Open thread.

We can discuss why I want the ruling based on the Spending Clause in comments, but the way I see it: These “transgender” cases are coming hot and heavy now. But cases that could trim executive power in favor of leaving States latitude? Less common.

So what follows is my view on why I’d like to see a ruling based on the Spending Clause.

Title IX is passed under the Spending Clause. If I understand correctly, when a statute is passed under the Spending Clause, federal funding cannot be withheld from a state based on unclear statutory terms; the statute must clearly specify the conduct or condition that could result in funds being withheld. There is an argument that WVa violated Title IX. But, given all the strum and drang over what ‘sex’ means under Title IX and the fact that Congress almost certainly did not mean “gender identity,” one would argue that it is at least possible they did not mean “gender identity.” If they did mean “gender identity”, they could have said so way back when, and they certainly could get together and modify the bill to say so clearly now. So WVa should not be subject to withdrawal of Federal funds for interpreting it to mean “biological sex.” And so their bill does not violate Title IX.

With regard to statutes passed under the Spending Clause, what a ruling on spending would do (among other things) is:

• Preserve state latitude to do things unless Congress writes a bill that clearly withdraws that latitude, preventing funds-withholding coercion based on unclear statutory terms.
• Reduce administrative overreach to interpret statutes as doing more than they clearly say.
• Stop executive flip-flopping that happens when the preferences of the executive branch change — generally due to election of a new President.
• Force Congress to own its decisions, or admit its non-decisions.

That said, if the ruling is purely on the Spending Clause, this does a few things some people might not like:

• Applies to lots of cases outside the “transgender” issue. This could be seismic.
• Allows states like California to pass the law it likes on the transgender issue. After all, if what Title IX says about “sex” is unclear, Cali can’t have its funding yanked either.
• Trims all executives—including the current sitting one—of the power to use an ‘unclear’ statute to carry out their preferred interpretation.
• Makes legislatures have to pass laws about transgender rights instead of trying to twist laws that were not even thinking about the issue when they were passed.
• Leaves the “equal protection” argument alone, for the time being, and leaves defining what “sex” means until later. But while doing so, it forces Congress to be explicit in future laws.

My prediction: If the ruling is based on the Spending Clause, Congress will not modify Title IX to tighten the language. This is because too many Congress critters would be thrown out of office.

For the time being, I am holding my opinion on how they would rule on the “equal protection” argument in Idaho. But I’m pretty sure my feeling is: Equal Protection does not require a carve-out for transgender students in women’s sports. I can discuss that later.

But I like the Spending Clause issue for now. I also don’t think I’ll get exactly what I want. But you can’t have everything.

Note:

• Transcripts to the oral arguments are here: here.
• Google’s summary of “Spending Clause”:
  

  The Spending Clause, found in Article I, Section 8, Clause 1 of the U.S. Constitution, grants Congress the power to levy taxes, duties, imposts, and excises to pay debts, provide for common defense, and promote the general welfare, serving as the foundation for federal taxing and spending authority. This clause allows Congress to attach conditions to federal funds, influencing state policy (e.g., setting drinking ages) as long as conditions are clear, related to the program, and not coercive, as defined by Supreme Court cases like South Dakota v. Dole.

So maybe this shut down is ending.

• Senate moves forward: The Senate has voted 60-40 on a key step toward reopening the government.[…]

• What happens next? There is more to come before the government can reopen. Any one senator can delay consideration of the package for several days, plus the House will have to return and adopt the deal struck in the Senate before it gets sent to President Donald Trump’s desk.

There are, of course, important national issues related to this. But I’m making this all about me: I have a reservation at 10:55 am Tuesday. I dance Wednesday– my guess is I’ll start near 9 am. I dance Silver; Bronze usually dances first. Honestly, they should also schedule the men only first (they don’t have elaborate hair/makeup). But they don’t seem to think of that.

No idea when my hair appointment is.

With luck, money will be flowing out of Federal purses by Monday night.

Open Thread.

The controversy continues. One thing is clear: Sydney Sweeney knows how to engage. She’s a master at not falling into leading-question traps.

NEW: Sydney Sweeney refuses to apologize for the American Eagle jeans ad after GQ's Katherine Stoeffel continuously tried to get her to cave.

Stoeffel: I mean, Trump tweeted about the jeans ad … that just seems like a very crazy moment…

Sweeney: It was surreal.

