a) Experimental protocol
The operation of this application has two drawbacks. First, a fraction of the users. First, tinder free feature don’t let scroll through the whole list of people who have liked a profile. Only members who have purchased Gold feature will be shown a grid of people who have swiped right. The number of stored swipes/likes is usually indicated with the following symbols: 1, 3 +, 10 +, 25+, and 99+. So despite knowing the exact number accurately, we can have a relative estimate of the size of the set of swipes stored.
Since my dummy profiles are on the free tier, I have had to swipe through a list of opposite-sex profiles without knowing which potential matches have liked them.
Second, we cannot know if that every user we liked was also presented with our dummies profile and, as such, these values offer a lower-bound on potential matches. Therefore it does not allow us to establish the acceptance/rejection rate. Only the raw (and approximate) amount of swipes/likes stored by profile.
I browsed profiles within a 100 mile radius and clicked like for each one.
Settings:
Maximun distance: 100 miles.
Age range: 18-55+
Localitation: Western small-sized town (Population: 151.000).
Data-collection period: 5 days.
Dummy profiles:
This mating experiment is composed of four dummies profiles. One highly attractive girl (female Z, r> 8 on a 1-10 scale ), one medium-attractive girl (female G, r≈ 5); and one highly attractive guy (male Z, r> 8); One medium-attractive guy (male B, r≈ 5). It has not been inserted any dummy into the below-medium and moderate attractiveness spectrum.
a) High-attractiveness dummy female profile (r>8 points on a 1-10 scale):
- Female Z:
Highly attractive female Z
b) Medium-attractiveness dummy female profile (r≈ 5):
- Girl G
c) High-attractiveness dummy male profile (r>8 points on a 1-10 scale):
- Male Z
Highly attractive male Z
d) Medium-attractiveness dummy male profile (r≈ 5):
- Male B
Medium attractiveness male B
b) Results
- Female Z
Likes = 338 Matches + Likes (+99) ≥ 437
- Female G
Likes = 310 Matches + Likes (+99) ≥ 409
- Male Z
Likes = 228 Matches + Likes (+3) ≈ 231
- Male B
Likes = 3 Matches.
TABLE OF RESULTS
Is that dude a 5? He looks like a 6 but yeah those results are depressing for him. I guess it’s good to know I get more mutual likes than he does off Tinder lol. He really could lose weight though that’d help for sure.
He was rated in a survey on this blog with a “2.59” on a 1-5 scale (i.e “5.2” on a 1 -10 scale). But as you say he could be maybe a “6” if we could assess a ranking through male population Z-scores. His body is not endomorphic, although he has a slightly plumpy face. Have you seen how’s the chick he got in his real life? Lol
@Tyrion
I see thanks for the response!
@ Travis,
What I meant, it is curious as a male mating loser in a Lek context (e.g. male B), has been able to subvert sensory/cognitive female bias towards a more favorable accommodation by courtesy of a non-Lek environment (Not so prone to empower female polygynous delicacies).
For example, non-Leks social network metrics include: the focal’s reach (the number of friends of friends, the number of individuals connected to a focal via two steps); clustering coefficient (the tendency for a focal’s friends to also be friends with each other); exclusivity= 1/(degree of a focal’s partners); male–female degree correlation (males that mate with many females tend to mate with females who also mate with numerous other males) then this should substantially reduce overall sexual selection; centrality (i.e. proportion of shortest pathways between all individuals in the group that go through a given individual), etc.
All very interesting. Someday we should discuss this topic.
@ Tyrion
So if I understand correctly you are saying he is getting/or gets better results offline than online via his social circle, recreational activities, etc?
@ Travis,
Yes, social circle is social network. Keep in mind that online dating, bars, nightclubs, etc are Lek environments, a scenario where the number of potential sample men is huge, and average female experiences enough highly attractive males to be choosy.
If all females choose mate within a Lek landscape (if there was no underlying social network structure), then all females would be choosy and sexual selection would be very strong. An example of how social network structure can affect adaptive behavior, in this case, female mate choice.
Outside from mating Leks, social network structure can be particularly interesting when we consider not just direct nodes connections among individuals, but also, indirect connections. The idea is that outcomes can depend on not just who a focal interacts with, but also with whom the focal’s nodes interact.
