Calculating the Odds of Finding a Spouse in the Muslim Community

Special thanks to Shehbaz Aslam, Brendan Avent and Kashev Dalmia for proof-reading this article for grammatical, mathematical and logical errors.

XKCD: Dating Pools

So I’ve recently noticed that a majority of my Muslim friends are single and have been single for a long time. I’ve heard a variety of different explanations from different people as to the cause, some saying it’s because they’re too picky, some saying because they’re too old, others speculating it has to do with the cultural and religious background we come from.

After talking to a few of my friends about what they were looking for in a partner, I felt like they had unrealistic expectations. Being an engineer, I wanted to see if I could prove this. I’m not sure if I can quantify their potential for love, but I can quantify the number of people that meet their standards.

Recently I was listening to an older podcast from This American Life about Valentine’s day. It interviewed some Harvard graduate students about how they calculated the likelihood of finding girlfriends using a variation of The Drake Equation.

Fig 1. The Drake Equation

The Drake Equation is a way to estimate how many planets in our universe have intelligent life. It works by taking the total amount of planets and multiplying it by factors that limit how many could have intelligent life.

We can apply the same principle to finding a partner. For example, let’s say you’re at a singles event with 100 people. Half of the singles are men, and you’re only interested in women. You would multiply the total number of singles by ½. Now you’re left with only 50 people to choose from.

Before we move forward

I wanted to mention that the numbers presented here are an approximation. There are a number of contributing factors that could skew the results that I’m going to document here:

  • Some data used is older census data from 2010 and 2006. The population of each ethnicity could have increased or decreased during the last decade.
  • Some census data included multi-racial ethnicities and some did not.
  • Demographic data for California was used for a large portion of these calculations. There isn’t a guarantee that those numbers would fit perfectly with Muslim populations. Each demographic might have a different education, gender trend, etc…. I used a corresponding demographic trend if I found it.
  • The Drake equation is useful to us for these calculations because it assumes each factor is perfectly independent from the others. But not all factors are truly independent. For example, we multiply the percentage of women who are single in California by the percentage of women who are between the age of 18–24. The number that calculation generates will likely give a number less than the actual number of single 20 to 24-year-olds. This is because we used the average marital ratio for California and not for women in the age range of 20–24. Women aged 20–24 are more likely to be single.

Case 1: The woman my parents want me to marry

So every person with immigrant parents knows that your parents would love to pick out your spouse. They’ve got a very specific set of requirements, being from the woman’s ethnicity, religion, education, looks and more. I thought I’d calculate the odds of finding the girl my parents approve of.

My parents first requirement is the women be Iranian (Persian). So let’s denote to be the number of possible Persian women. First, we take the number of Persians in the Greater Los Angeles area, which is estimated to be around 300,000¹. Then we multiply it by the female to male ratio in California which is around 50%².

Pʷ = 300,000 * 0.5 ~= 150,000

Now we have the number of female Persians in Los Angeles. Let’s solve for the number of them that are single. The marital ratio in California is also 50%².

Pʷ = Pʷ * 0.5 ~= 75,000

The next requirement would be for them to be Muslim. A survey done by the Public Affairs Alliance of Iranian Americans found that around 30% of Iranian American’s identify as Muslim³.

Pʷ = Pʷ * 0.3 ~= 22,500

Alright, now we have to consider age. My parents’ want someone younger than me, so we’ll say anyone within the age range of 20–24 which is around 8%⁴.

Pʷ = Pʷ * 0.08 ~= 1,800

Bachelor’s education or higher is a requirement as well. Thankfully, Iranian Americans are some of the most educated in the country with around 51% having a bachelor’s or higher⁵.

Pʷ = Pʷ * 0.51 ~= 918

Attraction is important. This number is hard to quantify, but let’s say for argument’s sake I’m not picky and I’m attracted to 1 in 3 women.

Pʷ = Pʷ * 0.33 ~= 306

I’m terrified looking at that number. To put it in perspective, there are around 19 million people in the Greater Los Angeles area. This means only 0.0016% of the population of greater Los Angeles would fit my parents’ criteria. And we haven’t discussed a lot of other factors, like whether we get along or whether I fit her requirements.

Case 2: Male Pakistani-American, 27 years old

From here on out, I’m going to state demographic factors and the final calculation.

This example is based on an interview with a friend. I asked him about his preferences and this is what he stated:

  • Must be Pakistani Los Angeles area preferred (estimated to be 150,000⁸)
  • Must be female. ~= 50%²
  • Must be single. ~= 50%²
  • Must be Muslim, ~= 90%.
  • Age range 20–30 ~= 15%.
  • At least a Bachelors ~= 60%¹⁰
  • Attraction 1 in 50 women or 0.02%

Let’s use Pᵏʷ to denote the number of Pakistani women that would match my friend’s profile.