Stoeffel:… pic.twitter.com/F9ihaeMqeq

— Collin Rugg (@CollinRugg) November 6, 2025

When asked about the U.S. President tweeting the ad, Sweeney picks up the interviewer’s cue of a “crazy moment” and replies simply, “It was surreal.”

The interviewer then frames a follow-up as if suggesting Sweeney might feel grateful that powerful people were on her side: “I wondered if you felt that way.” Sweeney pauses, looks thoughtful, and explains that she hadn’t been thinking about it—she was working 16-hour days and had put her phone away. She then discusses her busy schedule. When asked if she worries people might avoid her movies because of perceived political views, her answer is simply: No.

Throughout this exchange, Sweeney conveys the impression of being unrehearsed. Perhaps she truly hasn’t thought much about Trump tweeting her ad—and perhaps she has. Either way, it reads as authentic.

After that, Sweeney delivers a masterclass in handling leading questions. The interviewer tries to get her to discuss the criticism of the ad—the “genetic” controversy. Importantly, the question is framed as giving Sweeney an opportunity to respond rather than putting her on the spot. Sweeney ignores the leading part and engages the opportunity framing, answering:

I think that when I have an issue that I want to speak about, people will hear.

I have to say: Sweeney either has terrific media instincts, very good handlers, or both. Not delivering political analysis or commentary is precisely the correct move. There are plenty of people whose profession is to deconstruct these things. Defending or criticizing the ad herself would serve her no purpose. For now, she leaves the politics to others.

Her consistent refusal to answer also ensures continued interest from interviewers hoping to be “the one” to get her to engage politically. I predict that if she ever does, it will be in an autobiography decades from now—and entirely on her terms.

The existence of male, female, and true hermaphrodite trees and plants comes to mind when people are discussing “what is sex?”, “what is gender?”, and “do their meanings differ?”. The correct answer is that they do differ. Sex is biological; gender is a role. Today, I’ll focus on biological sex.

So what is biological sex? Let’s start by asking this question:

When biologists discover a new species that reproduces sexually, how do they decide which of the two sexes is male, which is female, and whether, possibly, individuals are both simultaneously?

The answer is that they examine the gametes an individual produces. In sexually reproducing species, there are two types of gametes. Some species are observed to have one larger, nutrient-rich, usually non-motile gamete (i.e., egg) and one smaller, non-nutritive, mobile gamete (e.g., sperm, pollen). These species, with differently sized gametes, are called “anisogamous.” The individuals whose reproductive organs create the smaller gametes only are called males. The ones that create the larger gametes only are called females. Individuals that create both are biological, or true, hermaphrodites.

It may be worth noting that hermaphrodites may be sequential hermaphrodites, or simultaneous hermaphrodites. When a hermaphrodite is sequential, it is referred to as male during periods when it creates the smaller gametes, and female when it creates the larger, nutritive gametes. The identification of its current state of maleness or femaleness depends on the gametes it is currently producing.

Sex in humans

Humans are an anisogamous, sexually reproducing species. As such, biologically, humans are nearly always either male or female. Most humans are male or female, though disorders of sex development and other conditions do occur. Those whose sexual organs are designed to produce sperm are males; those that produce ova are females.

True hermaphroditism in humans — ovotesticular disorder (DSD) — is extremely rare, occurring in roughly 1 in 80,000–100,000 births. Other disorders exist: Androgen Insensitivity Syndrome (AIS), congenital adrenal hyperplasia (CAH), and Klinefelter syndrome (47, XXY). Sometimes, the XYY chromosomal variation occurs.

It is worth noting: even when disorders, conditions, or syndromes occur, the biological sex of the individual is based on the gametes their body produces. Their biological sex does not depend on their phenotype — the appearance of their body or organs. However, since gonads produce gametes, and testes produce sperm while ovaries produce eggs, those with testes are males, and those with ovaries are females.

With respect to disorders: individuals with XXY (Klinefelter syndrome) have testes, which may produce sperm, though function may be impaired. XYY individuals also have testes and, generally, normal sperm production. In all three cases, the bodies are designed to produce sperm, though that production may be impaired. So all of these individuals are biologically male. Individuals with congenital adrenal hyperplasia (CAH) have a typical complement of gonads—ovaries in genetic females (XX) and testes in genetic males (XY).

