines the conclusion:
The observed asymmetry between left-handedness in our sample and left-handedness in the general populace is not coincidental. It is also not a result of how the survey was constructed, and not a result of an error in measurement. If we were to conduct another survey, the results would most likely be similar, possibly even more pronounced.
Questions
There is a reason for this and I wonder what it is.
- Why are left-handed people more drawn to this game?
- Is there a reason that right-handed people leave this game/don't find it in the first place?
- Maybe left-handed people that play this game are more inclined to use the forum?
- Are right-handed people prone to not answering surveys?
Possible explanations
A greater male tendency toward left-handedness and a predominantly male player-base explain the observed phenomenon
- B1rd
Left-handedness, being correlated with increased spatial ability and reaction time, gives left-handed users an inherent advantage compared to right-handed users, subsequently making them more likely to stick with the game due to their greater success
Voting is more appealing to people when their vote represents something that is generally considered interesting or unusual, causing an elevated number of left-handed people to vote in the thread.
Discuss.
The percentage of left-handed people in the general populace is somewhere around 10% (between 7% and 15%, depending on the study). However, in the thread it is 19% and that with a sample size of 407 (lower if leaving out the votes for "Both"). I strongly believe that this is not a coincidence, that there is a reason for the disproportional amount of left-handed players.
SPOILER
We can calculate the probability of this being a mere coincidence, using this formula
(1 - formula)
n = 346 (sample size, aka number of people that completed the survey, not accounting for everyone who voted "both")
p = 0.15 (general probability of being left-handed)
k = 78 (number of left-handed people according to the survey)
this leaves us with a final probability of ~0.01% (the probability that the results are a coincidence). And this is using the value 0.15, which is already on the upper end of all estimates regarding the prevalence of left-handedness. If we use the more accurate value of 0.1, the probability of this being a coincidence drops far below 0.00001%.
(1 - formula)
n = 346 (sample size, aka number of people that completed the survey, not accounting for everyone who voted "both")
p = 0.15 (general probability of being left-handed)
k = 78 (number of left-handed people according to the survey)
this leaves us with a final probability of ~0.01% (the probability that the results are a coincidence). And this is using the value 0.15, which is already on the upper end of all estimates regarding the prevalence of left-handedness. If we use the more accurate value of 0.1, the probability of this being a coincidence drops far below 0.00001%.
Objection 1: The survey is inaccurate
SPOILER
Some People in the thread interpreted the survey not as "are you left-handed or right-handed", but as "with what hand do you play?".
The consensus in the thread is that, if you interpret the question this way, the answer is determined by what hand you use to aim.
Looking at the comments, it is clear that many naturally left-handed people actually use a mouse/tablet with their right hand. On the other side, there is a lack of right-handed people that aim with their left hand. Even if you say that there is probably some right-handed people that answered left-handed, this is more than compensated for by the left-handed people that gave either "both" or "right-handed" as their answer.
The consensus in the thread is that, if you interpret the question this way, the answer is determined by what hand you use to aim.
Looking at the comments, it is clear that many naturally left-handed people actually use a mouse/tablet with their right hand. On the other side, there is a lack of right-handed people that aim with their left hand. Even if you say that there is probably some right-handed people that answered left-handed, this is more than compensated for by the left-handed people that gave either "both" or "right-handed" as their answer.
Objection 2: You left out the votes for "both"
SPOILER
The people who voted for "both" are most likely the people that interpret the question as "with what hand do you play" and think to themselves: "but I use both". There is no reason to assume that people who follow this line of thought are dis-proportionally more likely to be right-handed, hence leading to their votes getting lost and tilting the ratio in favor of left-handedness.
Furthermore, even if we assume that all the 56 votes for "both" were right-handed people and increase our sample size to 402, the result would still be statistically significant. (1 - 0.9931 < 0.05)
Furthermore, even if we assume that all the 56 votes for "both" were right-handed people and increase our sample size to 402, the result would still be statistically significant. (1 - 0.9931 < 0.05)
Objection 3: It is just a coincidence
SPOILER
I addressed this through testing the statistical significance. The probability of the results being coincidental is less than 0.1%. Of course there will always be a very small chance left that exactly this probability kicked in here, but it is unreasonable to assume that it did.
Conclusion: Taking all of this into consideration, we'd expect the real k (number of left-handed people in the thread) to be even bigger than the results of the survey suggest, which further underlines the conclusion:
The observed asymmetry between left-handedness in our sample and left-handedness in the general populace is not coincidental. It is also not a result of how the survey was constructed, and not a result of an error in measurement. If we were to conduct another survey, the results would most likely be similar, possibly even more pronounced.
