This is from CampusReform, about a psychology experiment (the kind of experiment that has led to  the "replication crisis" in psychology) about avoiding "stereotypes".

Profs: Stop saying 'girls are as good as boys at math'

By Celine Ryan, Campus Reform, July 12, 2018

• A Stanford University professor wants people to avoid saying that “girls are as good as boys at math,” arguing in a recent study that this can "unintentionally perpetuate bias."
• Instead, the authors of the study suggest saying that "girls and boys are equally good at math," or that "boys and girls are equally good at math."

A new Stanford University study warns against claiming that “girls are as good as boys at math,” saying this may “unintentionally perpetuate bias.”

“Some well-meaning statements can spread stereotypes,” proclaims the headline of a university press release touting the study, highlighting the study’s conclusion that such sentences “frame one gender as the standard for the other."

"Gently ‘nudging’ participants through more implicit messages might...be a more effective strategy for counteracting gender stereotypes and encouraging more egalitarian behavior."   

“On the surface, the sentence tries to convey that both sexes are equal in their abilities,” the university explains. “But because of its grammatical structure, it implies that being good at math is more common or natural for boys than girls.”

The study, titled “Girls Are as Good as Boys at Math” Implies That Boys Are Probably Better: A Study of Expressions of Gender Equality,” was conducted by Stanford psychology professor and Associate Dean for the Social Sciences Ellen Markman and researcher Eleanor Chestnut, a postdoctoral scholar at New York University.

The researchers first asked participants to indicate whether they believe girls or boys are naturally more skilled at math, finding that 67 percent attributed more natural math ability to boys.

They then had participants read a paragraph from a study showing a lack of gender differences in math ability, but altered the wording to frame the information in one of four ways: “girls do just as well as boys at math,” “boys do just as well as girls at math,” “girls and boys are equally good at math,” and “boys and girls are equally good at math.” 

After reading their text, participants were asked which gender they believed to be more naturally competent in math. [More]

So this is all about how you phrase the idea that boys and girls have equal math ability, to see if you can convince students. But there's not much talk, even at CampusReform, about whether girls and boys have equal math abilty, although to her credit, CampusReform's Celine Ryan does refer to the "dubious claim that both sexes “are equally good at math”.

But they're not. Math nerd John Derbyshire wrote once that

On the wall in the main room of the Mathematics Library at Columbia University are four huge pictures of mathematicians. Left to right, they are: Henri Poincaré, Sofia Kovalevskaya, Carl Friedrich Gauss and Emmy Noether. Male, female, male, female — perfect sexual equality. Except that neither Noether nor Kovalevskaya were even close in mathematical powers to Poincaré and Gauss.

I can quantify that statement. My colleague Charles Murray (Losing Ground, The Bell Curve) has a book coming out soon in which all the great names of all intellectual disciplines are ranked by the number of references to them in other people's work. He tells me that the rankings — among mathematicians, that is — for Poincaré, Kovalevskaya, Gauss and Noether are, in order: 25, 106, 3, 95.

If you want an academic citation for something everybody used to know, see this from Mankind Quarterly:

This study analyzes gender differences in scholastic achievement of 15-year-olds in 75 countries that participated in the OECD’s PISA (Programme of International Student Assessment) tests. The cultural and socioeconomic determinants of three outcome measures were investigated: (1) overall achievement (math, reading, science average); (2) gender-typicality of achievement profiles (higher reading combined with lower math scores for females); and (3) the male/female variance ratio as a measure of the extent to which male scores are more variable than female scores. Females achieved higher reading scores than males in all countries, usually by a wide margin. Males achieved higher mathematics scores in 61 out of 75 countries, and gender differences in science were negligible.

Gender Differences in School Achievement across Cultures: An Analysis of Results from PISA 2000-2012
Gerhard Meisenberg, Mankind Quarterly, Volume 57, Number 2, 2016

Of course, in attempting this whole psychological influence test on young people, the experimenters did have the subjects read a paragraph "from a study showing a lack of gender differences in math ability.”

This quote, by "Miss Manners" is relevant:

"Miss Manners has a simple way of dealing with such figures provided by sociology or its gullible younger sibling, the polls. She just dozes off the minute she hears the phrase "studies show ... " and wakes up in time for the study that shows the opposite."[Dressing Down Slipping Up, By Judith Martin, September 18, 1994]