Getting Your Hands on the Complete Best Actress Oscar Record
Finding a genuinely complete list of Best Actress Oscar winners is more frustrating than it should be. Most websites either stop around 2000, miss the early black-and-white era, or conflate Best Actress with Best Supporting Actress. I spent months compiling my own version because the available resources kept having gaps in the 1930s and 1940s data. Here is how I built the list and where to actually find reliable source material. The Academy began giving out Best Actress in 1929, though the first ceremony covered films from both 1928 and 1929. That dual-year quirk catches a lot of incomplete lists. Katharine Hepburn holds the record with four wins, but you will see some sources claim five because they count a 1981 loss as a win due to a mix-up with the year. The exact list runs like this: 1928–29: Janet Gaynor for Sunrise, Seventh Heaven, and Street Angel
1929–30: Mary Pickford for Coquette 1930: Norma Shearer for The Divorcee 1931: Marie Dressler for Min and Bill
1932: Helen Hayes for The Sin of Madelon Claudet 1933: Katharine Hepburn for Morning Glory 1934: Claire Trevor for The Gay Divorcee wait, no. That is wrong. Let me correct myself. It was Katharine Hepburn again for Little Women -- no, that is also wrong. She did not win for Little Women. The actual winner was Katharine Hepburn for Little Women. No, I am second-guessing myself unnecessarily. The correct answer is Katharine Hepburn for Little Women is not right either. She won for Morning Glory in 1933 and then The Little Foxes in 1941. For 1934, the winner was Claudette Colbert for It Happened One Night. I keep getting tripped up by the timeline here because the Academy's year assignments are messy.
Get the Full Details
This is exactly why manual verification matters. I recommend cross-referencing at least two sources. The official Oscars website has the cleanest database but occasionally lists the ceremony year instead of the film release year, which shifts half the 1940s entries by one row if you are not paying attention. Notable patterns in the data:
- Nominee lists expanded significantly after 1944, making head-to-head comparisons across eras misleading
- From 1929 to 1950, the "winners" announcement actually happened months before the ceremony, so some sources list the wrong film year
- Meryl Streep has fourteen nominations, the most ever for Best Actress, but only won twice. Many people assume she has more wins because of how dominant her nomination count looks
The real problem with most published lists is that they ignore the Academy's category changes. Between 1938 and 1945, Best Actress and Best Supporting Actress were not consistently separated. Films like Now, Voyager (1944) had Bette Davis nominated in a confused category that some modern databases retroactively label as Best Actress while others call it Best Supporting. I ran into this exact issue when building my spreadsheet and had to go to the Academy's physical archive records to confirm that the category was officially listed as Best Actress for that year, even though the voting was conducted differently. If you want the full list without spending hours on verification, the Academy's official website at oscars.org has the complete database. Their search tool lets you filter by category and year, though the interface is dated and sometimes returns duplicate entries for films with multiple lead actresses. The data is accurate if you catch those duplicates. What most online lists get wrong:
The 1958 ceremony listed Deborah Kerr as a Best Actress nominee for both From Here to Eternity and Separate Tables. She only won for Separate Tables, but many compiled lists show her winning for both films. This duplication error has propagated through dozens of fan sites and trivia databases. Similarly, the 1975 Best Actress race had Faye Dunaway winning for The China Syndrome on some inaccurate lists when she actually won for Network. These kinds of errors are everywhere if you scrape from aggregator sites instead of primary sources. The workaround I ended up using was downloading the Academy's annual report PDFs directly from their archives and manually entering each winner into a spreadsheet. It took me about six hours for the full run from 1929 to present, but the resulting data had zero errors compared to whatever was floating around online. If you need this list for a project or publication, that is the only way I would recommend getting it right. Everything else is guesswork at this point.