There is a reason for this and I wonder what it is.
- Why are left-handed people more drawn to this game?
- Is there a reason that right-handed people leave this game/don't find it in the first place?
- Maybe left-handed people that play this game are more inclined to use the forum?
- Are right-handed people prone to not answering surveys?
- B1rd
SPOILER
According to http://psycnet.apa.org/psycinfo/2008-11487-004 , the approximated male to female odds ratio of being left-handed, is 1.23
Considering this, the numbers of the survey-thread are a lot more likely to be caused by mere coincidence as opposed to underlying mechanisms.
Considering this, the numbers of the survey-thread are a lot more likely to be caused by mere coincidence as opposed to underlying mechanisms.
SPOILER
The human brain has two hemispheres ("sides"), that are responsible for different things. For instance, one side is more responsible for mathematical functions, one more for verbal functions etc.
This is also known as Lateralization of Brain Function
One hemisphere of the brain is also in charge of the motor-functions of the opposing side of the body. For example, if you get brain damage in your right hemisphere your left side of the body will be more likely to lose function than your right side of the body.
Handedness supposedly gives insight in which hemisphere of the brain is more dominant. For left-handed people, this would mean that their right hemisphere is more dominant.
Interestingly, the right hemisphere is also responsible for spatial functions, such as 3D and 2D mapping.
And in fact, there are studies that found a statistically significant link between functional laterality and ability in spacial thinking.
http://www.ncbi.nlm.nih.gov/pubmed/15341431
http://www.ncbi.nlm.nih.gov/pubmed/7187636
Spacial thinking is an important part of playing this game, and an improved ability in this department is generally beneficial for performing well in osu.
I suggest that overall, left-handed people have an advantage when it comes to playing osu, which leads to them getting better faster, and quitting less frequently (explaining the results of the survey).
Further studies suggest that left-handed males have an advantage over right-handed males in terms of reaction time (no difference reported in females). Although this should only become relevant in higher levels of play, when the reaction time approaches the limits of what is humanly possible.
http://www.tandfonline.com/doi/abs/10.1 ... 0390220367
This is also known as Lateralization of Brain Function
One hemisphere of the brain is also in charge of the motor-functions of the opposing side of the body. For example, if you get brain damage in your right hemisphere your left side of the body will be more likely to lose function than your right side of the body.
Handedness supposedly gives insight in which hemisphere of the brain is more dominant. For left-handed people, this would mean that their right hemisphere is more dominant.
Interestingly, the right hemisphere is also responsible for spatial functions, such as 3D and 2D mapping.
And in fact, there are studies that found a statistically significant link between functional laterality and ability in spacial thinking.
http://www.ncbi.nlm.nih.gov/pubmed/15341431
http://www.ncbi.nlm.nih.gov/pubmed/7187636
Spacial thinking is an important part of playing this game, and an improved ability in this department is generally beneficial for performing well in osu.
I suggest that overall, left-handed people have an advantage when it comes to playing osu, which leads to them getting better faster, and quitting less frequently (explaining the results of the survey).
Further studies suggest that left-handed males have an advantage over right-handed males in terms of reaction time (no difference reported in females). Although this should only become relevant in higher levels of play, when the reaction time approaches the limits of what is humanly possible.
http://www.tandfonline.com/doi/abs/10.1 ... 0390220367
Voting is more appealing to people when their vote represents something that is generally considered interesting or unusual, causing an elevated number of left-handed people to vote in the thread.
SPOILER
As I mentioned before (in the last of the 4 questions I raised), it might not be valid to extrapolate the results of the survey to the general osu-playerbase. B1rd's theory claims that the phenomenon can be explained using this stance. According to him, the minority (left-handed players) is more interested in sharing their handedness than majority (right handed players). How strong this effect is, and if it explains the asymmetry completely, is open to discussion. Personally, while I think that it is a very valid point, I still believe that the asymmetry is too pronounced to be explained just by this, especially considering the points raised in objection 1. (many left-handed people voting for both and right-handed).
B1rd's theory can be challenged by selecting a random sample by hand and collecting their data directly, instead of letting the "sample choose itself"
B1rd's theory can be challenged by selecting a random sample by hand and collecting their data directly, instead of letting the "sample choose itself"
Discuss.