The only disorder in which an individual is not specifically male or female is ovotesticular disorder (DSD); these individuals are true hermaphrodites.

Assignment at birth

Let me now discuss the notion of “assignment” at birth. What is really happening during “assignment”? What is intended?

At birth, doctors, midwives, and perhaps even the cab drivers who deliver the baby—because there was insufficient time to get to the hospital—will often inspect the newborn and, based on that inspection, decree, likely with some joy, “It’s a boy,” or “It’s a girl.” Along with that, they may also decree, “Ten fingers! Ten toes!” Perhaps, occasionally, whoever is in attendance will be stumped. The ensuing, “Yikes. I can’t tell!” is likely less joyful, as would be “Wow! Twelve toes!”

What is going on here? Well, what’s happening is that the attendants are doing a physical inspection of the phenotype—the outer physical appearance—and making a guess at the baby’s biological sex based on that phenotype. If the baby has a penis and external testes, they diagnose the sex as “boy,” that is, “male.” If the baby has a vulva and no penis, they diagnose the sex as female. Those diagnosing the biological sex usually do so in good faith, and usually do so correctly. However, misdiagnosis can occur, particularly in cases of disorders of sex development or other atypical phenotypes.

After the diagnosis, whoever fills out the birth certificate will check the “sex” box according to their diagnosis: male or female. Some states permit “undetermined,” “X,” or similar. So, they fill out the box to reflect their guess or diagnosis of the sex based on the child’s appearance.

Even in rare cases where underlying gamete potential cannot be diagnosed unambiguously based on external genitalia, the baby’s biological sex — defined by the gametes its body is designed to produce — remains a distinct attribute.

I don’t wish to stray into “gender” too far; I’d like to discuss that in a later post. But it is true that the diagnosis of the child’s sex is one of the things that can affect how parents relate to the child, and later what roles or behavioral expectations parents or others have for the child. Biologically, the baby’s sex exists independently of how the box is checked. To the extent that “gender” is assigned, it occurs socially, through the parents’ or other people’s response to the diagnosis of biological sex. But this does not change the baby’s sex. The baby’s sex is a biological attribute and is not “assigned” by doctors, parents, or society.

In summary: sex is a biological attribute that can be determined by the gametes an individual’s reproductive system creates. Some disorders or syndromes can affect how an individual looks, but their sex remains determined by the gametes they produce. An individual’s sex is not “assigned” at birth; it exists prior to birth and remains unchanged by the diagnosis doctors or others make at birth.

I invite you to discuss this. You may discuss gender also, but I intend a later post on what gender is. Briefly, gender is not biological; it is a social role.

Just google Ian Robert’s and read the stuff. The mind boggles!

Apart from all the criminal stuff….don’t k-12 systems write schools for transcripts? When interviewing for school superintendent? You can have schools send them directly; that’s a reasonably good protection against forgery. Maybe he figured out an appropriate fake name? Enquiring minds want to know.

Supposedly, the school district hired a firm to vette him and

In a statement released last week shortly after Roberts’ arrest by ICE, the district said that Roberts was hired following a vetting process conducted by Ray and Associates, Inc., a national educational executive leadership search firm. The district said that they received all required documentation, including an FBI background check, and that no disqualifying issues were found.

Yeah. I know the government shutdown is the story du jour. Carry on. Open thread.

1. Long time regularly visitors. This will be defined as more than 10 approved posts. (I may later do and more than 3 in the past year. It’s limiting at all that is the issue.
2. Me on the “front end”. That’s what you see here.
3. Me on the “back end” where I can respond inside comments in blue. This will allow me to occassionaly make sure an answer is near a question. (Like “how do I do ‘x’?” could have the answer inserted immediately after it, and it could be clear mine was the answer.) Also, I could fix people’s incorrect html!
   It will also allow me to include responses to “you know whose” comments inside his comment when I feel like releasing it which would never be immediately. That way, I don’t need to quote and respond and derail the thread.
   It’s actually really easy for me to get this feature going if I only allow me to do it. But it makes all of you want me to let you add color too. So I want to do that.
4. Greek letters are going to be a problem. Some ways of entering them are easy at the screen– but the database won’t store them. I am not screwing with the database for maintainability reasons. (WordPress updates regularly. Sometimes that involves automatic changes to the database. I would need to detect the problem and then go in and “fix” it when this occurred at whatever random times WP updates.)
   I do have a latex plugin activate– but it requires me to add latex tags to the post. When those are present, we’ll all be able to enter Greek letters in Latex. This post does not have the tags entered. If you know unicode, you might be able to succeed with that, but it’s cumbersome. Finding the letter elsewhere and cutting and pasting will probably not work.

