Study: Are gender and fetish preferences independent or dependent?

[HornedSerpent]

I would vote but I'm none of these!  My fetish is nose shape/size.  And it is gender-dependent heavily for me.  (I like it on men)

[Sophie<3]

18 hours ago, AnonyMeows said:

 

What is your hypothesis, by the way?

 

17 hours ago, Kaze wo Hiku said:

Ok I FINALLY voted. And yes what is your hypothesis?

Thanks for voting, Kaze!  And sorry, I was going to post it all properly once enough votes have come in. But these are my hypotheses:

H1: There's a certain level of dependence between the variables gender and fetish preferences. H0: The variables gender and fetish preference are independent.

Like I said, the data is pointing towards H0 at the moment.

[Anonymouse]

Allow me to make an informal observation: there seems to be a lot of female identifying members here in general, or they're at least more likely to respond to a survey slash help out a fellow member.

[Sophie<3]

1 hour ago, AnonyMeows said:

Allow me to make an informal observation: there seems to be a lot of female identifying members here in general, or they're at least more likely to respond to a survey slash help out a fellow member.

That's what I thought, too, Anony! It shouldn't really impact the statistical test that much (I think), because what it does is compare expected frequencies (under H0 conditions, so independence) to observed frequencies. Meaning that it takes proportions into account. But obviously, a bigger n, so more replies, is always better, meaning that more replies from male identifying people would be highly appreciated!

This makes me wonder, though... what is the gender ratio on here in general? Do we know? Can we find out? Are there substantially more female identifying members than any of the other genders or does it just seem that way?

 

[Anonymouse]

1 hour ago, Sophie<3 said:

That's what I thought, too, Anony! It shouldn't really impact the statistical test that much (I think), because what it does is compare expected frequencies (under H0 conditions, so independence) to observed frequencies. Meaning that it takes proportions into account. But obviously, a bigger n, so more replies, is always better, meaning that more replies from male identifying people would be highly appreciated!

This makes me wonder, though... what is the gender ratio on here in general? Do we know? Can we find out? Are there substantially more female identifying members than any of the other genders or does it just seem that way?

 

Oh I know how it works, trust me, I've taken three statistics classes. It can be such an interesting topic of study.

Good question! I wonder if there's a way to get basic demographics of the overall population on the forum without combing through every single member profile for information. 

[bloom]

My hypothesis: Men are less likely than women to click on threads that won't have any sexy content in them. (NB folks excluded from this because there are not enough of us to make such a sweeping statement.) If I'm right, though, then unfortunately there is no way to measure this because we made a thread poll about how likely men vs women are to click on a thread like this.

You used to be able to use the forum search function to sort by gender, but it seems like that's no longer a search option, unfortunately, unless I'm blind and just can't find it.

[Cinnamon829]

Interesting poll and very professional looking. Good call, Sophie

I'm a female who prefers allergies (well, I can't really complain about cold sneezes cause I'll take anything I can get, but the whole illness/germs thing puts me off a little). I wonder why more males prefer allergies over colds while it's the opposite for females. Hm.

[bloom]

17 minutes ago, Cinnamon829 said:

I wonder why more males prefer allergies over colds while it's the opposite for females. Hm.

My guess would be that women are more likely to be into caretaking and that's a lot more associated with colds than it is to allergies. But I could be wrong!

[Anonymouse]

28 minutes ago, Junia said:

My guess would be that women are more likely to be into caretaking and that's a lot more associated with colds than it is to allergies. But I could be wrong!

Sort of my thought too.

[Joal 555]

I generally prefer random sneezes/sneezes without cause, so I voted for "I identify as male and like both equally/ cannot make up my mind" 

I wouldn't actually say I like either, but they are definitely equal for me.

 

 

 

[Mr. Black Cherry Berry Tea]

Yeah, I voted neither because cause of the sneeze is functionally irrelevant for me. But yeah, my casual unscientific guess would be that colds trend female and allergies trend male. Or maybe just the former and male-identifiers would be split evenly. Interested to see if anything statistically relevant emerges though! 

