Generational Differences

What it is

The whole question turns on separating three explanations that produce identical-looking data.

A cohort effect says the birth group itself carries the difference, which is what someone means when they say a particular generation does not want to work. An age effect says people change as they get older, in every cohort, which is what the life course hypothesis proposes when it holds that work becomes more important until around forty and less so afterwards. A period effect says the moment of measurement explains the result for everybody alive in it, which is what the historical hypothesis proposes: a recession, a pandemic, a housing market.

A survey taken at one moment cannot tell these apart, and almost every generational claim in circulation is a single cross-sectional survey. The usable form of the test is a single sentence. If a claim about a generation could be restated as a claim about being twenty-four, or about being alive in a particular year, it is not yet a claim about a generation.

In effect

The evidence base runs in one direction. Costanza, Badger, Fraser, Severt and Gade pooled twenty studies and eighteen generational pairwise comparisons across nearly twenty thousand participants and found essentially nothing. Zabel and colleagues found the same for work ethic. Cort Rudolph, Rauvola, Costanza and Zacher later assembled the debunking directly, myth by myth.

The genuinely useful half of the concept is the prejudice rather than the difference. Generationalism, glossed as a modern and socially acceptable ageism, is real in a way the underlying differences are not: sanctioned ambivalence and openly voiced assumptions about people of particular ages, in professional settings where equivalent statements about other groups would not be made. The differences are not there. The prejudice organised around believing in them is.

There is a cautionary example in how easily the tool gets put down. A textbook treatment that deploys this meta-analysis to show generational patterns dissolving into age and period effects asserts, two pages earlier and unsourced, that a particular wave of resignations was almost exclusively among young people. That is a cohort claim of exactly the kind about to be dismantled, and one that a period explanation fits far better. A correction stated in one place and not applied in the next is not yet a correction.

What it does not say

It does not say that young and old workers are identical. They differ by age, and age effects are real. What fails is the attribution of those differences to the birth cohort.

It does not say nothing changes over time. Period effects are real too, and a labour market or a pandemic changes everybody in it.

It does not say generations are a meaningless idea in sociology. Mannheim's concept is about a cohort marked by shared formative events, which is a narrower and more defensible claim than the marketing version.

And the popular myth lists should be cited with care. A widely reproduced set of ten debunked generational myths is frequently credited to the wrong paper, and belongs to the later debunking article rather than to the work-ethic meta-analysis.

---