If you try putting in right now it will probably not work for you. I’ll let you know when it might because I’ll want testing. I am watching what “hooks” get activated as people do comment, and in what order. ChatGPT is certainly confused. But I don’t need you to bother trying to put the fonts in for the time being. I’ve just got error logs that show what is “triggering”. Merely commenting normally leaves some breadcrumbs.

Hopefully, I’ll get these powers in. As usual: open thread.

In this post, I’m going to quote from the (currently reserved) comment in which he describes a general statistical test. He applied the method to hypothetical data. I’ll also be correcting some of the numerical errors that crept into his analysis of that hypothetical data. I will also be releasing that comment and placing it in the comments thread for this post so you can see the full comment.

On to Joshua’s comment! I’ll skip the first bit as irrelevant to the proposed statistical test and start with this.

Joshua’s method extracted from comment

Oh, and it’s cute that you’d want to take the time to show me a test, but really don’t waste your time. I can just ask Grok to run some numbers for me:

Joshua then post Grok’s answer to a question Joshua asked; the question itself is not. So we will call the entity responding “Joshua-Grok”. Content from Joshua’s comment is in blockquotes. I will break that up and add my own observations.

Joshua-Grok’s answer begins:

Expected Proportion Under the Null

If trans people make up 1% of the population, we expect 1% of shooters to be trans by chance. For a sample of n shooters, the expected number of trans shooters is n × 0.01.” .

That is indeed what we expect if individual trans individuals become shooters at the same rate as non-trans individuals.

2. Binomial Test Setup

Let n be the number of school/church shooters in the sample (e.g., 50 or 100 over a decade).
Let k be the number of trans shooters observed.

The probability of observing k or more trans shooters by chance, given a population proportion of 0.01, is calculated using the binomial distribution:P(X ? k) = 1 – P(X < k), where X follows a binomial distribution with parameters n and p = 0.01.

This would seem to tells us something about the situation Joshua wanted Joshua-Grok to analyze. It seems to be school or church shooters (as opposed to other possible shooters.) Joshua wants Grok to assume 1% of the general population is trans. The variables ‘n’ and ‘k’ are the total number of shooters and total number of trans shooters respectively.

I’ll now tease out his notation so we agree what it is. I’ll also translate this into R commands. This will allow those familiar with R to repeat the calculations.

1) The notation P(X ? k) contains a ? which likely means WordPress stripped something. Based on the wording “P(X ? k)”, we can infer P(X ? k) must be P(X ? k). That is the probability that “k or more” will be observed out of whatever sample size “n” is later used and single probability “p” are used. Joshua-Grok says it will use p=0.01.

2) Based on the above, P(X < k) must mean "the probability of observing fewer than k shooters" given the parameters which will be n shooters, with the probability that any individual shooter is trans is 0.01". Joshua appears to be rounding to three decimal places. So if n=50 and the population of trans is 0.01 in R P(X < k) would be calculated as P(X < k) = round(pbinom(k-1,50,0.01),3); The reason for k-1 is that, pbinom calculates "the probability of observing k or fewer", Joshua is saying just "fewer". Joshua-Grok continues:

We want the smallest k where P(X ? k) < 0.05, indicating statistical significance.

This statement tells us Joshua-Grok has picked a single sided test. He doesn’t say why he picked a single sided test, but this is his prerogative. Some might have picked a two sided test. He’s also picked a confidence level of 5%. That’s also his prerogative and pretty conventional.

Joshua-Grok now starts to calculate.

3. Calculating the Threshold

Let’s test for different sample sizes (n) to find the minimum number of trans shooters (k) that would reject the null hypothesis at p < 0.05.Sample Size: n = 50 (e.g., 5 shootings/year × 10 years):

Okie dokie! Let’s do that!

Expected number of trans shooters: 50 × 0.01 = 0.5.

Yes.

Using a binomial calculator (or cumulative distribution function), we compute P(X ? k) for increasing k until p < 0.05:

For k = 2: P(X ? 2) ? 0.264 (not significant).

For k = 3: P(X ? 3) ? 0.080 (not significant).