[Sophie<3]

2 hours ago, AnonyMeows said:

Oh I know how it works, trust me, I've taken three statistics classes. It can be such an interesting topic of study.

Oh, sorry! I'm sure you know more than I do, then! I've only taken two classes, and I'm not the best!  I try, though. And I agree that it's interesting!

2 hours ago, Cinnamon829 said:

Interesting poll and very professional looking. Good call, Sophie

Oh gosh, thanks!  

 

And I agree with the general idea that caretaking is what appeals to women about colds. It's true for me, anyway.

[Elektra]

Well...Okay, I  voted, but here's the thing : I like colds because the degree of miserableness most of the time higher, which gets me on the H/C level every time. Also more tendency for coughing in colds, which I like too.
On the other hand, am *not* into contagion, not because it grosses me out or anything (*waves to germ lovers*) but because having a low immune system I turned into a kind of germophobiac over the years, and it sorts of kills the move when I wanna care for someone. So caring for s.o with allergies is made easier.

PS : I don't see how i could have fitted all this into my vote

PPS : what do you mean I'm complicated and can't make a choice?

[David]

Male here, and I definitely prefer allergies because it's something that...like, can be avoided if need be.  I love the helpless aspect of allergies but don't really enjoy the dull, constant suffering of colds.

[Sophie<3]

Alright guys, guess what. There are still 3 categories below 5. Basically, we're missing four votes in the right places. But I realized that 93 people have already voted now, which is amazing!! And I'm sure you're just as tired of waiting as I am. Also, I'm sure there's a great statistical reason behind the "no cells below 5 units" rule, but at this point in time, I don't see it making a huge difference for us, so I just did the calculations anyway.

Test: Chi-Squared Test of Independence
Purpose: To check whether the variables Gender and Fetish Preference are independent/ do not affect each other
Hypotheses:

H0: The variables X (Gender) and Y (Fetish Preference) are independent

H1: The variables X (Gender) and Y (Fetish Preference) are dependent / There is some dependency between the variables X and Y

Statistical Analysis:

(Note: I hope nobody takes offense at the word "other". I personally hate it, but needed to put something short and couldn't think of a good alternative. :( Sorry! I hope it's okay!)

Observed Frequencies:

	Female	Male	Other
Colds	22	4	6
Allergies	16	15	3
Both	15	8	4

 

Expected Frequencies under H0 conditions (independence):

	Female	Male	Other
Colds	18.2365	9.2903	4.4731
Allergies	19.3763	9.8710	4.7527
Both	15.3871	7.8387	3.7742

 

Chi Square value: 8.2366

(Note: I rounded to four decimal places to make things a little easier for us.)

(This is the formula, in case anybody cares. )

Compared to the critical chi square value with (k-1) = 2 and (m-1)= 2, with alpha 0.05 (so chi square (4, 0.95)) = 9.488

9.488 > 8.2366, meaning that H0 cannot be rejected! The data suggests that the variables X and Y are independent of each other.

 

Notes: Bummer. BUT I feel like we're actually approaching significance here! See, our chi square would have to be greater than 9.488 for me to be able to reject the null hypothesis.
And basically, we're not that far off, you know? The way it works is that the bigger the difference between the observed and the expected frequency, the bigger our chi square value. And there were notable differences for some cells. The female preference for colds over allergies and the male preference for allergies over colds became quite visible, I'd argue!

But hey, we need to be scientific. So at this point, I really have to say that the traits appear to be independent. If more people decide to vote, though, I'd be happy to perform the test again.  But like I said, no clear dependence at this point.

 

[Sophie<3]

UPDATE:

I feel a bit horrible for doing this, but like everyone doing science, I'm desperate for significant results! :lol: So I just reduced the table to the classic 2x2 set-up in the name of science. And guess what. We have significant results!!

