Do zodiac signs predict personality? What studies show

“Leos are born leaders, Pisces are dreamy and emotional”: you have probably heard something like it. mind.io’s zodiac sign pages also describe how the twelve signs are usually portrayed in Western astrology, but from the start they present this as culture, read for fun. So is there any scientific basis for a link between your birth sign and your personality? This guide answers that question with research.

The short answer is that research has mostly found no meaningful link between zodiac signs and personality. In a double-blind experiment published in Nature in 1985, astrologers given only birth charts did no better than chance at picking the matching personality test result (Carlson, 1985), and a 2020 study of the personality scores of more than 170,000 people in China found no meaningful differences in personality by sign (Lu et al., 2020). A review of the research also describes how knowing your sign’s description in advance can tilt your self-image towards it, so there are reasons why it can feel accurate (Dean & Kelly, 2003).

This guide covers how the sign dates were set and what Ophiuchus is, what the experiments with astrologers and the large data sets found, why it feels accurate, when a sign becomes a label, and how to enjoy it without taking it as fact. “Zodiac sign” here means the Western sun sign set by your birthday; the East Asian zodiac animals, set by your birth year, are an entirely different system. A side-by-side comparison with MBTI, blood type and saju is in the guide comparing MBTI, blood type, zodiac signs and saju. mind.io does not offer a test that judges personality by zodiac sign.

Your Western zodiac sign is set by where the Sun stood on your birthday along the ecliptic, the path the Sun traces across the sky over a year. According to NASA’s science site for children, the Babylonians, more than 3,000 years ago, divided this path into 12 equal parts like slices of a pizza and picked one constellation for each slice; since they already had a 12-month calendar, each month got a slice of its own (NASA Space Place, 2016).

Yet even in the Babylonians’ own ancient stories there were 13 constellations along the path, and they left one, Ophiuchus, out. The real constellations differ in size and shape, so the Sun lines up with Virgo for 45 days but passes Scorpius in only 7, and it is also aligned with Ophiuchus for about 18 days each year (NASA Space Place, 2016). The astronomy site EarthSky gives those dates as roughly 30 November to 18 December (EarthSky, 2026).

The sky has also shifted over time. Because Earth’s axis no longer points in quite the same direction, a birthday between about 23 July and 22 August meant being born under the constellation Leo when the Babylonians set up the twelve signs, but 3,000 years later the Sun of someone born on 4 August lies in front of Cancer, one constellation earlier (NASA Space Place, 2016). If you have heard that “Ophiuchus has been added as a 13th sign”, the NASA page did not set up a new sign; it explained why the ancients left one out and why the sky has moved.

So has your sign changed? The signs used in horoscopes and astrology are not the constellations themselves but 12 equal 30-degree sections of the ecliptic. They are measured from seasonal markers such as the December solstice point, so the sign of Capricorn begins at the solstice around 21 December (EarthSky, 2026). The dates used in horoscopes are therefore tied to the seasons and stay the same, apart from boundary days that can shift by about a day from year to year; what has changed is the constellation behind the Sun on those dates. Ophiuchus is not on the list of horoscope signs (EarthSky, 2026), and mind.io’s zodiac sign date table follows the twelve sections too. NASA adds that, unlike astronomy, astrology’s claims are not based on scientific evidence (NASA Space Place, 2022).

One direct way to test zodiac personality is to ask astrologers to get it right. Shawn Carlson, of the physics department at the University of California, Berkeley, published such an experiment in Nature in 1985. It tested whole birth charts drawn from the date, time and place of birth, not only the sun sign, and it was designed with the help of astrologers as well as scientists and statisticians. Advising astrologers nominated by an astrological organisation, the NCGR, accepted the design as fair and then predicted that astrologers would get at least half of the cases right (Carlson, 1985).

The core of the experiment worked like this. Volunteers took the California Psychological Inventory (CPI), a standard personality test (Carlson’s paper calls it the California Personality Inventory), and the astrologers received one person’s birth chart together with three personality test results: one belonged to that person and the other two were picked at random from other participants, so a blind guess would be right one time in three. The CPI itself was chosen because the advising astrologers judged its traits closest to those astrology can discern (Carlson, 1985).

In Carlson’s analysis, across 116 charts, the astrologers’ first choice was the right personality test result at a rate of 0.34, almost exactly the chance level of 0.33. That was 3.3 standard deviations below the 0.5 the astrologers had predicted, and by his analysis they were not right more often when they rated a result as a strong fit (Carlson, 1985).