Pᵏʷ = total people * gender * single * muslim * age * education * attract

Pᵏʷ = 150,000 * 0.5 * 0.5 * 0.9 * 0.15 * 0.6 * 0.02 = 60

He’s got such a low number of potential matches I’m willing to speculate it’s nearly impossible for him to find someone with his desired criteria. Let’s expand his search to include the United States and Canada. The United States has an estimated Pakistani population of 453,000⁷ and Canada has 155,000¹¹.

Pᵏʷ = (453,000 + 155,000) * 0.5 * 0.5 * 0.9 * 0.15 * 0.6 * 0.02 ~= 246

In the 350 million people in the United States and Canada, he only has 246 potential matches. The largest factor weighing him down after the small sample size is his pickiness. Let’s see what happens if we make him less picky and change his attraction level to 1 in 10 women instead of 1 in 50 women.

Pᵏʷ = (453,000 + 155,000) * 0.5 * 0.5 * 0.9 * 0.15 * 0.6 * 0.1 ~= 1230

He increases his odds by a factor of 5. But, that number is still abysmally small. I’d like to reiterate that we haven’t even considered chemistry or how many of the 1230 women’s requirement he would meet.

Case 3: Female Indian-American, 29 years old

Here is the profile of another friend I interviewed who wants to get married. Here are her requirements:

  • Preference for those of South Asian decent. i.e. Indian(120,000¹³) and Pakistani (150,000⁸).
  • Must be male. ~= 50%²
  • Must be single. ~= 50%²
  • Must be Muslim ~= Indian 10%¹², Pakistani 96.5%
  • Age range 24–35 ~=16%⁴
  • At least a Bachelors ~= Indian’s 70%¹², Pakistani 60%¹⁰
  • Attraction 1 in 10 or 0.1%
  • Height preference of 5'8" or taller ~= Indian and Pakistani men 5%ᴬᴮ.

Let’s use Pᵏᵐ to denote the number of Pakistani men that would match my friends profile and Iᵐ to denote the number of Indian men.


Pᵏᵐ = total people * gender * single * muslim * age * education * attract * height

Pᵏᵐ = 150,000 * 0.5 * 0.5* 0.965 * 0.16 * 0.6 * 0.1 * 0.05 ~= 17

Iᵐ = total people * gender * single * muslim * age * education * attract * height

Iᵐ = 120,000 * 0.5 * 0.5* 0.1 * 0.16 * 0.7 * 0.1 * 0.05 ~= 1.68


Almost non existent. Let’s remove her height preference and see what happens:

Pᵏᵐ = 150,000 * 0.5 * 0.5* 0.965 * 0.16 * 0.6 * 0.1 ~= 340

Iᵐ = 120,000 * 0.5 * 0.5* 0.1 * 0.16 * 0.7 * 0.1 ~= 33

Now let’s remove her ethnicity preference. There are about 500,000 muslims in the greater Los Angeles area¹⁷. Let’s say Mᵐ will denote the number of Muslim men.

Mᵐ = 500,000 * 0.5 * 0.5* 0.16 * 0.6 * 0.1 ~= 1200

Let’s expand the search to include the United States and Canada. The total estimated population of Muslims in the United States is 3.3 million and the total population of Canada is 1,053,945.

Mᵐ = (3,300,000 + 1,053,945) * 0.5 * 0.5* 0.16 * 0.6 * 0.1 ~= 10,449

You can see that her odds got significantly better the moment we got rid of her height preference and expanded the search to include all Muslims instead of just Pakistanis and Indians.

Case 4: American male, 27 years old

This last profile is fictional. I made it up to reflect the standards of an average educated American in Los Angeles. Let’s give him the following preferences:

  • No ethnicity preference ~= 18.7 million²⁰
  • Must be female. ~= 50%²
  • Must be single. ~= 50%²
  • No religious preference ~= 1
  • Age range 20–30 ~=15%.
  • At least a Bachelors ~= 31%²¹
  • Attraction 1 in 10 or 0.1%

Let’s say W will denote the number of women this person would match with.

W = total people * gender * single * religion * age * education * attract

W = 18,700,000* 0.5 * 0.5 * 1 * 0.15 * 0.31 * 0.1 = 21,738

So the average American male had 2 times the number of potential partners in Los Angeles compared to the total number of partners my Indian friend (Case 3) had within the entire United States and Canada.

Analysis

The Pakistani male faces incredible odds

Let’s compare the results of Pakistani-American male (Case 2) vs American male (Case 4)

The main differences between the American male vs the Pakistani male are the following:

  • The American male has no ethnicity restrictions which increases his starting candidate pool to 18.7 million versus the Pakistani’ males meager 150,000.
  • The Pakistani male is 5 times more picky by being attracted to 1 in 50 women versus the American males 1 in 10 women.
  • The American male has no religious preference.
  • The percentage of candidates with a bachelor’s degree within the general American population is 29% less than within the Pakistani community.