Here are the tables for the 2x2:

Observed Frequencies:

	Female	Male
Colds	22	4
Allergies	16	15

 

Expected Frequencies:

	Female	Male
Colds	17.3333	8.6666
Allergies	20.6666	10.3333

 

The new chi square value is: 6.9305

Compared to a critical value of 3.841 (with alpha = 0.05).

6.9305 > 3.841, meaning that I can reject the Null Hypothesis (with an error probability of 5 %). The data suggests that the traits Gender (Male or Female) and Fetish Preferences (Colds or Allergies) depend on each other!

After this, I got a little cocky and checked the critical value for alpha = 0.01. And it's 6.635.

Meaning we still have significance at this level, because 6.9305 > 6.635.

So that means, I can now say that gender and fetish preferences are dependent of each other with 99% confidence!  

Honestly, I'm so sorry I had to exclude anybody who either doesn't fit the male/female gender binary or has no clear preference for either colds or allergies. You guys are all amazing and I'd like to say thank you so much again for your votes!  It just appears that the effect only becomes visible in a 2x2 set-up, because only people who identify as female generally show a preference for colds over allergies and only those identifying as male generally show a preference for allergies over colds.

So now, this suspicion of mine has been confirmed. It's canon, so to speak.

PS: I hope anyone else is excited about this, because I sure am!  

Edited July 14, 2016 by Sophie<3

[•.*°•☆. Q .☆•°*.•]

I feel like by excluding people from your results you at that point give a somewhat biased and inaccurate portrayal of the fetish community's preferences in relation to gender?

like rather than answering "does gender effect fet preferences" you then actlly answer "does gender effect fet preferences *in forum members who are decidedlymale or female and who absolutely are for either allergens OR colds" which is all in all a different question to the original by quite a lot?

 

[bloom]

I dunno, I think it's pretty fair to exclude nonbinary people from an attempt at gathering data like this, simply because a. nonbinary people are not exactly a dime a dozen to begin with (even though we seem to have a pretty surprisingly significant presence on the forum), and b. 'nonbinary' is a very, very wide umbrella and people who are nonbinary experience our genders in scads of different ways. (Not that binary people don't, but I feel like the experiences of nonbinary people are often even more nebulous and confusing.) Don't get me wrong, 'nonbinary' is a great umbrella term in most other contexts, but I think in a case like this it casts a pretty wide and confusing net vis a vis collecting data in this way.

Sophie, I'm really glad you made this thread and compiled the data like this, even though you had to use a 2x2 table instead of a 3x3 one. I love pretty much any discussion and analysis of gender in relation to the fetish so this thread has been right up my alley  

[Sophie<3]

9 hours ago, Quing Nerd said:

I feel like by excluding people from your results you at that point give a somewhat biased and inaccurate portrayal of the fetish community's preferences in relation to gender?

like rather than answering "does gender effect fet preferences" you then actlly answer "does gender effect fet preferences *in forum members who are decidedlymale or female and who absolutely are for either allergens OR colds" which is all in all a different question to the original by quite a lot?

 

Fair point! Sorry, I got super excited there about the 2x2 for a moment, but you're right, of course! There appears to be a very specific effect, namely a dependence between gender and fetish preferences when looking only at male/female people and cold/allergy preferences. This I personally consider very excited!
But it's still true that when taking all people and all preferences into account, the two appear to be independent.

See, the best thing would have probably been to consider my initial test preliminary research and then start a new thread for the 2x2. I would've gladly done this, too, but I felt like people were probably getting sick of me pestering them for votes.  

9 hours ago, Junia said:

Sophie, I'm really glad you made this thread and compiled the data like this, even though you had to use a 2x2 table instead of a 3x3 one. I love pretty much any discussion and analysis of gender in relation to the fetish so this thread has been right up my alley  

Thanks so much!  I had a lot of fun with it!