Marilyn vos Savant

Early life and the childhood test

Marilyn vos Savant was born Marilyn Mach in St. Louis, Missouri on August 11, 1946. Her father Joseph Mach was a German immigrant; her mother Marina vos Savant was Italian. She adopted her mother's family name as her professional surname in adulthood, an unusual choice for an American writer of her generation.

Her Stanford-Binet test was administered when she was 10 years old at her elementary school in 1956. The score - 228 - is derived from the ratio-IQ formula used in the 1937 Stanford-Binet revision: mental age divided by chronological age, multiplied by 100. At age 10 her measured mental age was reported as 22 years and 10 months, producing the headline number.

She married twice before her current marriage to Robert Jarvik, the cardiac surgeon who invented the Jarvik artificial heart. She worked in her family's clothing business in St. Louis through her twenties and only began writing publicly in her mid-thirties, after winning an Omni magazine intelligence-quiz contest in 1985 that brought her national attention.

The Guinness World Records listing

In 1986 the Guinness Book of World Records added a category for "Highest IQ" and placed vos Savant at the top, citing her childhood Stanford-Binet result. She held the listing through three editions (1986, 1987, 1988-89) until Guinness discontinued the category entirely in 1990, on the grounds that IQ scores from different tests could not be meaningfully compared and that ratio-IQ scores at the extreme range had unreliable confidence intervals.

The Guinness retirement of the category was widely covered in the press and is sometimes mistakenly read as a retraction of her score; it was not. Guinness explicitly stated that no single number could legitimately be called the world record. The decision was a methodological one, not a comment on any individual.

Vos Savant has acknowledged this directly in print, noting in her column that the Guinness listing reflects test conventions of the 1950s and is not a current measurement. She has also taken several adult tests, including the Mega Test designed by Ronald Hoeflin, on which her score was again at or near the top of the published distribution.

The Monty Hall problem (1990)

In her September 9, 1990 Parade column, a reader asked vos Savant a probability question: in a game show, a contestant chooses one of three doors. Behind one door is a car; behind the other two, goats. The host (who knows what is behind each door) opens one of the unchosen doors to reveal a goat, then asks the contestant whether they want to switch to the remaining unopened door. Is it advantageous to switch?

Vos Savant's answer was yes - switching doubles the probability of winning, from 1/3 to 2/3. She was attacked by hundreds of professional mathematicians, including a now-famous letter from a PhD statistician at George Mason University insisting she was wrong and that she should "confess." Approximately 10,000 readers wrote in disagreement, of whom around 1,000 had advanced degrees.

She was correct. The result, now standardly called the Monty Hall problem, has been formally proven and is a staple of introductory probability courses. The episode became a case study in how mathematical intuition can fail at the boundary between conditional and unconditional probability. Several of the original objectors later wrote retractions.

The Ask Marilyn column and other writing

The "Ask Marilyn" column has run continuously in Parade Magazine since 1986 - more than 35 years and over 1,800 individual columns. The column answers reader-submitted questions across mathematics, logic, science, and general reasoning. Selected columns have been collected into several books including The World's Most Famous Math Problem (1993) and Ask Marilyn (1992).

Beyond Parade, vos Savant has published on Fermat's Last Theorem (a book-length critique of Andrew Wiles' proof - which she later acknowledged was substantially weaker than the original proof), the Riemann Hypothesis, and applied probability. Her books have been widely-read but mixed in critical reception; her strength is plain-English explanation rather than original mathematical research.

She has been a contributing editor to Parade since 1991 and was inducted into the National Women's Hall of Fame in 1988. She lives and works in Manhattan, where her column has been continuously produced since the 1980s.

Criticism and methodology debates

Vos Savant's 228 figure has been criticized on technical grounds for decades. The most substantive critique: ratio-IQ scores at extreme ranges are calculated from very thin standardization samples. There are simply not enough children in the population at any given age scoring near the ceiling to derive precise confidence intervals. Most modern psychometricians treat any childhood IQ over about 180 as approximate rather than precise.

A second critique is that the early Stanford-Binet (1937 revision) and adult deviation IQ tests are not on the same metric. A 228 on a 1956 ratio-IQ administration does not translate to a 228 on a WAIS-IV. The closest defensible cross-walk would place her measurement in the deviation-IQ range of roughly 170-180, which is still at the extreme right tail but not a singular world-record figure.

Vos Savant has been thoughtful about this in print. She has neither defended the 228 as a current measurement nor dismissed her own measured ability; her stated view is that the score reflects the testing conventions of her time and that what matters is the work, not the number.

Timeline

• 1946Born Marilyn Mach in St. Louis, Missouri.
• 1956Stanford-Binet administration, age 10. Reported score: 228.
• 1985Wins the Omni magazine IQ contest; takes the Mega Test.
• 1986Begins "Ask Marilyn" column in Parade Magazine. Added to Guinness Book of World Records.
• 1988Inducted into the National Women's Hall of Fame.
• 1990Monty Hall problem column; receives ~10,000 letters of disagreement; her answer is later proven correct.
• 1993Publishes The World's Most Famous Math Problem on Fermat's Last Theorem.
• 2026Ask Marilyn column still running after 40 years.

Frequently asked questions

Comparable scorers