Linguistic Diversity and Trust

Andrew Leigh has a discussion paper out on linguistic/ethic diversity and trust. I’m commenting on the 7-page general summary and paper for lay audiences published in Dialogue – I haven’t seen the other paper he mentions.

I’m worried about a number of aspects of the paper and apparent suppositions of causality, in particular the idea that diversity causes mistrust. Some

  • Linguistic diversity and ethnic diversity are not the same thing!

Both ethnic diversity and linguistic diversity are used in the paper. Thus Indigenous populations are presumably treated as a single group for this purpose, despite the presence of 100+ different Indigenous languages, Kriol and English. This would be an issue in regional centres (e.g. Dubbo), where the population is linguistically fairly homogenous (depending on how you treat dialectal diversity) but racially diverse. Given the small sample sizes, this could make a difference (although I don’t have the data to test this – maybe Andrew could depending on the richness of the original data set).

In particular, neighbourhoods where many languages are spoken tend to have lower levels of trust, suggesting that the main issue may whether people can communicate effectively with those living nearby.

Not necessarily, given that linguistic diversity patterns themselves are greatly diverse. For example, there is a very well documented pattern of language shift amongst immigrant communities, where the first generation tend to be non-English dominant, the second generation are bilingual, and the third are English-dominant or English-only. Linguistic diversity can be high without communication barriers, and frequently is so.

Furthermore, highly linguistically diverse communities have multiple origins:

  1. community-wide multilingualism (Milingimbi, historical Martha’s vineyard)
  2. large influx of recent immigrants from many places
  3. long-term immigration where bilingualism has remained stable (rare in Australia, as everywhere)
  4. regular immigration from different areas over different times (East London).

Such different patterns typically have rather different demographics, and different income distribution. The correlation with a raw measure of linguistic diversity is incidental.

Linguistic diversity is strongly correlated with age of establishment of community. As mentioned above, there’s a well-established pattern of language shift in immigrant communities. It’s typically a three-generation process, but in some communities it’s a two-generation process, and in others (e.g. Dutch immigrants to Australia and the US the shift is completed within a generation). Thus with the exception of Aboriginal communities, a high proportion of linguistic diversity is also strongly correlated with recent immigration. That’s surely going to have an effect on “trust” levels too.

There is a further problem with the question used to gauge “trust”, which was “generally speaking, you can’t be too careful when dealing with most Australians”. It’s hard to parse for people without fluency in English, so that could affect results for areas of high linguistic diversity.

Secondly, there are well-studied problems with this type of question when asked of minority versus majority groups. In a nutshell, majority group respondents interpret the question as applying to the majority, whereas minority respondents either interpret it as applying only to the majority or to their own group. That’s probably less of a problem here because of the “most Australians” part, but it is still another potential confounding factor. That is, the “object of trust” may not be equivalent for different groups.

Finally, I’d want to see something more about the notion of “trust”.


4 responses to “Linguistic Diversity and Trust

  1. I didn’t really digest all of what you were arguing or what you were arguing against, but it seems relevant to say that I feel like there’s a general habit of monolingual people not trusting people who speak other languages. Bilingual or multilingual people would generally be less close minded wouldn’t they? Or am I just making assumptions?

  2. David Marjanović

    The 2nd paragraph is cut off.

  3. Claire, thanks for your thoughtful feedback.

    As to language classifications, the paper is really identifying effects based on the 23% of Australians who are born overseas, rather than the 2% who are Indigenous. Also, because I’m using the Census, the language definitions aren’t as precise as one might like. Here’s the relevant paragraph from the paper:

    “Two measures of ethno-linguistic heterogeneity are used. First, I calculate a measure of ethnic fractionalization based on country of birth. The census classifies respondents as born in Australia, or in 22 broad regions of the world (eg. Northern America, Southern Asia). Second, I form a measure of linguistic fractionalization based on whether the respondent speaks a language other than English at home. Those who speak only English at home are classified as one subgroup, 24 other languages are separately identified, and a final category identifies ‘other’ languages. ”

    On the notion of trust, here’s the best I could do to validate the measure:

    “What is the ‘Generally speaking, you can’t be too careful in dealing with…’ question really measuring? Glaeser et al (2000) survey a group of Harvard undergraduates, and then ask them to play a trust game. They find that the question used in the General Social Survey (‘Generally speaking, would you say that most people can be trusted, or that you can’t be too careful in dealing with people?’) actually measures how trustworthy the respondent is, rather than how trusting of others they are. Similarly, Karlan (2005) finds that Peruvian respondents who agree that most people can be trusted are less likely to default on their microcredit loans. It is likewise possible that the question used in this paper is actually measuring how trustworthy rather than how trusting the respondent is. Readers who wish to interpret the results in this paper as such should feel free to do so, though it makes little difference when interpreting the effect of neighbourhood-level factors.”

    Your point about how the trust question is interpreted by people who don’t speak English as a first language is an important one.

    In case you’re interested, the academic version of the trust paper is at

  4. Andrew, I can’t quote sources for this but the census data is better for major community languages than for Indigenous languages. I suspect that the (overseas-born) community language data will underreport diversity, whereas the inidgenous census data tends to overreport speakers because there’s no differentiation between “speakers” and “owners”.

    This paper’s a nice complement to your recent immigration and economic success paper, by the way.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s