I thought you were arguing against the relative importance of looking at the numbers.
I completely agree with you that just because the numbers are small and may run into problems with statistics that trans people should not be discarded as a rounding error.
The ease at which I recall one can show a statistically significant increase in suicide rates of trans people really emphasizes the need to not ignore the issue even if the group you are studying is too small to have a statistically significant number of trans individuals.
As to the primary analysis - that of bias is media I donāt think she will have any issue on sample sizes.
I donāt think itās unimportant, just that itās not the only thing that matters. Iām just sick of being told that itās the ONLY thing that matters.
Even if we were to assign rights based upon population size - the pop of trans people is estimated at .6% - this is much larger than many other populations. Jewish people are .2% of world population- Armenians less.
Thereās a long list of ethnicities and characteristics smaller than the trans population- we donāt seem to be making that argument elsewhere.
Letās dumb it down for those of us whose eyes glaze over when presented with math problems (like me)ā¦
Consider the issue like a rope bridge over a steep chasm, and every makeup of person represents a section of ropeā¦LGBTQ+, cis/hetero, white, black, tan, Muslim, you-name-it.
If weāre not out there fixing frayed rope, i.e. ā supporting even the least represented sections of the whole, then we all go tumbling into the river.
Great work on the data visualization, Andrea. Also, very interested in: How can āabsence of dataā be handled in the visualization? Example: how are transgender individuals who donāt self-identify as trans shown in data? (a significant percent of transgender people merge into gender binary). i.e., could The Atlantic already have several transgender journalists on staff, but nobody knows who they are?
The headcount aspect of the dataviz has to be independently verifiable, so if thereās some way in the future to verify they have staffers who are not public about their identities, that could happen. For now, that canāt happen. And yes, ātransā as a binary and even an identity will go away in the future.
Has the notion of āindividual privacyā in medical or personal identity been turned into a quaint notion? The wholesale looting of private data in our social media, medical, and government records probably made this a rhetorical question already
Most targeted groups canāt hide the characteristics that make them the targeted group in the first place (more importantly. they shouldnāt have to).
Further, you seem to presuppose that āhidingā oneās personal identity is done as a free choice, as opposed to out of necessity or fear.
The comment was more about the data privacy issue.
Should private data be utilized at all?
Ethically, should a researcher sacrifice someoneās personal data, in an effort to achieve the greater good of eliminating or exposing hate or bias, in society or a specific company or organization?
Could the quest for gathering such data end up doing more harm than good?
Or, are there ways to ethically study the subject without the study itself doing harm or modifying the subject of the study?
That is a pretty good summary of the whole p<0.05 thing, but I will say again, I do not think it is terribly applicable to this sort of study. I am not fluent enough in statistics to guess what might be, though. Certainly the lack of trans hires over a period of time is enough to raise the question, hence the original premise. Can it be measured? Of course it can. Will it be measured? Well, from the sounds of it, it just might. Looking forward to it!
* And not to reduce the many layers of Ioannidisā argument to the size of p-values, thereās more but thatās an important component when we are using statistical inference to make decisions.
It could all start with a simple and innocuous phone call from a researcher to the corporate spokesperson of a company. It might go something like thisā¦
Researcher: āHow many transgender employees do you have?ā
Corporate Spokesperson: āIām not sure of that number, Iāll have to get back to you.ā
Researcher: āThanks.ā
2 months later:
Corporate Spokesperson: āWe added your transgender question to our annual employee review and new employee questionnaire. We ask every employee āAre you or have you ever been transgender?ā, so we should have some new data for you within a few months.ā
Researcher: āBut, butā¦ā
Corporate Spokesperson: āDonāt worry, we are also doing DNA tests on everyone, so we can verify that the data is correct.ā
Researcher: āPlease, thatās not what I wanted know.ā
Corporate Spokesperson: āIt isnāt? Well, Iāll send you the database of all our transgender employees, so you can pick which data you want from that.ā
This isnāt any different than any other EEO data that employers collect. Providing that information is voluntary on the part of the employee.
ā 1. Why is Stanford University asking employees to designate their veteran and/or disability status online in Axess ?
Stanford University is a federal government contractor and as a recipient of federal funding (funding for research, student loans, etc.), Stanford is required to collect and report information on its applicants and employees to assess its affirmative action efforts.ā
āWe are an Affirmative Action, Equal Opportunity Employer. Our employment decisions are made without regard to race, color, religion, gender, national origin, age, disability, marital status, veteran or military status, or any other legally protected status. The purpose of this Employee EEO Self-Identification Form is to comply with federal government record-keeping and reporting requirements.Periodicreports are made to the government on the followinginformation.The data you provide on this form will be kept confidential and used solely for analytical and reporting requirement purposes. This form is processed and maintained separately from your personnel file and is not used to make decisions about the terms and conditions of employment. Completion of this form is optional and voluntary. We appreciate your assistance.
ā
Also - the notion that employers donāt know this for most trans employees with todayās background checks is quaint.
ā¦ the constant vilification of trans people in parts of the British media is a disgrace. We should all be mortified by how devious hate groups with anti-trans voices that base their poorly constructed arguments on bigotry and rightwing ideology have been presented. Parts of the British media have allowed themselves to be infected by hate and swayed by the misguided concerns of a small but vocal group of people.
The comparison between the discussion around trans rights in Iceland vs. the UK is a real eye opener to how things could be if projects like this one - and work that builds on them - are successful.
Before a question can be answered, or even analyzed, there has to be someone willing to ask it. In areas like this, that can be the scariest place to be, asking a question TPTB do not really want to have answered. More power (and money) to @AndreaJames project, and I look forward to seeing where it goes.