For k = 4: P(X ? 3) ? 0.019 (significant at p < 0.05).

Proportion: 4/50 = 8%. So, if 8% or more of shooters are trans, the result is statistically significant.

If P(X ? 2) means what Joshua says it means “The probability of observing k or more trans shooters by chance”– the correct results for the calculation above should be:

> round(1-pbinom(c(2,3,4)-1,50,0.01),3)
[1] 0.089 0.014 0.002

We can now replace Joshua’s numbers with the correct values and also correect the declarations of significance.

Proportion: 3/50 = 6%. So, if 6% or more of shooters are trans, the result is statistically significant.

But maybe some of you don’t trust my R. So I can also ask Grok:

Thanks! I’m going to divert a little to something “Joshua” claimed. He was interested in a coin that got 0.01 heads in a single flip. He wanted to flip it 50 times. And he wanted the probability of getting the following three things:

“k or more” heads, for k values of 2, 3 and 4.
Can you give me those values? (You need not explain.)

Initially, Lucia-Grok gave me ~ 0.091, 0.014, and 0.002. Those did not exactly match my values from R, but they were close. I told Grok what I got using R and asked it if it had approximated. Grok gave me a long winded answer which ended with “The correct probabilities, using the binomial distribution, are: [..] 0.089 [..] 0.014 [..]. 0.002 . These match your R output.”

Back to Joshua-Grok which repeats the analysis, but this time with a different number of shooters. The new sample size is n=100.

Sample Size: n = 100 (e.g., 10 shootings/year × 10 years):Expected number: 100 × 0.01 = 1.

Testing values of k:

For k = 4: P(X ? 4) ? 0.189 (not significant).<\li>

For k = 5: P(X ? 5) ? 0.081 (not significant).<\li>

For k = 6: P(X ? 6) ? 0.029 (significant).<\li>

How do these match what I get with R?
> round(1-pbinom(c(4,5,6)-1,100,0.01),3)
[1] 0.018 0.003 0.001

Correcting Joshua’s text we get:

Sample Size: n = 100 (e.g., 10 shootings/year × 10 years):Expected number: 100 × 0.01 = 1.
Testing values of k:
For k = 4: P(X ? 4) ? 0.018 (significant).
For k = 5: P(X ? 5) ? 0.003 (significant).
For k = 6: P(X ? 6) ? 0.001 (significant).

As before, I asked Grok to estimate.

Joshua next moved on to a different claim . He now wanted to flip the same coin 100 times. He wanted the probability of k or more heads for k values of 4, 5 and 6. Can you give me those?

As before, Grok first used a Poisson estimate and returned (0.015, 0.003, and 0.001).

So I asked “Did you do this exactly? Or are they approximated?” It admitted to approximating and now returned these numbers as correct. (0.018, 0.003, and 0.001). These match my R values– which is not surprising because Grok hooked in R to do the calculation. Grok even gives the code!

Back to Grok answering Joshua:

Proportion: 6/100 = 6%. If 6% or more of shooters are trans, the result is significant.

Well… except that 6 out of 100 was not the threshold for significance. From the following we can see the threshold for significance is 4. So the correct proportion is 4%.

Joshua-Grok now repeats but with n= 20 shooters!

Sample Size: n = 20 (e.g., 2 shootings/year × 10 years): Expected number: 20 × 0.01 = 0.2.

Testing values:For k = 2: P(X ? 2) ? 0.017 (significant).

Grok says 0.017. Let’s check:

> round(1-pbinom(c(2)-1,20,0.01),3)
[1] 0.017

Hey! The Joshua-Grok got a number right!!!!

Joshua-Grok then generalizes its results. But bear in mind, Joshua-Grok’s conclusions for n=50 and n=100 are based on incorrect results for the cumulative probability. So these generalizations are based on wrong numbers for those two cases. Neverthless, quoting it:

4. Generalizing the Threshold

The required percentage decreases as the sample size increases because larger samples provide more statistical power. Roughly:

For n = 20, you’d need ~10% (2 trans shooters) to reach significance.

For n = 50, you’d need ~8% (4 trans shooters).

For n = 100, you’d need ~6% (6 trans shooters).