The experiment has limits. In the first part, participants picking their own reading from three written by astrologers also scored at chance, but they could not pick out their own personality test results better than chance either, so Carlson wrote that this part could not be used to rule against astrology (Carlson, 1985). And Suitbert Ertel, of the University of Göttingen in Germany, argued in a 2009 reanalysis that the astrologers’ matching was slightly better than chance at a borderline level (p = 0.054), with a similar result on the 10-point ratings (p = 0.04), so that Carlson’s negative verdict could not stand, while himself concluding that these results were not enough to regard astrology as verified (Ertel, 2009).

Nor is this the only such experiment. According to a review by Dean and Kelly, more than forty studies have asked astrologers to match birth charts with personality profiles or case histories (nearly 700 astrologers and 1,150 birth charts), and a meta-analysis of them gave a mean effect size of 0.051, indistinguishable from chance (p = 0.66). The authors concluded that nothing there suggests astrologers can perform usefully better than chance (Dean & Kelly, 2003).

Whatever the astrologers’ skill, if each sign had its own personality, simply sorting many people’s personality scores by sign should show differences. A 2020 study in the Journal of Personality and Social Psychology, a journal of the American Psychological Association, analysed the answers of 173,709 people aged 18 to 60 on an advertisement-free Chinese personality-test website. Participants answered 40 adjective items measuring the Big Five personality traits, and the researchers worked out each person’s sign from their birthday (Lu et al., 2020).

Agreeableness, conscientiousness, emotional stability and openness did not differ meaningfully by sign. Only extraversion seemed to differ, but that came from the season of birth and disappeared once season was taken into account. Grouping the signs by element (fire, earth, air, water) or by “masculine” and “feminine” signs showed no differences either, and the adjectives usually attached to each sign (such as “critical” for Virgo) were not especially high for that sign. The results were the same for believers in astrology alone and for non-believers alone (Lu et al., 2020).

The season effect itself was tiny. People born in summer were slightly more extraverted than those born in other seasons, but the difference was an effect size (d) of only 0.02 to 0.03, which the researchers also called small (Lu et al., 2020). That is a finding about seasons, not support for the twelve signs. The same team also analysed the personnel records of 32,878 employees of a large Chinese company, and sign did not predict their job performance scores (Lu et al., 2020). The first data set, though, came from Chinese respondents who chose to visit an online personality-test site, so it is not a sample that represents a whole country.

There is an even sharper test. If astrology were right, “time twins”, people born at almost the same moment in the same place, should resemble each other. Dean and Kelly used data gathered by a study unrelated to astrology on 2,101 people born in London on 3 to 9 March 1958, on average 4.8 minutes apart. They compared 110 measures taken at ages 11, 16 and 23, including IQ, reading and arithmetic scores, teacher and parent ratings of anxiety, aggressiveness and sociability, height and weight, occupation and marital status, and found no tendency for people born one after another to be more alike. Of the 110 measures, 5 came out statistically significant, close to the 5.5 expected by chance (Dean & Kelly, 2003). All of them were born under Pisces, yet, in the authors’ words, they evidently had few similarities.

Some studies looked beyond the sun sign to the moon sign and the ascendant. In a 1996 study of 190 first-year university students, the only significant difference was a higher mean extraversion score for people with both the Sun and the Moon in “positive” signs than for those with both in “negative” signs, and the authors concluded that their findings generally did not support the theory that signs determine extraversion and emotionality (Clarke et al., 1996).

It is not that no study ever reported a link between signs and personality. A 1978 study by Mayo, White and Eysenck gave a personality test to 917 men and 1,407 women and, as astrological writings predict, found that people born under the odd-numbered signs (Aries, Gemini, Leo and so on) were more extraverted, and people born under the water signs (Cancer, Scorpio, Pisces) scored higher on emotionality (Mayo et al., 1978).

According to Dean and Kelly’s review, however, this link between signs and extraversion disappeared in later studies when the participants had no prior knowledge of astrology. That was taken to show that knowing how your sign is described can tilt your self-image towards it (Dean & Kelly, 2003). In other words, the sign did not shape the personality; the sign descriptions may have seeped into the way people describe themselves. In the Chinese study above, though, no personality differences appeared even among believers alone (Lu et al., 2020).

