Everything you need to know about my data, how I got it, and what I did with it
I am a survey and data noob. I know nothing about doing surveys, I just do them and then look at basic correlates.
The data came partially from my own traffic (twitter, tumblr) and partially from my friend Eric’s traffic, redirected from his personality testing website.
I removed all answers that reported not being in a serious relationship, as well as a very small number (<10) of answers that appeared suspiciously inconsistent (e.g., a fundamentalist monogamous trans 65 yo with 3 active partners). This left me with a sample size of 993; about 550 cis women and 396 cis men, with the rest reporting as trans or other. For all gender information I used cis women and men, due to the low trans number. Identified (not birth) gender was used for determining straight/gay relationships.
The data on relationship length was very detailed, and I reduced it to 7 categories of length for most graph purposes – 1 month, 8 months, 1.5 years, 3 years, 6 years, 12 years, and 20+ years. (Categories were “at least” – so someone in a 2 month relationship were categorized as 1 month.)
I also am unsure what numbers are needed to reach significance, so take all discussions about importance with a strong grain of salt. I tried to feel out significance by seeing the correlation between questions I strongly expect to be consistent, and by looking at how strong the correlations are compared to others.
Nearly all questions were asked on a 1 (disagree) – 5 (agree) scale. Assume this unless I clarify otherwise.
I measured 7 different scores, 4 positive and 3 negative. Each score was the average a few similar questions. Expand the score to see exactly which questions contributed.
My relationship makes me happy
I am in love with my partner
My decision to enter into a relationship with my partner was a good one, in hindsight
My relationship brings me emotional stability and security
My relationship brings me practical (in regards to finances or life plans) stability and security
I share the same hobbies with my partner
My partner and I agree on whether or not we want children
I am satisfied with the division of labor in my relationship
*I am least confident about this one, and grouped the questions together afterward
Do you consider this relationship to be the most promising, most important, or best relationship you’ve ever had?
My relationship is the most fulfilling thing in my life
My relationship takes priority above everything else
My partner completes me
, and Excitement
My partner causes me to feel butterflies, nervousness, or romantic excitement
I feel thrill and excitement in my relationship
I feel a great deal of passion in my relationship (not including sexual passion)
My relationship brings me entertainment and novelty
I sometimes worry that my partner will leave me for someone better
I am more needy than my partner
I am insecure in my relationship
I feel jealousy or possessiveness
My relationship causes me anxiety
My relationship causes me grief or sorrow
I take care of my partner more than my partner takes care of me
I suspect my relationship might not last much longer
I fight with my partner
My partner and I have different values
I have broken up or decided to break up with my partner
I would like to leave my partner
I sometimes wish I had never begun a relationship with my partner
, and Undesirability
I doubt I could find someone else who completes me as much as my partner does[
My partner is more desirable (to general society) than I am
My partner could have had chosen to date people who are better (worth more, more desirable) than I am
negative of:I could date someone better (worth more, more desirable) than my partner if I wanted to
I chose to combine these questions for ease of presentation.
and good things:
The lowest sample size was 31, for men in 20+ year relationships. Sample sizes for 12-20 and 20+ year categories were between 35-45ish per gender, each.
Everything good decreases over time, with maybe a slight increase between the 12-20 and the 20+ year marks (due to kids leaving the house?)
In hindsight I dislike the practical score and am going to ignore it. It isn’t present in the Men and Women direct comparison graphs.
Men peak in nearly everything at the 3-6 year mark, actually – or maybe it’s just a climb from their sudden drop at the 1.5-3 year period. This is strange to me, especially because women don’t show the same dip-and-climb.
Could this be due to children? Do couples have kids during the 1.5-3 year period, leading to men handling it poorly, thus leaving the relationship before the 3-6 year mark, leaving only the remaining happy men to bring the data back up? Does this indicate that there’s a ‘danger period’ for men, and if they get past it they’re good?
0-3 years, women report higher priority and lower happiness, compared to men. This flips at 3 years in, when men report being less happy and giving the relationship higher priority.
Women report higher excitement overall, with a more stable decline.