For example, a micro social network structure is set up when a focal male is connected to some female nodes; always considering to how many female nodes he’s linked, how many interconnections link these female nodes with other male nodes, etc.
Said in a more colloquial way, a focal male would have an optimal landscape if he is linked to a female node (Imagine she is a neighbor, a co-worker, a classmate or something similar), which she is not in turn linked to more male nodes that could offer a competitive mating obstacle for the focal man.
B is a lucky guy, I guess he had a very favorable social network, and his girlfriend was quite disconnected (low encounter-link frequency with highly attractive males).
In other types of network conditions, or approaching females within a Lek environment, hardly he would have found an willful cute girl.
That hypothetical female could be restricted to pair off his focal node, due to the scarcity of available male nodes around her immediate network environment (as long as we establish that she will not be visiting Leks environments, of course).
I replied to you on Youtube, but I preffer to try and write something more coherent here.
As I said in my comment, I think that the participant’s ratings might be biased by the characteristics of the population that voted them.
I don’t believe Male B to be unattractive, but as I said, I think he looks significantly older than the other participants, and his pictures are worse.
Female Z’s pictures are staged, and you can tell. She looks good in the and like she puts work on how she presents herself.
Female G’s pictures are of worse quality, but in most of them, she looks like a “put together” woman. Also, it’s worth mentioning that OkCupid experiment about how the disparity on opinions towards a female’s attractiveness, caused her to have much more success Online dating. Which means that 2 women might be rated an average of 5, but the one that got more disperse ratings from men, is more succesful. This woman is clearly not conventionally attractive, has short cooured hair, and other characteristics that probably make her a “Oh, heck yes” or a “Oh, God no”, to men. In her case, that’s prefferable over getting all “Oh she is cute I guess”.
Male Z’s pictures just show a very attractive man who is not trying too hard, but puts some efford, is social and seems to know how to have fun. Also, Dog picture bonus. Why? 1. Halo effect, 2. He chooses good angles, shows pictures with others smiling, has a picture with a Dog (It’s been proven to improve your chances online), and is very clearly physically fit.
Male B’s pictures show a much older guy, with angles that don’t flaunt him, make him have a double chin, he hardly smiles in any of them. He looks very serious and not super friendly in many of those. He does have a good style and haircut and such, but it’s just… his pictures don’t make him any favour!
I think the population of refference that rated this people is not reliable for this kind of experiments, on a dating app that’s based solely on people’s appearence. Taking a wild guess, the majority of the voters were men. This is personal belief, but I think men are more forgiving of other men’s looks.
@ Fish,
“[ ………..I don’t believe Male B to be unattractive, but as I said, I think he looks significantly older than the other participants, and his pictures are worse.”
Attribution bias. Male B is 28 years old and Male Z 32 years old.
” [ ………. Male Z’s pictures just show a very attractive man who is not trying too hard, but puts some efford, is social and seems to know how to have fun. Also, Dog picture bonus. Why? 1. Halo effect, 2. He chooses good angles, shows pictures with others smiling, has a picture with a Dog (It’s been proven to improve your chances online), and is very clearly physically fit.
Male B’s pictures show a much older guy, with angles that don’t flaunt him, make him have a double chin, he hardly smiles in any of them. He looks very serious and not super friendly in many of those. He does have a good style and haircut and such, but it’s just… his pictures don’t make him any favour”
Red herring argument (i.e. Ignoratio elenchi). I acknowledge the statistical power of these amateur studies is improvable, as long as the pictures are not standardized (i.e. taking them under nearly identical conditions). I collect images for several sources (While obtain standardized images would require funding). So your donation will be welcome Lol.
Anyway you can’t see the Forest for the trees!
“I think the population of refference that rated this people is not reliable for this kind of experiments, on a dating app that’s based solely on people’s appearence. Taking a wild guess, the majority of the voters were men. This is personal belief, but I think men are more forgiving of other men’s looks.”
Irrelevant, since only one of the 4 stimuli (male B) were surveyed in one rating trial on a old post. By the way, which lacks any utility for the present experiment. The claim here was determine the mate choice frequencies for average looking morphs on a dating site along with two high-attractiveness individuals. Which gives results tuned with the rest of the mating experiments.
