Oops

Making consistently correct decisions is hard. That’s why it’s remarkable that this gov has somehow identified 100% of the correct decisions by doing the opposite every time

Israeli soldiers using sexual assault to force Palestinians out of West Bank, report says Experts say attacks, also carried out by settlers, are leading girls to quit school and enter early marriages

www.theguardian.com/world/2026/a...

Don't be shy to take on a little two-week side project. These five months will be the most precious three years of your academic journey.

With the help of the Sandy Hook families, The Onion has reached a long-awaited deal to take over InfoWars.

We've enlisted the help of @timheidecker.bsky.social, who will be InfoWars' Creative Director.

Please stand by for more.

Mamdani hasn't had time to really think about all that space he now has, because he spends most of his time at City Hall and around New York City. He tries to keep a semblance of his old life by getting around the city on foot, by bike or train. "If you spend every single day driving around in a tinted window security detail, you will have a very specific view of the city," he said. "You actually meet other New Yorkers and you break out of the bubble that so many have come to expect of politics, where politicians only seem to be spending time with other politicians or the people who donated to make them politicians."

I'd say that this applies to anyone traveling in a car in any city. Being in a car versus walking, riding, or taking transit fundamentally changes how you view a place and your relationship to it. I wish more leaders set this kind of example.

www.npr.org/2026/04/16/n...

An image of the article "Desistance": A Multimethod Review of the Literature on Gender Identity Variability in Transgender and Gender Diverse Youth

New publication alert! After four years of analysis, synthesis, and careful writing, I am pleased to announce a brand-new article, “Desistance”: A Multimethod Review of the Literature on Gender Identity Variability in Transgender and Gender Diverse Youth (1) 🧵
psycnet.apa.org/record/2027-...

Life is better on foot! After living somewhere walkable (Toronto), then moving somewhere not walkable (ATL suburb), the latter felt like living on an island, where you have to hop in your boat to the next island over to do anything.

Israel escalates attacks on medics in Lebanon with deadly ‘quadruple tap’ Lebanese health ministry says killing of 91 healthcare workers shows ‘total disregard’ for international law

More deliberate attacks on medical personnel. This is a monstrous state
www.theguardian.com/world/2026/a...

Every online getting company has exactly the data they need to identify people with gambling problems, and the technology and know-how to maximally enable that problem. They also know exactly how to spot the rare skilled gamblers and ban them from playing. Zero economic value. It's basically theft.

Special thanks to my post doc mentor, Candice Monson, and the STRONG STAR Consortium for their support. And to Nicole Pukay-Martin for huge contributions in helping flesh out the analysis universes. (not on bsky)
Data and annotated R code: doi.org/10.17605/OSF.IO/AFHQP

The multiverse isn't a replacement for a strong, a priori analysis plan, & not appropriate for all trials. It's for the (common) situations where flexibility already exists: post hoc analyses, deviations from preregistration, unexpected problems, or genuinely equivalent options.

Figure showing multiverse results for PTSD symptoms during treatment phase

Figure showing multiverse results for fear of intimacy during treatment phase

Across 24 tx-phase universes: PTSD symptom improvement was fully consistent: both groups improved, w/ no group differences. Relationship satisfaction was partially consistent. Fear of intimacy was highly dependent on analysis approach. The multiverse differentiated robust from fragile results

Table 2 from manuscript detailing multiverse dimensions

We demonstrate the approach using a small RCT comparing prolonged exposure vs cognitive behavioral conjoint therapy for PTSD, which had serious real-world challenges: 65% dropout in one arm, only 4 people at 12-month follow-up, and a common challenge of txs differing in session count and length

Table 1 from manuscript detailing multiverse terminology, expanded decision types from Del Giudice & Gangestad (2021), and decision-making guidelines

We expand Del Giudice & Gangestad's (2021) framework to trials and introduce "Type D" decisions, where a best-practice option is itself flawed (e.g., ITT when dropout is severe and differential). All options are flawed, so converging evidence across them is stronger than a single result

The multiverse approach (Steegen et al., 2016) runs all reasonable options simultaneously, crossing every decision to create analysis "universes." Doing so shows which results hold up and which are fragile. It has gained traction in observational research but is almost never used in clinical trials

2/7 Even well-planned trials face decisions w/o a single right answer: handling high or differential dropout, comparing treatments that differ in dosage/length, time function, etc. Picking 1 option is standard, but hides whether conclusions would hold up under a different defensible choice.

Title page and abstract. Abstract text: "Making decisions regarding data processing and analysis are crucial steps toward extracting insights from data in clinical trials. Trial registries like clinicaltrials.gov promote transparency about these decisions and encourage making them in advance. However, clinical studies often face decisions with multiple reasonable options outside the bounds of preregistration, such as when studies conduct post hoc analyses, deviate from preregistered plans, or simply were not preregistered. Additionally, even a priori decisions often have multiple reasonable options from which to choose. Methods that maximize transparency and minimize bias in such situations are needed. This paper advocates for applying a “multiverse” approach to analyzing such data from clinical trials. The multiverse approach simultaneously selects and analyzes the various reasonable options for each decision and presents results across all analysis “universes.” We highlight common challenges and decisions when analyzing clinical trial data, review and expand upon the multiverse approach and show how it can address these challenges, and demonstrate the approach using data from a small randomized psychotherapy trial for posttraumatic stress disorder. In the example presented, results were fully consistent across the multiverse for one outcome (posttraumatic stress symptoms), partially consistent for another (relationship satisfaction), and mostly inconsistent for a third outcome (fear of intimacy). The multiverse approach is a flexible and transparent analysis option for clinical trials in the presence of uncertainty regarding data processing and analytic choices."