And bad things:
Men report much higher undesirability and women report much higher insecurity. Both of these things even out over time, ending up swapped at 20+ years. Is this consistent with the trope that men age better?
Men report higher Badness at early and late stages, with women most unhappy right in the middle. This is relatively close compared to the other two scores.
and good things:
There were very low sample sizes for 5/5 polyamory; only 10 responses for women and 20 or so for men. Interpret the far right side of the graph with skepticism.
I asked people to rate how monogamous or polyamorous they identified, with 1 being monogamous and 5 being polyamorous.
The overall trend seems to indicate that medium-poly relationships do the worst, particularly for men. Full poly relationships beat out full mono relationships on happiness, but come slightly short on excitement and priority.
Women see a similar jump on happiness, but this occurs at 4/5 poly, not the 5/5 poly required for men. Women also don’t have to be as poly as men to report excitement increases, and seem to get total larger excitement boosts from polyamory. They also don’t get the 5/5 poly boost on priority that men do, though again, we’re dealing with low sample sizes here so we don’t know for sure.
It seems that men have a higher poly dedication threshhold to start reaping rewards, while women can be happier at mid-ranges of poly.
and the bad:
It appears polyamory widens the gender insecurity gap, with women getting a slight increase and men a bigger decrease. Polyamory does seem to (be correlated with) lower feelings of undesirability. The Bad score is the coolest, with women peaking at 3/5 and men peaking at 4/5.
and good things
the X axis marked religiousness, with 1 being ‘not religious’ and 3 being ‘very religious.’
Men show generally more positive effects from increased religion, with the strongest difference being the Priority score. The more religious a man is, the more his partner takes importance in his life – by over a half point! Women’s priority remains low and stable regardless of her religiousness.
Nonreligious men are less excited about their relationships than nonreligious women, but this flips for the religious, where men report higher excitement. The same goes for happiness.
So in general, religion has minimal effects on women but strong positive effects on men. I’m really not sure how to interpret this and would love to hear theories in the comments.
and the bad
Turns out religion doesn’t help men’s insecurities that much, with religious men reporting greater insecurity and religious women feeling more secure their heathen counterparts. Very religious women and men report equal insecurity; it’s pretty rare to see that gap closed.
The Bad score sees a slight drop for both, but more pronounced in men. Religious men report a higher sense of undesirability than nonreligious men, with religious women reporting only a slight increase.
So basically, religion might help men with happiness and excitement, but correlates with more insecurity and undesirability, while reducing women’s insecurity.
I also asked about abuse. The Abuse Score consists of the average of questions that asked about physical harm ever and physical harm over the last month, emotional manipulation as a pattern and emotional manipulation recently, and sexual assault.
The more religious a man is, the more likely he is to report both abusing and being abused. Religious women only report being abused more, with a drop in giving abuse (but still not as low as nonreligious women).
Everybody reports being abused more over time. Women report being abused more than men, with an exception in the beginning of the relationship. Do people wait longer to abuse women than men?
Up until 1.5 years, men are more likely than women to report both being abusers and being abused. Women then start to report higher levels of giving and receiving abuse (compared to men), but after this the received abuse stays stable (with women as more abused), and the abuser switches again back to men.
For polyamory, women report a weird sharp decline at 4/5 poly in both abuse given and received, but men stay high in both of these. I don’t know what this means.
Some notes: I haven’t looked at the difference between emotional and physical abuse.
It’s also likely that some groups are more likely to admit abuse than others (e.g., maybe religious people are much more likely to consider and report ‘regular’ things as emotional manipulation). It’s possible that low scores are just people in denial.
Also fascinating is the correlations between self esteem and abuse. The higher reported self esteem, the lower the chance of abuse given and received – until we get to 5/5 self esteem, where we enter Narcissist Land, apparently. 5/5 Self Esteemers report a sharp increase in abuse given, and a slight increase in abuse received.
Things are pretty even here, but anomalies include the drop in priority score in men between self esteem points 1 and 2. Men who have 1/5 self esteem are much more likely to rate their relationships as most important (with questions like ‘most fulfilling thing in my life, completes me, best relationship I’ve ever had).