What is interesting, is where objective data suggests something meaningful of how mating choices are distributed in a population, in such a way that lends conclusion to female skew choices being aberrantly *hipergamic* through measures of kurtosis on mating leks (e.g. Tinder, Pof, Badoo, Okcupid, Lovoo, Pubs/nightclubs, etc).
Therefore mating distribution for males granted by skewed female preference functions give a qualitative understanding of the real distribution given the Gini coefficient near to G=1.
And this is exactly what we should expect as sexual selection *must * be zero sum, as female function is limiting (with scale invarient, isomorphic manifestations of a contrary agency).
Human sexual selection implies greater selectivity in tending larger female populations towards smaller male populations, if these preferences can be interpreted as qualitative filter thresholds.
So forget your personal incredulity for a moment, and realize studies lies data supports something universally meaningful of how matings are distributed in a population, in such a way that lends conclusion to less varience between females and an obscene variance between males. Unless a female is exceptionally ugly, she will have sexual access to the same pool of males as her betters: A chubby girl G achieves the same mating pool as the hot girl Z and, please note twice as much as the model-looking male!. While 98.7% of the female population rejects an average looking male as a potential partner!
Consider Pareto’s principle as an analogy
‘A chubby girl G achieves the same mating pool as the hot girl Z and, please note twice as much as the model-looking male!. While 98.7% of the female population rejects an average looking male as a potential partner!’
I definitely agree with the majority of your findings, but there are implications in this statement which seem quite unfounded. First off, ‘where does 98.7% of the female population rejects an average looking male as a potential partner!’ come from (but this is minor). My main issue is with the apparent claim that averagely attractive women experience interest from men to the degree which highly attractive men experience interest from women.
There are plausible reasons why a highly attractive male might receive less likes than a medium attractiveness female aside from variance in intersexual attraction. These include men being more numerous on the platform, or men being more active on the platform.
@Ryan Coons,
“I definitely agree with the majority of your findings, but there are implications in this statement which seem quite unfounded. First off, ‘where does 98.7% of the female population rejects an average looking male as a potential partner!’ come from (but this is minor).”
Let N be the number of female users to whom the highly attractive male profile has been displayed.
Where n is the number of female users to whom the medium attractive male has been displayed
Being x (= 231), the number of receptive users (swiping right) for highly attractive man.
we know that N≥ x (= 231).
And y= 3, the number of receptive users (swiping right) for average attractive man.
Let’s assume that N≈n, Being the unknown sampling algorithm, but we can infer, for the sake of rationality, that profiles are incorporated in a random order/sequence. The average order of presentation rate should be approximately equivalent, since both male profiles share the same geographic location and are co-existent in the time of existence and frequency of use online.
x (min) ≤N≥Nmax
As N≈n, for convenience we use n≈N = x (min) = 231
Let’s figure out acceptance/rejection rate for the lower limit: Where Y (= 3) would be 1.3% of acceptance { for n≈N =x (min)} and 98,7% swiping left/rejection.
We could deduce that surely N will take a value higher than xmin.
“My main issue is with the apparent claim that averagely attractive women experience interest from men to the degree which highly attractive men experience interest from women.
There are plausible reasons why a highly attractive male might receive less likes than a medium attractiveness female aside from variance in intersexual attraction. These include men being more numerous on the platform, or men being more active on the platform.”
Being S the size of male sample users to whom the medium attractive woman has been displayed.”
Being t the number of receptive users (swiping right), t = t1 + t2.
And t1 number of matches (= 310), t2 unknown (≥99), t would be t≥409.
Therefore t> x. So the dating pool for the average looking woman is larger than for the highly attractive man into an online dating app
I agree with you that likey S> N, since typically OSR (operational sex ratio) is male biased in mating leks. Which would lead us to a circular reasoning, since sexual dimorphism over differential pairing gradient creates such imbalances/distortions in the mating market, also generating male biased OSRs within Leks.
Why female G has 4 times more message that the hottie ?
@ghju,
I guess many of G’s matches felt that it was more feasible to get an answer from her. While the hottie matches believed that she would be more inaccessible, even was a bot, catfish, etc. (keep in mind that the vast majority of men do not usually get reciprocity from beautiful women, and when such a match is achieved online, it usually generates well-founded suspicion).
This is just sad 😦