If we use the correct numbers the above becomes:

The required percentage decreases as the sample size increases because larger samples provide more statistical power. Roughly:

Back to Joshua-Grok who continues:

However, these percentages are sensitive to the sample size and the true population proportion. If the trans population is 0.5% instead of 1%, the threshold percentages would be slightly higher (e.g., ~10% for n = 50), and if it’s 2%, they’d be lower (e.g., ~6% for n = 50).

Bold mine.

Of course the percentages of shooters that would make us reject the hypothesis that trans become mass murders at the same rate as non-trans are sensitive to the number of mass shootings observed in a period and true population proportion of trans which may not be 1%.

But, before we debate if the proportion of trans in the general population is doing, let’s discuss what Joshua-Groke claims. Focus on the part I made bold. As a generalization, something is backwards. Because if the proportion trans in the general population is lower than 1% the threshold percentage of shooters required to reach significance is lower than if the percentage of trans is 1%. You don’t even need to do the numbers to know this. After all, if there were 0.00001% trans in the general population, you’d marvel if 2 out of the 50 randomly draws of people were trans! Likewise if the proportion of trans is higher, the threshold value to reject the proposition they commit mass murder at the same rate is higher. Say 50% were trans. You would be unsurprised if 50% of the shooters were trans. That’s about what you would expect!

So what are the right numbers for 0.005% trans?

For the 0.5% trans, the threshold for 50 is 2, and 2/50 is 4% not 10%. For 2% trans the threshold is 4, and 4/50 is 8%. Here’s the R.

> round(1-pbinom(c(1,2,3,4)-1,50,0.005),3)
[1] 0.222 0.026 0.002 0.000

> round(1-pbinom(c(1,2,3,4)-1,50,0.02),3)
[1] 0.636 0.264 0.078 0.018

Ok… so I’ve also been replicating with Grok. Naturally, I asked Grok. What does Grok say? Now things get a little interesting.

Turns out Grok makes mistakes!

Suppose the probability of heads on a single toss of a coin is 0.5%. If we toss it 50 times, what is the probability that we would get “k” or more heads. Please do this for k=c(1,2,and 3).

Grok came back formatted nicely but basically told me c(0.221, 0.080 and 0.013). It should be evident these do not match c(0.222 0.026 0.002). Grok conveniently also gave me R code to run for myself.

I ran the code Grok gave me and told it the R code it supplies gives c(0.222 0.026 0.002). Those match my final numbers, not Grok’s.

Grok redoes stuff, tells me his numbers and accused me of having the wrong value for “p” or “n” stored. His code snippet provided both values! So, as if! So, I give Grok simpler code with “p” and “n” hard wired in. Once again, R returns my previous numbers. I tell Grok Grok still decides it’s right!! It does post details of what it does. I read that long, long long discussion and tell Grok,

Hmmm…. for P(X=1) you do P(X=1)=50*0.005*0.995^49. Then you say that’s 0.14095.

My R says
> 50*0.005*0.995^49
[1] 0.1955559

It took a little more bullying on my part, but finally Grok found it’s mistake. It had computed 50*0.005*0.78271435 = 0.1409514, which is wrong. Feel free to check. Grok had stored that as a result and re-used that number for subsequent calculations.

Interestingly, this is an incredibly unsophisticated mistake. Grok was able to do all sorts of nifty binomials, explain and so on. And then, just like any human, it is capable of the equivalent of a transcription error. Once it makes that error, and “writes that number down”, it just steam rolls ahead with the wrong number.

Once it corrected its mistake, Grok replicates my numbers within rounding.

Your output is essentially correct, and the differences are due to R’s rounding behavior (e.g., 0.221199217 rounds to 0.222, 0.001158029 rounds to 0.002). My previous calculations for k = 2, 3 were incorrect because I used an erroneous P(X = 1) ~~ 0.1409514 , likely from a miscalculation of (0.995)^49 or a typo in my manual process.

So, Grok can do great stuff. It can also go wrong and really stick to that. You have to check things with R!

Returning from the Grok makes mistakes interlude

Yes, the results of the threshold value (whether by number of %) depends on the sample size and the probability– in the trans case, the proportion of trans in the population. Estimating either are actually very contentions matters; but Joshua tells me the proportion of trans is “(roughly 0.5-1.5%, per Gallup and Williams Institute estimates”. Well, that’s a factor of 3. And is that range even wide enough? Who knows?