Here are some interesting comparisons:

  • The American male has 362 times (21,738 vs 60) more potential partners in the Greater Los Angeles area than the Pakistani male.
  • The American male has 88 times (21,738 matches vs 246) more potential partners in Los Angeles than the Pakistani male had within all the United States + Canada.
  • 1 in 860 women in the Greater Los Angeles area might be a potential partner for the American male, versus 1 in 311,666 women for the Pakistani male.
  • If we make the Pakistani male as picky as the American male, 1 in 62,333 women in the Greater Los Angeles area become potential partners.

The number of potential partners for the Pakistani male is astonishingly lower than the American male. To put it into perspective, for each woman the Pakistani male finds in Los Angeles, the American male could have found 361 more (assuming he had enough time).

Eliminating ethnic preferences helps significantly

Eliminating ethnic preferences for the Indian-American woman (Case 3), her potential partner pool increased 3 fold within the Greater Los Angeles area (373 → 1200).

Searching globally makes an impact

  • Looking within the United States + Canada for matches instead of local matches expanded the Pakistani male’s (Case 2) candidate pool 4 fold (60 → 243).
  • The Indian-American female (case 3) had 10 times a larger candidate pool (1200 → 10,449) when searching within the United States + Canada for a Muslim partner (no ethnic preferences).

Conclusion

You can see that each case (besides the fabricated one at the end), had an abysmally low number of potential partners. What’s shocking is that these numbers d0n’t consider more restricting requirements like having the partner be a doctor/lawyer/engineer or making above a certain income.

Including these requirements would cause the total number of potential partners to drop to zero. This is why a significant amount of my Muslim friends are single. Their requirements are unrealistic. The person they’re looking for doesn’t exist.

I’m not saying you should give up, but you should take a look at what’s really important to you. Because at the end of the day, a lot of your requirements don’t matter as much as you think.

Footnotes

A. I had to calculate the percentage of men of Indian origin who were 5'8" or taller. To do that found that the average height of an Indian man was 5'5" (168 cm) with a standard deviation of 2.71 inches (6.9 cm)¹⁵. I then subbed those numbers into the equation for a Normal distribution and integrated that with the bounds 5'8" (176.784 cm) to infinity to find the percentage of those surveyed who were 5'8" or taller¹⁶.

B. Unfortunately, I could not find this statistic for Indian men in the United States. There’s a potential for Indian men in the states to be taller because of better nutrition. I also was forced to use the statistic for Indian men for Pakistani men as well because I couldn’t find specific height information about Pakistan.

References

  1. http://www.bbc.com/news/magazine-19751370
  2. https://censusreporter.org/profiles/04000US06-california/
  3. http://www.paaia.org/CMS/Data/Sites/1/PDFs/2012surveyofiranianamericans-militaryaction-electronic.pdf
  4. https://factfinder.census.gov/bkmk/table/1.0/en/ACS/15_5YR/S0101/0500000US06037
  5. http://www.payvand.com/news/04/jan/1191.html
  6. https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=CF
  7. www.migrationpolicy.org/sites/default/files/publications/RAD-Pakistan.pdf
  8. http://www.pakconsulatela.org/pakistani-community/
  9. https://www.cia.gov/library/publications/the-world-factbook/fields/2122.html?countryName=Pakistan&countryCode=pk&regionCode=sas&#pk
  10. http://www.ide.go.jp/English/Publish/Download/Dp/196.html
  11. http://www12.statcan.ca/census-recensement/2006/dp-pd/tbt/Rp-eng.cfm?LANG=E&APATH=3&DETAIL=0&DIM=0&FL=A&FREE=0&GC=0&GID=837928&GK=0&GRP=1&PID=92333&PRID=0&PTYPE=88971,97154&S=0&SHOWALL=0&SUB=0&Temporal=2006&THEME=80&VID=0&VNAMEE=&VNAMEF=
  12. http://www.pewresearch.org/fact-tank/2014/09/30/5-facts-about-indian-americans/
  13. https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_15_1YR_B02018&prodType=table
  14. http://www.pewsocialtrends.org/2012/06/19/the-rise-of-asian-americans/
  15. http://journals.sagepub.com/doi/pdf/10.1177/156482651103200103
  16. https://www.wolframalpha.com/input/?i=integrate+1%2Fsqrt(2+*+pi+*+(6.9)%5E2)*e%5E-((x-165.2)%5E2%2F(2*+(6.9%5E2)))+from+x%3D176.784+to+infinity
  17. http://www.scpr.org/news/2015/12/11/56162/socal-s-muslim-community-geographically-ethnically/
  18. http://www.pewresearch.org/fact-tank/2016/01/06/a-new-estimate-of-the-u-s-muslim-population/
  19. http://www12.statcan.gc.ca/nhs-enm/2011/dp-pd/prof/details/page.cfm?Lang=E&Geo1=PR&Code1=01&Data=Count&SearchText=Canada&SearchType=Begins&SearchPR=01&A1=Religion&B1=All&Custom=&TABID=1
  20. https://factfinder.census.gov/bkmk/table/1.0/en/PEP/2015/GCTPEPANNR.US41PR
  21. https://www.census.gov/quickfacts/fact/table/CA/EDU685215#viewtop