What population has the least divorce?

1. Definitions matter: crude rates versus “at-risk” measures

One reason there is no single answer is methodological: crude divorce rates count divorces per 1,000 total population and can be depressed by many unmarried people or a young demographic, whereas refined rates or divorce-per-married-women measure the population actually at risk of divorce—each has weaknesses and yields different “least divorced” populations ^[7].

2. The global bottom: Sri Lanka and other low-crude-rate countries

International compilations identify Sri Lanka as having the lowest crude divorce rate—about 0.15 divorces per 1,000 people in 2020—and list Vietnam and Guatemala also among the lowest; multiple commercial and data aggregators reiterate Sri Lanka’s position as the lowest in those datasets ^{[1] [2] [8]}. These country-level lows are plausibly driven by cultural, legal, and reporting differences, and they reflect crude counts rather than the share of marriages that end in divorce ^{[8] [7]}.

3. Age: older adults are the least likely to divorce in annual rates, but ‘gray divorce’ is rising

When broken down by age, the lowest annual first-divorce rates are among the oldest cohorts—reports show the lowest first-divorce rates among people 65+ or even 75+ in recent U.S. data (for example 4.2 per 1,000 at 65+ in one profile and 5.0 per 1,000 for men 75+ in another) ^{[3] [4]}. That said, scholars note a sharp rise in “gray divorce” among those over 50 in recent decades, so older adults both register the lowest cross-sectional rates and the fastest increases over time ^{[9] [4]}.

4. Race/ethnicity and social segments: Asians and religious/affluent groups report lower divorce exposure

Analyses of U.S. demographic patterns show Asian adults with the lowest first-divorce rates relative to other racial groups in the data reviewed, and segmented studies list Asians among the groups least likely to have been divorced, alongside higher-income adults and certain religious identifiers like Catholics or evangelicals in some compilations ^{[3] [4] [10]}. These patterns likely reflect complex mixes of selection, cultural norms, socioeconomic status, and marriage timing; the sources differ in whether they report incidence (divorces per 1,000 married) or prevalence (share ever divorced) ^{[3] [10] [7]}.

5. Geography within the U.S.: Puerto Rico, Vermont and other low-rate places

At the state and territory level, Puerto Rico appears among the lowest in both marriage and divorce crude rates in multiple Census presentations, with generally low recorded marriage and divorce rates attributed in part to demographic shifts like outmigration of young adults ^{[5] [6]}. Other U.S. analyses using refined measures find states such as Vermont or New Hampshire at the lower end of divorces per 1,000 married women in particular years, underscoring that “least divorced” depends on the chosen denominator and year ^[11].

6. Caveats, conflicting signals and the implications for the question

All sources warn that differences in measurement, cultural/legal factors, and demographic composition drive the variation: a country with very low crude divorce may simply have fewer married adults or social barriers to divorce, and within the U.S., low annual divorce rates for the elderly coexist with a growing trend of later-life divorces; similarly, racial group differences may reflect socioeconomic or selection effects rather than immutable cultural traits ^{[7] [9] [8]}. Therefore, the direct, defensible answers are: by country (crude rates) Sri Lanka ranks lowest in available international compilations; by age group, people 65+ or 75+ register the lowest annual first-divorce rates in U.S. data; by race/ethnicity, Asian Americans show the lowest first-divorce rates in the cited U.S. profiles; and by U.S. geography, Puerto Rico and some New England states rank among the lowest depending on the metric used ^{[1] [2] [3] [4] [5] [11]}.