The Easiest Astrological Claim to Test
Among the many things astrology claims, “your sun sign — determined entirely by your birth month and day — correlates with your personality” is structurally the simplest to test. It requires only two pieces of information: birth date (to determine sun sign, which for most people is unambiguous — only people born on the one or two “cusp” days each year near a sign boundary require more precision) and a personality measure. No birth time, no birth location, no chart calculation, no practitioner interpretation. Twelve categories, applied to everyone, with clearly defined boundaries.
This simplicity is exactly why sun sign has become the most-tested astrological claim by a wide margin — and also why it’s the claim astrologers most consistently describe as the least important factor in a real reading. This creates an odd dynamic: the part of astrology that’s easiest to test empirically is also the part most astrologers would tell you matters least, which means the research that’s easiest to do addresses a question that’s least central to the practice being studied.
The Research Timeline
Sun sign research has a long history, predating the Big Five studies discussed in the companion piece. Early work in the 1970s and 1980s, including studies by Hans Eysenck — one of the twentieth century’s most influential personality psychologists, and someone whose early work was sometimes cited by astrologers as supportive — initially reported some correlations between sun sign categories (particularly the traditional “introvert” and “extravert” sign groupings) and personality measures.
These early positive findings did not hold up. Follow-up research, including work specifically designed to test whether the apparent sun-sign effects in Eysenck’s early data were artifacts of self-selection — people who already knew astrology and had absorbed the trait descriptions for their sign — found that the effect was concentrated entirely among people familiar with astrology, and vanished in samples of people who didn’t know their own sun sign’s traditional description. Eysenck and his collaborator David Nias eventually concluded, in their own later work, that the initial findings reflected self-attribution rather than any astrological signal: people who knew they were “supposed” to be extraverted (as a Leo, say) rated themselves as more extraverted, not because Leos are more extraverted, but because they’d been told they should be.
This early episode set a pattern that’s recurred throughout the subsequent decades: an apparent positive finding, on closer examination, traces back to participants’ prior knowledge of astrological trait descriptions rather than to any relationship between birth date and measured personality.
Why the Same Question Gets Asked Again
Given that this question has produced consistent results since at least the 1980s, it’s worth asking why sun sign research continues to be conducted at all — why, periodically, a new study tests essentially the same hypothesis with essentially the same null result.
Several factors contribute. First, sun sign research is cheap and easy relative to almost any other personality research question — birth date is often already collected as demographic data in large surveys conducted for entirely different purposes, which means testing sun sign against whatever personality measures that survey happens to include requires no new data collection at all. This makes sun sign a low-cost “bonus” analysis that researchers can run on existing datasets, sometimes almost as an afterthought to a study designed around a different question entirely.
Second, large-scale longitudinal and survey datasets continue to be created for unrelated purposes — health studies, economic surveys, educational research — and each new large dataset represents, in principle, a fresh opportunity to check whether sun sign shows up as a factor, with the statistical power that a fresh, large sample provides. Each of these represents a legitimate, if usually minor, test — and because these large datasets are valuable resources studied for many purposes over years, sun sign sometimes gets tested as one variable among very many, almost as due diligence, by researchers who aren’t necessarily invested in the outcome either way.
Third — and this is more about how findings get amplified than about why they’re conducted — any study that does find something, even by chance (and with thousands of datasets being analyzed for many variables, some will produce significant results purely by chance, a problem closely related to the multiple-comparisons issue discussed elsewhere in this series), tends to receive disproportionate attention relative to the much larger number of null results. A single study reporting a sun-sign correlation gets covered; the dozens of datasets where no one bothered to report the (null) sun-sign analysis don’t generate news.
What “Birth Month” Actually Does Correlate With
It’s worth being precise about something the sun-sign literature has actually found, because it’s easy to conflate with astrological claims when it isn’t one.
Birth month does show small but real correlations with some outcomes — but the explanations for these correlations are environmental and social, not astrological, and they don’t correspond to the personality traits sun-sign astrology assigns. The relative age effect discussed in the companion piece on Nine Star Ki is one example: children born just after a school-enrollment cutoff date tend to be relatively older than classmates born just before it, producing small but measurable effects on educational outcomes that have nothing to do with zodiac signs and everything to do with school administration.
Season of birth has also been linked, in some studies, to small effects on certain health outcomes — plausibly related to maternal nutrition, sunlight exposure, or seasonal infection patterns during pregnancy and early infancy, varying by hemisphere and latitude in ways that wouldn’t track a fixed zodiac calendar at all (a “Scorpio” born in the Southern Hemisphere experiences the opposite season from a “Scorpio” born in the Northern Hemisphere, which is a long-standing problem for any claim that seasonal effects underlie sun-sign astrology specifically).
These real, if modest, birth-month effects are sometimes invoked — by both proponents and critics of astrology — as if they were relevant to the sun-sign debate. They’re not, in either direction. They don’t validate sun-sign astrology, because they don’t correlate with the traits sun-sign astrology actually assigns, and their causal mechanisms (school cutoff dates, hemisphere-dependent seasons) are completely unrelated to anything astrology proposes. But they also mean that “birth date has zero effect on anything” isn’t quite the right takeaway either — birth date matters for some outcomes, through mechanisms that have nothing to do with the zodiac.
The Question That Doesn’t Go Away
What keeps the broader “personality by birth month” question alive, beyond the practical research-cost factors above, is probably something closer to what’s been discussed throughout this series: the gap between what controlled research can show and what individual experience seems to show.
Sun-sign astrology remains one of the most widely recognized and casually used frameworks for describing personality in everyday conversation, across cultures where it’s culturally salient. People continue to find that their sign “fits,” in the Barnum-effect sense discussed elsewhere, regardless of what the aggregate research shows — and this persistent subjective experience, multiplied across a very large population, creates ongoing folk-level interest in “is there something to this?” that academic research, however consistent its findings, doesn’t dissolve.
This is, in its way, the most interesting finding to come out of fifty years of sun-sign research — not the null result itself, which has been remarkably consistent, but the fact that the null result hasn’t reduced the cultural salience of the underlying belief at all. The research answers a narrow empirical question. It doesn’t touch the broader psychological and cultural reasons the question keeps feeling worth asking.