Everybody’s scores fall a bit between 4/5 and 5/5 Self Esteem, except for women’s Excitement Score, which keeps going up, while men’s only drop. Weird.
Gender Differences by Question
Scores are differences, marked by who got it higher. Respondents voted on a 1-5 scale; the higher the score, the more they agree.
The first initial is the gender that agreed with the question more. The number is the difference in ratings. The numbers in parentheses are the absolute averaged numbers, by gender.
F I am more needy than my partner: .65 (F 3.19, M 2.54)
M I am polyamorous: .52 (F 1.66, M 2.18)
M My partner is more desirable (to general society) than I am: .51 (F 2.74, M 3.25)
F My partner causes me to feel butterflies, nervousness, or romantic excitement: .32 (F 3.31, M 2.99)
M My partner could have had chosen to date people who are better (worth more, more desirable) than I am: .32 (F 2.83, M 3.15)
F I feel jealousy or possessiveness: .30 (F 2.60, M 2.30)
F I have broken up or decided to break up with my partner: .28 (F 2.27, M 1.99)
F I share the same hobbies with my partner: .25 (F 3.11, M 2.86)
F My partner and I agree on whether we want children: .25 (F 4.13, M 3.88)
F I feel a great deal of passion in our relationship: .24 (F 3.68, M 3.44)
M I take care of my partner more than my partner takes care of me: .22 (F. 2.84, M 3.06)
M I am insecure in my relationship: .21 (F 2.47, M 2.26)
M My partner and I have different values: .21 (F 2.67, M 2.88)
F My relationship causes me anxiety: .16 (F 2.70, M 2.54)
F I am sexually compatible with my partner: .16 (F 3.85, M 3.69)
M I am satisfied with the division of labor in my relationship: .15 (F 3.42, M 3.57
F I feel thrill and excitement in my relationship: .13 (F 3.38, M 3.25)
F I doubt I could find someone else who completes me as much as my partner does: .13 (F 3.41, M 3.28)
M I have performed nonconsensual sexual acts upon my partner: .13 (F 1.10, M 1.23)
F I am in love with my partner: .10 (F 4.23, M 4.13)
F My relationship causes me grief: .10 (F 2.31, M 2.21)
M I intentionally physically and nonconsensually harm my partner: .9 (F 1.15, M 1.24)
F My partner has performed nonconsensual sexual acts upon me: .9 (F 1.28, M 1.19)
M Sometimes my partner is emotionally manipulative: .8
F My partner has exhibited a pattern of emotional manipulation in our relationship: .8
M I suspect my relationship might not last: .7
F My relationship makes me happy: .6
M My decision to enter into a relationship with my partner was a good one, in hindsight: .6
M I have exhibited a pattern of emotional manipulation in my relationship: .6
F Our relationship is very sexual: .5
F My relationship brings me emotional stability or security: .4
M I could date someone better (worth more, more desirable) than my partner if I wanted to: .4
F I fight with my partner: .4
F My relationship brings me entertainment and novelty: .3
M Amount of serious partners: .3
F I would like to leave my partner: .3
F I sometimes worry that my partner will leave me for someone better: .3
M My partner completes me: .2
F My relationship is the most fulfilling thing in my life: .1
M Sometimes I am emotionally manipulative: .1
My relationship brings me practical/financial security: 0
My relationship takes priority in my life: 0
My partner has intentionally physically and nonconsensually harmed me: 0
I sometimes I wish I’d never begun the relationship: 0
I have intentionally physically and nonconsensually harmed my partner: 0
In summary, time is not kind. Relationships show an almost universal decrease in everything good the longer they go on.
Poly is hard, and you have to go all the way to make it work – especially for men. Religion is also great, if you’re a man.
Women get more excited and insecure, men feel undesirable.
I’m going to write a blog post with much more conjecture about the differences in how men and women approach relationships. But for now, if you want to see the rest of the individual graphs, check them out here. The “relationship length” is sorted by months, and with slightly different average points.
If you want to look at the data yourself, download here.