Two familiar habits of mind add to this. One is the Barnum effect, taking a vague description that fits almost anyone as being about you. In 1949 the psychologist Bertram Forer handed his students what he said were individually written personality sketches, which were in fact all the same text, and the students rated it as a good fit for themselves (Forer, 1949). The other is confirmation bias, remembering the moments that match what we expect and letting the others slip by. “Such a Leo, born for the stage” sticks in your memory, while the shy Leo friend is easily forgotten.

Plenty of people do believe in it. In a 2018 Pew Research Center report, 29% of American adults said they believe in astrology (Pew Research Center, 2018), and in the 2018 US General Social Survey, asked whether astrology is scientific, 58% said it is not at all scientific but 33% said it is sort of scientific and 6% very scientific (National Science Board, 2020). So there is nothing strange about nodding along to a zodiac description. That feeling, though, is not evidence that your sign shapes your personality.

Zodiac talk is not always used lightly. The 2020 research team above also looked at what stereotypes the signs, imported from the West, had created in China. Astrological stereotypes were salient in China but not in the United States, and in China Virgo was the most negatively evaluated sign, stereotyped as fussy and disagreeable. The researchers thought this was likely because Virgo was translated into Chinese with a name meaning “virgin” (Lu et al., 2020).

The stereotype turned into behaviour. In an experiment on a popular Chinese dating app, three male profiles with the same photo, nickname and age, differing only in sign, were shown for nine days: the Leo profile received 41 “likes” and the Libra profile 46, but the Virgo profile only 15. In simulated hiring experiments, both employed adults and human-resources professionals were less willing to hire a candidate whose résumé said he or she was a Virgo (Lu et al., 2020).

Yet in the same research, Virgos’ actual personality scores and job performance did not differ from those of other signs (Lu et al., 2020). A groundless stereotype can still change real opportunities once it is used to pick or rule out people. The research was done in China, and in the same research astrological stereotypes were not salient in the United States. Still, judging a date or a job candidate by sign can turn a joke into an unfair label for the person on the other end.

The same goes for zodiac compatibility. A study that followed marriages and divorces in Swedish population register data from 1968 to 2001 found no consistent evidence that couples deemed more compatible by astrology were more common among actual marriages or less likely to divorce (Helgertz & Scott, 2020).

Zodiac talk is a handy conversation starter for birthdays, for someone new, or for teasing friends and laughing together. What the research shows is that your sign cannot tell you someone’s personality, not that you should never talk about it. A few habits make it much easier to enjoy.

First, read the other signs’ descriptions too. If you are a Leo, put the Virgo and Pisces descriptions side by side with yours, and you will probably find quite a few lines that make you nod even though they are not about your sign. It is the easiest way to see the Barnum effect for yourself.

Second, turn the descriptions into questions. Instead of “You are a Scorpio, so you hold grudges, right?”, ask “How do you deal with it when someone hurts your feelings?”, and the same topic helps you actually get to know the person. Third, laugh off the descriptions that do not fit, and never use them to pick or rule out a date or a colleague. Zodiac sign compatibility is most fun read as a way to learn about each other.

Fourth, on days when you want to know yourself better, try a test that measures your current tendencies from your answers. The Big Five personality test measures the five traits the Chinese study above looked at, and the MBTI test shows a type built from four preferences. Unlike a zodiac sign, both give results from the items you answer, and the results can change when you change. How far you can trust the MBTI is covered in the guide to MBTI accuracy. A test result does not mean it is linked to your sign.

Q. Is there any scientific evidence for zodiac sign personality? A. Going by the research so far, the fair answer is no. Astrologers did no better than chance in a double-blind experiment where they matched birth charts to personality test results (Carlson, 1985), and the same held in an analysis combining more than forty similar studies (Dean & Kelly, 2003). A study of 173,709 people in China also found no meaningful personality differences by sign (Lu et al., 2020).

Q. But my sign’s description fits me perfectly. A. Feeling that way is natural. Knowing your sign’s description in advance can tilt your self-image towards it (Dean & Kelly, 2003), and the Barnum effect, taking a description that fits almost anyone as being about you, plays a part too. Read the other signs’ descriptions and you will likely find that those lines fit you quite well as well.

