New Post
Search
Help
ArchivesRe: Mr Susan
in reply to a message by Anneza
I assume you are looking at the SSA lists as your "charts."
The quality of the data here deteriorates the farther back in time you go. It is quite obvious from the names you have found that there simply were a lot of mistakes made in the gender code when the original data for Social Security was entered into the SSA computer system. There is no way that there were as many boys given the popular girls' names like Susan, Mary, Elizabeth, Deborah, etc. as the SSA data says.
To look at one early year at random: the SSA top 1000 list for 1890 has 9 of the top 10 names for girls (Mary, Anna, Elizabeth, Margaret, Emma, Florence, Ethel, Minnie, and Bertha) are also on the boys' list; and the top 11 boys' names (John, William, James, George, Charles, Frank, Joseph, Robert, Henry, Harry, and Edward)also appear on the girls' list. Though I have run across very rare examples of girls named George, Frank, James, and Harry -- and Florence was used in earlier centuries as a male name in Ireland --
it is highly unlikely that the above pattern reflects reality, but much more likely that it's the result of errors made in typing the gender code in the original SSA data.
The quality of the data here deteriorates the farther back in time you go. It is quite obvious from the names you have found that there simply were a lot of mistakes made in the gender code when the original data for Social Security was entered into the SSA computer system. There is no way that there were as many boys given the popular girls' names like Susan, Mary, Elizabeth, Deborah, etc. as the SSA data says.
To look at one early year at random: the SSA top 1000 list for 1890 has 9 of the top 10 names for girls (Mary, Anna, Elizabeth, Margaret, Emma, Florence, Ethel, Minnie, and Bertha) are also on the boys' list; and the top 11 boys' names (John, William, James, George, Charles, Frank, Joseph, Robert, Henry, Harry, and Edward)also appear on the girls' list. Though I have run across very rare examples of girls named George, Frank, James, and Harry -- and Florence was used in earlier centuries as a male name in Ireland --
it is highly unlikely that the above pattern reflects reality, but much more likely that it's the result of errors made in typing the gender code in the original SSA data.
This message was edited 1/18/2006, 9:48 AM
1
Replies
I certainly see what you mean ... but what I don't understand is why parents in the 1980s (not the 1890s!) suddenly started making a statistically significant number of errors with their Elizabeth, Melissa and Sarah daughters. Had the forms changed and become confusing? Otherwise surely the 70s parents would have been doing the same kind of thing, and the 90s ones?
Anyway, what you didn't say is probably as important as what you did: you haven't had direct personal experience of any boy named Lizzie, right? (And if right, then PHEW!)
Anyway, what you didn't say is probably as important as what you did: you haven't had direct personal experience of any boy named Lizzie, right? (And if right, then PHEW!)
1
It wouldn't be the parents making the errors, it would be the clerks at the Social Security administration who put the data into the computer. Entering the sex code into the computer probably originally involved typing in either a "1" or a "2", and it would have been very easy for a mistake to have been made. I have no idea why there would be more errors in the 1980s than in the 1970s or 1990s, but the accuracy in something like this might depend partly on how well trained the individuals involved were. It's quite possible that the 1980s data were actually entered first, with the 1970s data only being entered in later along with the 1990s while the system was becoming completely computerized.
I certainly haven't had any direct personal experience of a boy named Lizzie; but then I hardly know every person in the USA! :)
I certainly haven't had any direct personal experience of a boy named Lizzie; but then I hardly know every person in the USA! :)
1