As I previously wrote as I previously wrote when discussing analyzing data on the issue of whether or not trans individuals were more or less likely to commit mass murders of some sort, ‘Most of us actually touched on much of the “main” problem in any statistical analysis which is we simply don’t know numbers. There are some others– including cherry picking and also… I don’t know it’s name. It’s not the texas sharp shooter. It’s…. something to do with picking the observation and then only doing the statistic because you saw that thing. ’ I guess that’s called “post hoc theorizing”. Wikipedia says post hoc theorizing not so bad as long as you recognize and admit you are doing it and don’t pretend the analysis is based on a pre-planned experiment.

But at least now Joshua has suggested a “method”. I’m not sure he’ll continue to “like” it now that it gives answers he probably doesn’t like. And he’s suggested a range for p. So, there is that. We can now apply the Joshua-Grok method (with correct numerical results) to other problems. Of course, we can also criticize the method. I mean… “Why a one sided test?”

The results will be posted– but the link won’t be available for a while.

Tomorrow some people will be dancing, but I’ll be flying down to “the house of the mouse”.

Open Thread.

You may wonder: were we competing against the other people in this video? Nope. The two other am-am couples who competed in Senior Bronze Am-Am danced C1 so they aren’t in this heat. I see a younger am dancer dancing with her pro and I see a Sr. am gentleman dancing with his pro. You might wonder how I know which leads are ams and which pros? Leads wear the number. If it’s higher than 300 the lead is an am. Otherwise, they are a pro.

The gentleman wearing 367 is the current points leader in the WDSSeries Pro-Am something or other. That’s mostly a participation award. He’s at all the competitions and dances everything.

Lots of things are organized to facilitate “bragging” rights, provided the winners don’t reveal all. There are several “series” you can enter. The “series” that isn’t mostly a participation award is called “Best of the Best” series. You can’t really get those just by entering a lot of competitions. I have never won that. But we did win the am-am championship and I have won other championships. I’ve even won when we were contested. ( You can be placed 2nd out of 1 couple! And the judges will mark you 2 out of 1! It’s not just theoretical. )

Looking at it myself, I need to get “snappier”. I’m not sure how to do that since you aren’t supposed to lift your feet and “jump”! So you can’t get snappier that way. I already knew my styling is deficient. If you watch Mambo, you’ll see we got off time near 5:38. We stopped and restarted in this heat. We danced “on 2” often enough the judges didn’t ding us for that. We danced on time in our other heat.

Open thread.

I tend to think the outcome and reasoning of Skremetti leans in favor of SCOTUS deciding this is the domain of the legislature and then can ban transgender girls from cisgender girls sports. But this also hinges on interpretation of Title IX. Are regulations in Title IX relevent to Gender vs Sex in Title IX? And how? We’ll see.

As usual: Open thread.

U.S. ATTACKS IRAN: President Donald Trump posted on Truth Social on Saturday that the U.S. “completed a successful attack.”

Open thread.

I’ve never used Go Fund Me, so I gave a modest amount. (Turns out you can zero out the ‘tip’ and you can link to your bank to minimize the hit of fees.) But if you want to read, her letter is here. The link below is direct to the donation page. Otherwise, open thread.

P.S. Did Jim and I win our multi? Yes. We did.

Now some of you might wonder? How many people compete Am-Am? Answer: Very few. Very few. I’d sort of expected that a local Am-Am couple our age, the McBrides, would compete. They are competing a lot. However, one of the couple’s goals is to win the WDSS and the DCT series “top student couple” awards for the year. These require accumulating points in competitions that participate in those series. To get the greatest bang for their bucks they are only entering competitions that participate in both series. The upcoming competition participates in WDSS but not DCT! So, no McBrides to compete against.

Jim and I will learn if we are contested approximately 48 hours before the event! If we aren’t…. well, you can come in 2nd out of 1. So…. we still might not win. (In case you are wondering: I think you can’t come in 3rd out of 2. )

Open thread.

Facebook stories being Facebook stories, details were sparse. Also, the source is mostly her Dad and so things are worded in a way that is rather favorable to Becky. Among other things he characterizes her situation as a “visa mixup”. The story as told in Facebook had me scratching my head a little. One question was “Why would a British citizen traveling as a tourist be refused entry into Canada?” British citizens don’t need visas when traveling as tourists to Canada. What could the “mixup” possibly be?

I had to google to learn more. Becky evidently traveled to the US, was here a while, then she tried to enter Canada. That’s where things first went wrong. From the BBC.

