ban bossy – a case study in terrible statistics

Lately there has been a concerted push to ban the word “bossy”. The creator of this campaign quoted several studies in an effort to show that the term has negatively influenced young ladies. The problem with quoting statistics is that your arguments are easily torn to shreds if you do not understand what you are quoting. With statistics, people to need to understand “what is being measured”, “how is it being measured”, and “what are the likely implications” of that measurement. Crossing results from various studies is generally a terrible practice. But the Ban Boss bandwagon decided to go ahead anyways and get clobbered in the press.

Reason.com dissects the terrible use of statistics in detail:

“Girls are twice as likely as boys to worry that leadership roles will make them seem ‘bossy.'”

So what’s this about girls being twice as likely to worry that leadership roles would make them seem bossy? Well, the children who said they did not want to be leaders—again, fewer than one tenth of the total—were asked about the reasons for this lack of interest. In this small subsample, 29 percent of the girls and 13 percent of the boys agreed with the statement, “I do not want to seem bossy.” That’s about 2.5 percent of all girls, compared to just over one percent of boys. Truly, a dreadful scourge of future womanhood that calls for a massive social media campaign.

[Snip]

“The confidence gap starts early. Between elementary and high school, girls’ self–esteem drops 3.5 times more than boys.'”

The source for this is a 1991 study from the American Association of University Women, “Shortchanging Girls, Shortchanging America,” that amounts to 23-year-old junk science. Writing on the Psychology Today website in 2010, the late Susan Noelen-Hoeksma, a leading psychologist and a Yale University Professor, noted that “the study by the American Association of University Women was refuted by subsequent studies using large samples and better measures of self-esteem.” After reviewing the claims of a crisis in girls’ self-esteem and the relevant research, Noelen-Hoeksma concluded, “The phrase ‘much ado about nothing’ comes to mind.”

[Snip]

“By middle school, girls are 25 percent less likely than boys to say they like taking the lead.”

Ashe Snow, a columnist for The Washington Examiner, contacted the study’s author and learned that the question was only asked only once over the course of the study, so the wording which implies that girls become more leadership-averse as they get older is misleading. (“Change It Up” demonstrated the opposite.) Snow also demonstrates that another “Ban Bossy” claim—that “parents place a higher value on leadership for boys than for girls”—is based on a single statistic cherry-picked from a survey which, overall, shows that parents value leadership equally for their sons and daughters.

The default position when someone cites statistics should be skepticism. My wife and I always laugh at the 1 in 8 Americans experience hunger advertisements. In a country with no poverty and a major obesity problem among the lowest income earners, running out of Doritos is not struggling with hunger.

About christopher fisher

The blog is meant for educational/entertainment purposes. All material can be used and reproduced in any length for any purpose as long as I am cited as the source.
This entry was posted in Statistics. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s