This clearly isn't an especially rigorous analysis
Not especially, no.
Data samples consisting of two or five articles per keyword??
And 2 out of all 5 penalized "female" entries have nothing to do with social issues etc. (one is about female mice, another about mixed-sex animals).
Comparing stories on social justice against ARM and Intel is also cherrypicking, because we don't know whether this is supposed to be a bias against social issues, or just non-tech stories in general, or something else.
And what's your methodology of choosing keywords?
"Female" - which made it onto your list - gets penalized 5 to 1, all right.
But then in case of "girls" (omitted from your list) it's actually 0 - 3, telling us a different story.
And unlike with the "female" bunch, none of these articles is about mice. They're about how girls get better grades, how to get them into engineering and why it's easy to teach them code.
"Women" (plural) lose 13 - 4 and so the keyword is featured on your list, but "woman" (singular) wins 7 - 2, and surprise surprise, it's absent.
"equal", as a word stem (counting "inequality", "unequal" etc.) - penalized 2 times, not penalized 3 times. Holy smokes Batman, it rates better than "x86"!
Statistics, huh ;)
If this submission gets penalized, do we count it as bias against social justice, or junk science
Data samples consisting of two or five articles per keyword??
And 2 out of all 5 penalized "female" entries have nothing to do with social issues etc. (one is about female mice, another about mixed-sex animals).
Comparing stories on social justice against ARM and Intel is also cherrypicking, because we don't know whether this is supposed to be a bias against social issues, or just non-tech stories in general, or something else.
And what's your methodology of choosing keywords?
"Female" - which made it onto your list - gets penalized 5 to 1, all right.
But then in case of "girls" (omitted from your list) it's actually 0 - 3, telling us a different story.
And unlike with the "female" bunch, none of these articles is about mice. They're about how girls get better grades, how to get them into engineering and why it's easy to teach them code.
"Women" (plural) lose 13 - 4 and so the keyword is featured on your list, but "woman" (singular) wins 7 - 2, and surprise surprise, it's absent.
"equal", as a word stem (counting "inequality", "unequal" etc.) - penalized 2 times, not penalized 3 times. Holy smokes Batman, it rates better than "x86"!
Statistics, huh ;)
If this submission gets penalized, do we count it as bias against social justice, or junk science