“However, when she reached the border, Mr Burke said the Canadian authorities denied her entry as they were concerned she may try to work illegally.”

When the Canadians escorted her back to the US, the American authorities agree with the Canadian’s interpretation that she has been working and plans to work while traveling through the US and Canada. So, the American’s weren’t letting her stay in the US without a work visa either. But unlike Canada, the US couldn’t just refuse her entry because she’s already here. They put her in detention.

Now you might as “Was she working”? After all, if she wasn’t working, this is horrible unfair. She evidently did housework for host families while they provided her with accommodations. Her trip was organized through a program called “Workaway”, which sort of implies that the program organizing it know it’s considered “work” in some sense the word. Becky’s family evidently considers this to be some sort of “cultural exchange”. Maybe it is. Working as an Au Pair is considered “cultural exchange.” Doing so requires a work visa.

The reaction from Workaway to her plight is:

A spokesperson for the company said it was “very sad” to hear about what had happened but that it was “the responsibility of the host and person travelling to make sure they are within the law, have the correct visa or permit and insurance in place before travelling”.

[..]

“Unfortunately if these warning are not followed and someone attempts to enter the United States without the correct permissions or visas in place then the US border control are within their rights to refuse entry.”

Anyway, Becky is now detained in the US and is evidently unhappy about the conditions in detention. A Daily Mail article describes the food and other conditions:

‘She is sharing a cell, surviving on a diet of cold rice, potatoes, and beans (she is vegan), and has limited access to phone calls,’ [her father] wrote.

‘Visitors are restricted to speaking through a glass screen via telephone.

‘All her possessions have been confiscated, and she feels isolated and desperate to come home.’

I guess you don’t get some veggies in detention? I would have expected the dorm cafeteria classic “mixed vegetables” would be served. I’d guess she can’t surf the internet either. I’m sure I would not enjoy being in detention either. I’d want to get out.

How could that be accomplished?

Melissa Chavin, a US immigration lawyer based in London, said Ms Burke could be released either through voluntary departure or expedited removal, but the process of appearing in front of a judge as part of voluntary departure could be delayed because judges had been fired by the Trump administration.

Voluntary departure is experiencing big delays due to judges being fired or resigning. Ok… that’s slow. Potentially it is very, very slow. Slower than normal. Because Trump.

What about this “expedited removal” thing? As far as I can tell, expedited removal needs only a low levels administrator and can be supposedly be done in one day. Many articles describing it seem to thing the swiftness of the process is a bad thing because it doesn’t give people seeking asylum time to make their case. But I’m under the impression Cheryl Burke is not seeking asylum for persecution in the UK. So I would think she might jump at this chance. And Trump has evidently expanded the expedited removal process.

The 2nd Trump administration, however, has expanded it to the maximum, to apply to foreign nationals encountered anywhere in the United States who have been continuously present in the United States for less than two years.

She seems to have been here less than two years.

The new rule appears to be described here. As far as I can tell, it could apply to her case. I don’t know why immigration officials did not apply this process to her– the articles don’t say. It sounds like if it was applied, she would no longer be in the detention center. That would seem to be a nice thing.

But also, you evidently can’t ask for expedited removal. Google’s nifty AI thing currently says

No, you cannot request expedited removal; it is a process initiated solely by immigration authorities against individuals they believe entered the United States illegally and are subject to expedited removal procedures, meaning you cannot ask to be removed quickly through this process; if you are facing deportation, you should consult an immigration attorney to understand your options.

In any case, two downsides of expedited removal is the person removed by this process can not enter the US for 5 years or possibly longer and there is an order of removal put on your record.

A British Member of Parliament (MP) is evidently involved in some way and FCDO (Foreign, Commonwealth & Development Office) are evidently in contact with local authorities. Even though technically you can’t ask for expedited removal, I would think the FCDO might suggest it to an official who would then say, “Hey! Yeah. If I decide to do this, she’ll be out quickly! Problem solved!”

Speculating, perhaps the British MP or FCDO is asking to not use expedited removal? Dunno. Anyway, I thought people might find this story interesting. It’s a bit out of the ordinary relative to other immigration stories.

As usual, this is an open thread. Talk about whatever you like. (I’ll leave comments on the other thread open for a bit since perhaps we’ll want ones on immigration near the top of comments.)