Alternate dates of birth for our ancestors, perhaps ranging over several years, are common for many of us, and the reasons can vary considerably. A recent example in my own research came in the form of a deliberate change of birth date, with the sole intention to make the subject appear younger than her husband.
As shown in Albert A. Pomeroy’s History and genealogy of the Pomeroy family, Frances Pomeroy was born 22 November 1893, the daughter of Sanford and Mary (Lottimer) Pomeroy. Her father was an American artist living in Paris where Frances herself was born, and Frances never visited the United States until 1900, according to her 1915 passport application. Between 1908 and 1919, I found six separate records confirming her birth in 1893.
Then, in 1920, as summarized in the Pomeroy genealogy, she returns to the United States permanently to get married. From that point on her age “changes” by two years.
I found this new “1895” birth date in twelve records, including her death certificate and gravestone, where her daughter was the informant.
Frances’s husband Charles Warren Lippitt was born 15 May 1894. Frances was only six months older than Charles, but it appears that that was her reasoning to change her date of birth. Instead of both being 26, Charles and Frances were now 26 and 24, and Frances was a “respectable” eighteen months younger than her husband. The only record after 1920 that gives Frances’s correct 1893 birthdate is the above Pomeroy genealogy in 1922 (part 3). However, that date of birth had first appeared in part 2 of the same genealogy, published in 1912, when Frances was still using the 1893 date. Likely the genealogist kept the same date as had been earlier supplied and only updated her sketch with her marriage and how they met in France, and did not “update” her year of birth! Charles and Frances Lippitt divorced in 1937 and Frances still used the later date of birth on the 1940 census; her daughter supplied the same year on her mother’s death certificate and gravestone.
Being six months older than your husband may seem trivial to us, but this appears to have mattered to Frances in 1920! While not every age variation is as clear and deliberate as this one appears to be, it’s nonetheless an example of what could motivate an ancestor to “fudge” facts.
 Passenger lists in 1908 and 1919 provide an age; U.S. passport applications in 1915, 1916, 1918, and 1919 provide an exact date of birth.
 1920 New York City Marriage Certificate; Passenger lists from 1928, 1931, 1933, 1936; U.S. Passport applications from 1924; 1925 and 1935 Rhode Island State Census; 1930 and 1940 U.S. Federal Census; 1936 Social Security Application; Death Certificate and gravestone. As in the earlier passenger lists, only France’s age was given. Frances appears on lists of US. Citizens. Most people on these lists were given their full date of birth and U.S. birthplace. However, Frances (born in France) is only given her age with the notation that she is a U.S. citizen by virtue of being born to U.S. parents, or after 1920, by virtue of her marriage to an American citizen (citizenship rights for women in the U.S. at the time of their marriage in 1920 were dependent on the citizenship of their husbands).