The Museum of Statistical Anomalies by Brock Dubbels
August 22nd, 2009 7:27 PMMartha Whales Skelton Jefferson died at 34 -- although she died before Thomas became president, and even though he never remarried, he did have children with Sally Hemmings -- she was not considered a first lady.
and
Bess Truman died at 97, 10 years after her husband.
First ladies Ages of death
Median -- 70
Quartiles -- 60 and 81.5
Extremes -- 34 and 97
Another interesting piece -- Harriet Lane served as first lady to James Buchannan, but he was unmarried and she was his niece.
The only women close to Bess Truman is Nancy Reagan at age 88.
3 vote(s)
Terms
(none yet)9 comment(s)
Hrm. I'm going to insist on a higher standard for statistical anomalies, unfortunately. (My apologies--I'm kind of sensitive to statistics, and might be a little bit snarky as a result.) First, being distant from the median doesn't necessarily make it an outlier. One way that might help is to try and come up with a model for how long the first ladies live, like a regression based on date of birth (people live longer now). (I'd be tempted to also consider the fact that many of the first ladies are still alive, so use some analysis that incorporates that fact.)
But before all that, at least some sort of plot of length of life compared to time of birth--or even just a boxplot of the different lengths of life.
And before that, also provide a link to the data you did the analysis on (ideally in a nice csv)
But even before that, some consideration of the data. I mean--you first say that Martha Jefferson is not a first lady, then use her as one of your outliers. Also, if you're looking for first ladies with lives nearly as long as Bess Truman at 97, perhaps Lady Bird Johnson at 94? And if you consider Nancy Reagan...she's still alive, as is Betty Ford, who is older.
A couple parts of the task are missing, too: "create a display for it that elucidates its irregularity" and "Work with other players to create the Museum of Statistical Anomalies by combining your displays."
Interesting data set and idea--but I'm going to ask for more thought before I sign off on this as a good analysis. Good luck!
http://en.wikipedia.org/wiki/List_of_First_Ladies_of_the_United_States_by_longevity
I can see where the definition of first lady might only be reserved for women who participated in that role and does not include those who were dead before they had their shot. However, I think what makes them interesting for inclusion is whether they are discarded as outliers or included in history as important in the data set.
Why include Martha? Thomas Jefferson was so devastated he never remarried, although historians believe he may have had children with Sally Hemmings -- there are DNA matches for African Americans as possible descendants.
An outlier is a data point that is not consistent with the bulk of the data -- unusually high or low, and these women were at the extremes -- this data set comes from women alive in 2002.
Hannah Hoes van Buren could also be considered an outlier having died at the age of 36, but she never held that status either -- or Alice Hathaway Lee Roosevelt at age 22.
Between these three, Hannah, Alice, and Martha, there really are no others close to their young ages, and on the other end, the only women who come close to Bess is
Ladybird Johnson at 94.
The choice of data seems to be open here, and the idea is to look at anomalies, such as the lifespan of women who served as first lady. I believe that these women present a more interesting story and the effects on the men who became president. Here is what the Wikipedia offers about Alice Roosevelt:
"While he made some oblique references to Alice in the months after her passing, Roosevelt never spoke of her publicly again. He refused to have her name mentioned in his presence. When asked about her mother by Alice's daughter and namesake, she was referred to Roosevelt's sister, Ana/Bye/Bamie for information and learned details of her mother only from her aunt. So final was this decision to try to put Alice's loss out of his life, that she is not even mentioned by name in his autobiography.
The Roosevelts had one daughter:
* Alice Roosevelt Longworth (1884-1980). According to a number of historians, Roosevelt's need to leave behind or suppress his experiences with his first wife were a source of deep resentment by his daughter. She was unable to get him to talk about her mother in any meaningful way. Her rebellious life finds some explanation in this sad aspect of her relationship with her father.
In the immediate aftermath of his wife's death, Theodore turned the care of their newly-born infant daughter to his elder sister Anna, also known as Bamie, and embarked on a journey of personal discovery to his ranch in the Badlands of North Dakota. From this interlude Roosevelt would emerge a renewed man and would go on to the Presidency of the United States in 1901.
Alice Hathaway Roosevelt was buried in Green-Wood Cemetery in Brooklyn, New York, next to her mother-in-law."
The beauty of this simple analysis is that it allows for the integration of the qualitative that begins to investigate the stories of the families and those who lived on in positions of great privilege and power. The qualitative begins to breath life and intrigue into these women who might only be outliers in number but not in their longer effect on the presidency and their legacies -- perhaps more so than those who lived and chose china patterns, and perhaps less than those who sought influence and change with their position like Eleanor who became an advocate for civil rights.
Yes, much more complex tests could be run, and we could look at average life expectancy and use it as a meditating factor over time by grouping in eras -- industrial eras were and agricultural times were much harder for women--especially in regard to child birth. However, these women were often privileged and had a greater opportunity for success at birth than most women du to wealth and access to doctors.
I do believe however, that looking at quartiles will show us the extreme points.
The points you make are well-founded however and perhaps I will run a regression.
How might we collaborate to combine a display?
if and when i'm doing this task, you will get no csv's, my friend! There may come a museum, but no .csv! None for you!
That's okay--but not providing csv will only be a poison in your own heart, BU. I will try to lead by example. I will think of csv-providing as a parable for "Love your enemies, bless them that curse you." Hmm, statistics and religion...dangerous ground...
That is too much SAS, a MEAN statement and perhaps Deviant. I have no need for your .csv. These are generic micro-soft threats you pose of little significance. My P-value has never been higher and you are making a TYPE I error. Come back to the mean ye outlier.
:) I'm a sucker for statistics and for puns.
im not clear on what the anomaly is. am i just being epically duncish?
the task asked for identification of an outlier on an interesting data set and to do a statistical analysis and display. What I presented were presidential wives and first ladies who were outliers as a mixed-method presentation.
Thus young and old are the anomalies
Perhaps significant artifacts that these women owned could be acquired and used as Objects of Power.