The Tendency to Form Social Circles with People of Similar Values and Culture

[JohnL]

Maybe something from Ravenhill? sounds similar to the quote about to the amount of real Christians in the church being around 2 percent or so...

Yep it does sound like Ravenhill. Or one of his associates who traveled in the same circles.

[Neto]

There is a vast range within the idea of "cultural differences", and the same in terms of linguistic differences. Regarding the latter, are dialectal differences included, for case where another's speech patterns (pronunciation, cadence, and intonation) may make it difficult to understand, without sometimes repeatedly asking for the person to repeat what they have said. This creates an uncomfortable feeling, the fear of coming across as offensive.

If a person grew up as a minority person immersed in a dominant culture or language group, would you all say that makes more, or less, inclined to engage people from outside their own group?

I suspect more, because that is my experience. I grew up in a very small Mennonite community, one that was already somewhat cross-cultural. Then I attended a public school where the vast majority were American Indians. So I was a minority person within the 'white' classification, which was also a minority in the school and in the larger society, which was dominantly American Indian. I have also always been in social situations where a language I do not know is also regularly spoken in my presence. And that situation doesn't bother me. I attended Bible college in the central northern state of Minnesota, and at first their intonation patterns made me think they were disagreeing with what I had said, because the intonation of their common conversational interjection carried the implication of disbelief in my own dialect of English. (They would throw in a 'Probably' in the pauses as I was speaking, with the following intonation pattern: Normally this would be 'written' by showing lines superimposed over a word or phrase, but by using capital letters as high tone, and lower case as the low tones, I will write it as PROB-ab-LI-e.)

Another factor is 'ethnic smell', which I think it part of the reason some people do not seem to ever consider a person of a different ethnicity as a marriage partner. (I do not think this is all due to 'racial attitudes' or prejudice, although certainly cultural differences may account for a good part of this. It is obscured by the tendency for these cultural differences to be tied to specific ethnic identities.)

[Ernie]

It sounds like what is being said in that stat is essentially "A huge majority of people do not have the disposition it takes to be a truly effective foreign missionary." But the stat was being used in another thread to say "The vast majority of people don't like witnessing the presence of foreign-born people in the neighborhood they live in."

Many Americans do not like having foreigners in their neighborhood, especially if they don't speak English fluently, or the ladies cover their heads and dress modestly, or if the folks have a different skin color. Add in "undocumented" and it can increase the dislike.

I don't know what the percentages are for Americans who embrace such folks as fellow human beings, those who tolerate them, and those who despise them. I do know that it will vary widely across the US based on the local community dynamics.

One Plain Anabaptist brother told me that there that were people in his church who were mentally predisposed to think negatively about immigrants. But then an immigrant started attending church. The immigrant's attendance forced some in the church (who were devout believers in the Gospel) to adjust their thinking. Here was a human being sitting on the same bench or on an adjoining bench who needed friends. And so these people who used to have negative thoughts about immigrants, suddenly needed to decide whether they were going to let their theology change their minds/hearts/emotions, or let their minds/hearts/emotions change their theology. Thankfully some allowed for the former and it was a good experience for the church. But unfortunately, the latter often happens as well, particularly for those for whom Christianity is simply a cultural identity and not a living experience.

[Ernie]

It is fine to discuss immigration and the effect it has on people in this thread, but please no political discussion in this thread. Thank you!

[joshuabgood]

There is a very real sense that a community is the people who are in it. Therefore, when the people change, the community changes. From my perspective this isn't an ethical question. Like many things in life, there are some real losses and real gains possible depending on various contexts.

[Ernie]

Here is what ChatGPT has to say and some studies it references when asked the question,
"True or False? Only about 3% of the world’s population can truly form deep multicultural social circles?"

https://chatgpt.com/s/t_68f15366aa3c819 ... 54707513ce

[Szdfan]

Here is what ChatGPT has to say and some studies it references when asked the question,
"True or False? Only about 3% of the world’s population can truly form deep multicultural social circles?"

https://chatgpt.com/s/t_68f15366aa3c819 ... 54707513ce

So I asked ChatGTP the same question --

True or False? Only about 3% of the world’s population can truly form deep multicultural social circles?

Here's what I got:

False — there’s no credible evidence that only about 3% of the world’s population can form “deep multicultural social circles.”

Here’s why that claim is extremely unlikely:

I found no academic or reputable source making that “3%” assertion (or anything close) about deep multicultural social-connection capacity.

Social science studies show that many people’s close social circles tend to be relatively homogeneous (e.g. with respect to race, ethnicity, socioeconomic status) — a phenomenon known as homophily — but that doesn’t imply only a tiny minority can form cross-cultural bonds.

Some research suggests that exposure to cultural diversity in one’s environment correlates with greater ability (or inclination) to connect across cultural groups (“social brokers”) — but that is a statistical tendency, not a strict threshold that disqualifies 97% of people.

Definitions matter: What counts as “deep multicultural social circles”? If “deep” means intimate, trusted, cross-cultural friendships (versus casual acquaintances), then many fewer people might attain that, but even then, a 3% ceiling is arbitrary without evidence.

If you like, I can dig into studies of cross-cultural friendships, bridging social capital, or “social brokerage” and estimate what proportion of people do this. Would you like me to try that?