Q. Was Ophiuchus added as a 13th sign? Has my sign changed? A. No. The Sun passes in front of Ophiuchus for about 18 days each year, but the Babylonians left it out from the start to fit their twelve months and picked only twelve (NASA Space Place, 2016). The signs used in horoscopes are 12 equal 30-degree sections of the ecliptic tied to the seasons, so their dates stay the same apart from boundary days that shift by about a day from year to year, and Ophiuchus is not on the list of horoscope signs (EarthSky, 2026). Because Earth’s axis has shifted, though, the actual constellation behind the Sun on your birthday may be about one sign earlier than your horoscope sign.

Q. Is zodiac compatibility reliable? A. A study that tested it with Swedish population register data did not support astrological compatibility. In data from 1968 to 2001, there was no consistent evidence that couples deemed well matched by astrology were more common among married couples or less likely to divorce (Helgertz & Scott, 2020). Enjoy compatibility charts as a way to get to know each other.

Q. Is there a link between zodiac signs and MBTI? A. There is no evidence of one. The MBTI measures your current preferences from the items you answer, while your sign is set by your birthday and stays the same for life. Personality measured with the Big Five did not differ meaningfully by sign either (Lu et al., 2020). A detailed comparison is in the guide comparing MBTI, blood type, zodiac signs and saju, and the research on blood type personality is covered in the guide to the science of blood type personality.

Q. Is there a zodiac personality test? A. mind.io does not offer a test that judges personality by zodiac sign. The zodiac sign pages are a reading, for fun, of how the twelve signs are usually portrayed in Western astrology. If you want to explore your tendencies from your own answers, try the Big Five personality test. That does not mean the result is linked to your sign.

References: Carlson, S. (1985), A double-blind test of astrology, Nature 318(6045), 419-425, https://doi.org/10.1038/318419a0; Ertel, S. (2009), Appraisal of Shawn Carlson’s renowned astrology tests, Journal of Scientific Exploration 23(2), 125-137, https://journalofscientificexploration.org/index.php/jse/article/download/99/37; Dean, G., & Kelly, I. W. (2003), Is astrology relevant to consciousness and psi?, Journal of Consciousness Studies 10(6-7), 175-198, author’s version at https://astrology-and-science.com/D-psia2.htm; Lu, J. G., Liu, X. L., Liao, H., & Wang, L. (2020), Disentangling stereotypes from social reality: astrological stereotypes and discrimination in China, Journal of Personality and Social Psychology 119(6), 1359-1379, https://doi.org/10.1037/pspi0000237; Clarke, D., Gabriels, T., & Barnes, J. (1996), Astrological signs as determinants of extroversion and emotionality: an empirical study, The Journal of Psychology 130(2), 131-140, https://doi.org/10.1080/00223980.1996.9914995; Mayo, J., White, O., & Eysenck, H. J. (1978), An empirical study of the relation between astrological factors and personality, The Journal of Social Psychology 105(2), 229-236, https://doi.org/10.1080/00224545.1978.9924119; Helgertz, J., & Scott, K. (2020), The validity of astrological predictions on marriage and divorce: a longitudinal analysis of Swedish register data, Genus 76, 34, https://doi.org/10.1186/s41118-020-00103-5; Forer, B. R. (1949), The fallacy of personal validation: a classroom demonstration of gullibility, Journal of Abnormal and Social Psychology 44(1), 118-123, https://doi.org/10.1037/h0059240; NASA Space Place (2016), Constellations and the calendar, article last updated 13 January 2016, archived at https://web.archive.org/web/20161001233309/http://spaceplace.nasa.gov/starfinder2/en/; NASA Space Place (2022), What are constellations?, article last updated 6 December 2022, https://spaceplace.nasa.gov/starfinder2/en/; McClure, B. (2026), Is Ophiuchus the 13th constellation of the zodiac?, EarthSky, 19 August 2026, https://earthsky.org/constellations/ophiuchus-the-serpent-bearer-13th-constellation-zodiac/; Pew Research Center (2018), ‘New Age’ beliefs common among both religious and nonreligious Americans, 1 October 2018, https://www.pewresearch.org/short-reads/2018/10/01/new-age-beliefs-common-among-both-religious-and-nonreligious-americans/; National Science Board (2020), Science and Engineering Indicators 2020 (NSB-2020-7), Table S7-32: Public assessment of astrology, by respondent characteristic: 1979-2018, https://ncses.nsf.gov/pubs/nsb20207/assets/supplemental-tables/tables/nsb20207-tabs07-032.pdf