Masturbation handedness, porn fandom, age, and the color of that dress

The argument: which hand do people use to masturbate?

Working against my self-imposed theme, this first post is not from an argument with my brothers, but a friend at a bar. It will likely also set a record on one end of the maturity spectrum of these, which is yours to determine.

Somehow, some conversation led to discussion of which hand men use to masturbate. The antagonist in this story, in a context that I believe made sense, assumed people would use their dominant hand (so typically the right hand). I disagreed, given the dominant hand is needed to operate a computer (as porn and other stimuli are prominent these days). This formed a basic hypothesis: masturbation handedness is determined by habits while masturbation. Therefore, correlates of these habits would also be predictors. One chief correlate would then be age, due to the relatively recent advent of computers and remote controls etc. Argument ensued, and brought us here:

I collected responses in a 2-minute online survey from 189 people (67.8% male; 77.8% White; 30 year old on average with 8 year SD), in the United States (for $15). (Dataset available here - the things we sacrifice for open science).

To cut to it...

First, sanity check:

*From now on, in all of these figures, the y-axis is the likelihood of right-hand use to masturbate (so lower values on the y-axis mean more likely left-hand use, .5 being the boundary between, as a 50/50 chance of either hand). I am not posting stats as most folk won't care, but anything I am reporting is significant, unless I state otherwise, and is using standard logistic regression.

'Handedness' being the typical definition of handedness. This is interesting, actually. Left-handed folks are more divided than anticipated. These analyses have been run with many of the other correlates here discussed in the model, and seem to hold. If you think of any reason this would happen, let me know your thoughts. This is especially interesting given there are many reasons, even for a righty, to use their left hand in addition to handedness, as we see in a moment. Though some friends have pointed out ambidexterity is higher among lefties, so this could be the cause.

To the point. Age did relate positively to right hand use:

Importantly, looks like I was half-wrong, the data don't lie. We're looking at a decrease towards 50% left vs. right hand use among the youngest of the sample (range 19 to 57 years). 

Before we go there, first I must note, I wanted to measure the mediation of this by computer use - the idea being that age is explained because computers/remotes weren't around to demand your dominant hand back in the day.

At first glance, okay, how often they use a computer seems to mirror this same effect (used this variable more generally, but identical effect for porn use frequency):

However, interestingly, age does not relate to computer use to masturbate, and they both predict masturbation handedness over and above one another. Thinking about this more, this is not incompatible with the hypothesis, but more so forces us to flush it out more. I would think the most important mediator is whether they were computer users when they began. Collect another dataset? Maybe. For now we can glance at this:

And we can imagine that as time continues to pass, handedness should stabilize at some point, and lose its relation to age as computer-users become the norm.

To summarize, the relationship between age and masturbation handedness is real, and rather strong, with around 50% of the younger generation using their left hand, as opposed to the near unanimous righty preference of the older generation. While computer use is an equally potent predictor of masturbation handedness, it does not mediate the effect of age. For now, I'd speculate my prior statement, that what matters is if you used a computer when you learned, as older righties may have learned with the right, but now use computers. Impressively, these effects existed over and above one another, and many covariates, such as handedness and demographic information. This applies for each: age, computer use, and one below... 

As for alternate explanations, the few that have been recommended to me by friends are seem potential to a degree, but I am unsure they could fully underlie effects. First, there is the idea people use their non-dominant hand due to its unfamiliarity (feels as though it is someone else; similar to the child-humor sit-on-your-hand-so-it's-numb theory), which I doubt because the unfamiliarity would fade rather quickly. Second, there is a recent mention that in olden times folk used their left hand to do things that were 'impure', so this could be the case (which seems true; Roman left-handed masturbation; wiping after defecation nowadays). Or to strip morality from it, maybe people prefer their non-dominant hand because they simply don't want to dirty the hand they use for everything else. These are all speculation of course, so feel free to provide any input... it's a puzzler.

But before exiting this topic, another relationship seems to reflect a lot of this...

Just sayin'.  

___________________________________________________________

Before calling it quits after the main hypothesis test, let's have a glance at general demographics.

First, some general demographic differences I can't knock out controlling for anything... which makes them pretty interesting (all significant).

And last, but not least, without the slightest serious intent or hope, I had them answer this item...

And somehow...

I personally see Lavender and Bronze, but have to note the accurate ones (Black/Blue) are in the half/half righty/lefty camp. Might mean something...

There is so much more to analyze in this dataset outside of handedness, such as about general masturbation habits, their relations, and relations to demographics and such. I'll try to return to it sometime. A foreshadowing is the vast relationships being raised without a father has with many habits, and also demographic information. 

Hope this was as at least as interesting as it was weird.

~ Don

*Data collection was done with Qualtrics over Amazon Mechanical Turk. Analysis with R, figures via ggplot2.