[Szdfan]

So out of curiosity, I posted the answer that Ernie had posted and asked the following question:

Someone else also asked the same question to ChatGTP and got the following answer. Why do you think there's this discrepancy?

This is what my ChatGTP said:

Excellent question — and it’s a good example of how AI language models can sound authoritative even when the underlying data is more interpretive than empirical. Let’s break down why that other answer and mine diverged.

1. Different Standards of Evidence

My response: Treated the “3%” claim as a factual statement — something that should be supported by verifiable quantitative data (e.g., global surveys, demographic studies). Since no such study exists, I said “False.”

The other response: Treated “3%” as a plausible inference from indirect evidence (migration rates, homophily studies, trust data) and gave a reasoned interpretation rather than a literal fact-check.

In short:

My approach = Is there proof of this number?
Theirs = Could this number be a reasonable approximation based on patterns?

2. How AI “fills gaps” with synthesis

When no direct statistic exists, a model sometimes synthesizes related findings to build a plausible argument — e.g., using:

Global migration data (~3–4% of people live outside their birth country).

Friendship homophily data (80–90% same-ethnic close ties).
Then it infers: “Deep multicultural friendships are probably as rare as high cross-cultural migration rates (~3%).”

That’s a reasonable but speculative inference, not a published result.

3. What’s actually known

Empirical research does support that:

Most people’s close friends share race, language, and class.

“Deep” cross-cultural friendships exist, but they’re a minority pattern.
However, there is no global dataset that measures “deep multicultural social circles” or quantifies them at 3%, 5%, or 10%.
So, the “3%” number is an interpretive synthesis, not a measured fact.

4. Why both answers can coexist

If you’re asking a scientific true/false: it’s False — the 3% figure isn’t empirically demonstrated.

If you’re asking a sociological interpretation: it’s Reasonable but speculative — the world’s social networks are highly homogeneous, and only a small minority sustain emotionally deep multicultural ties.

5. In summary
Question Type Answer Reason
Literal / factual False No empirical study measures or supports the 3% statistic.
Interpretive / inferential Plausibly true (2–5%) Homophily, trust, and migration data imply it’s rare.

If you’d like, I can trace the main sociological studies that the “2–5%” synthesis probably drew from — so you can see exactly where each component (trust, migration, friendship homophily) comes from. Would you like me to do that next?

It sounds like a difference between "inference" based on data and empirical data. With Ernie, ChatGTP inferred (i.e. educated guessed) its answer based on a range of data. In my case, ChatGRP looked at specific studies on this question.

I think it's also important to note that Ernie's prompt --

True or False? Only about 3% of the world’s population can truly form deep multicultural social circles?

Is a fundamentally different question that John's interpetive comment --

Most people (about 97%) are not very interested in assimilating to other cultures or learning more about foreign cultures. 3% of people (such as myself) are. It is important we respect the 97% of people who would prefer to only associated with their own kind, often delineated by language, ethnic group, or religious leaning.

If we want to, we can break down how these two different statements are different and similar.

[JohnL]

There is a very real sense that a community is the people who are in it. Therefore, when the people change, the community changes. From my perspective this isn't an ethical question. Like many things in life, there are some real losses and real gains possible depending on various contexts.

I can see that here.
Here in the mountains. The “hollers” are where an entire extended family lives. They’ll marry someone from somewhere else and bring them back to the holler to start a new branch of the family. The holler was family and safety. It often resembled the clans of the Cherokee people. If a woman married a husband from outside the holler sometimes the husband just took on the holler family name. Up until the 1950s there were a lot of family hollers in these mountains and the broader mountain culture was shaped by them.
Then the changes came from a Federal project called the ‘war on poverty’. Some of holler kin sold their property because they were promised rich lives like the pictures and photos in the magazines that made their way into the local stores. The govt pushed for the families to sell and move to the foothills with promises of better jobs. You can guess what happened next. The land ended up in the hands of unscrupulous and greedy developers. The govt raised the taxes on property to force the remaining families to sell their holdings.
This bit of mountain history is part of a larger story of abuse of the mountain folks. The stories of this abuse have been handed down to the generations sitting at the supper table and explains the intense distrust of anything tied to the government and developers.

[Bootstrap]

These questions get at very different things. Ernie's 97% counts only people who have "deep multicultural social circles". This doesn't actually occur in surveys. When I asked GPT how it would interpret this, it said it would use this criterion:

A “deep multicultural social circle” would be implied by survey answers showing more than one close friend of another ethnicity/religion/culture (as opposed to “all/most” being the same).

I asked GPT to give me results for questions that were actually asked in surveys, and came up with answers like this.

1. Close Friends of a Different Race/Ethnicity

Pew Research Center (2023 American Trends Panel):
- 35 % of U.S. adults with at least one close friend say all of their close friends are the same race/ethnicity.
- 27 % say most are the same race/ethnicity.
- 22 % say some are the same.
- 8 % say only a few are the same.
- 5 % say none are the same.

Summary:
- 62 % have close friend groups that are "all or mostly" the same race/ethnicity.
- 38 % have at least one close friend of a different race/ethnicity.

Psychology Today (summary of similar results):
- 66 % report "all or most" of their close friends are the same race/ethnicity.
- That implies about 34 % have at least some close friends of a different race/ethnicity.

Note: The small differences between Pew and Psychology Today reflect rounding and slightly different survey wording.