1/7 Clinical trials should preregister analyses. But even careful plans leave room for defensible alternatives, and problems often arise that create decisions no one anticipated. Instead of picking just one defensible option, what if you ran them all? New paper in @collabrapsychology.bsky.social

AI seems to be the topic of the year — nearly every conversation I have in my role as academic lead for good research practice touches on it in some way. I’d like to lay out my developing thoughts for conversation and critique. (1/7)

My two most joy-evoking things on the internet these days are the bald subreddit (Reddit.com/r/bald; possibly the most wholesome place on the entire internet) at #1 and this account at #2

Detailed Recommendations for computing the reliable change Index: (1) Use a premade RCI threshold if one has been previously developed from a large sample of a comparable population. (2) If a premade RCI threshold is not available or there is a substantial population mismatch between the premade threshold and one’s study, (a) If N > 100, use the sample estimates of s and rxx to compute the RCI. (b) If N < 30, use estimates of s and rxx from a large-N study that best matches one’s population of study. Measure validation studies are a good default choice. (c) If 30 ≤ N ≤ 100, make a determination based on the study sample size and availability and representativeness of thresholds from a larger sample. (3) Do not use estimates of s and rxx from different samples. (4) Justify the choice of method for the RCI and its specific operationalization, and discuss limitations and biases that may result. Document such choices ahead of time in a study preregistration due to the impact that these choices may have on the strength of apparent evidence for change in a study.

7/7
Recommendations:
@tandfresearch.bsky.social

Figure 1. Percent of samples with [y] or greater deviation from the population value for the reliable change Index: simulated data. The y-axis represents the percent absolute deviation in the Reliable Change Index, and the x-axis represents the percent of samples that have a deviation as large or larger than the y-axis value. Text labels are provided for the worst 10% and 20% of samples, and for the median sample (50%). Example to aid interpretation: for the high reliability/more items measure (top left panel) when N = 10, 10% of samples (x-axis) have a deviation from the population value of 13% (y-axis) or greater, 20% of samples have a deviation of 10% or greater, and 50% of the samples have a deviation of 5% or greater.

6/7
An interesting nuance: Sampling error in the SD and reliability terms were positively correlated, so they partially cancelled each other out. The full RCI was less variable than either component alone. But this offsetting was incomplete and couldn't rescue small samples.

5/7
Avg error was >10% when N < 30, and the risk of excessive error in any single study stayed uncomfortably high until N>100. Also, small errors in reliability estimates sometimes had enormous effects: a .05 difference in reliability at the high end (.95 vs. .90) produced a 41% average deviation.

4/7
We ran Monte Carlo simulations + bootstrap resampling from real clinical datasets (8 measures, 2 datasets) to quantify how much sampling error distorts the RCI at different sample sizes. We used a 10% deviation threshold to flag excessive error.

3/7
The problem: Most studies compute the RCI using their own sample's standard deviation and reliability estimate. Both are subject to sampling error. When the RCI is over- or underestimated, it biases the classification of every individual in that study. We call this the "variable ruler problem."

2/7
The RCI (Jacobson & Truax, 1991) evaluates whether an individual changed enough to rule out measurement error. It's perhaps the most widely used method for evaluating individual change in clinical interventions, and has also been applied in personality, cognitive, and other domains

Title page and abstract. Abstract text: "The reliable change index (RCI) is a tool for evaluating change at the individual level. It compliments standard group-level change estimates and is central to evaluating clinical significance for interventions. In principle, the RCI provides common criteria for evaluating individual change. However, current practices use sample-specific estimates to create these criteria. Because sample estimates are subject to sampling error, these criteria are also subject to sampling error and therefore differ across studies. We illustrate how current practices can lead to differing criteria for reliable change and use simulations to identify the impact of sampling error on the RCI. Excessive error in the RCI began for the average sample when N < 30, and samples only comfortably avoided the risk of excessive error when N > 100. Finally, small errors in estimating a measure’s reliability sometimes had profound effects on the RCI. Recommendations on use of the RCI are provided."

1/7
The Reliable Change Index is supposed to be a common ruler for deciding whether a patient truly improved. But what if that ruler changes size from study to study? Recent paper in the International Journal of Social Research Methodology: doi.org/10.1080/13645579.2025.2585286

Don't get me wrong: I'm relieved that this case is shaping up as either 8-1 or 7-2 against the Trump executive order. But the case is a gift to the Supreme Court. By rejecting an outlandish position, it will earn credibility as apolitical, even as the Overton window moves far to the right.

Trump attends supreme court hearing on birthright citizenship case that could upend rights for millions The US president issued an executive order in 2025 that seeks to undo birthright citizenship, overriding the constitution

Landmark gathering of nation’s leading mathematicians to determine whether president can declare 2+2=5
www.theguardian.com/us-news/2026...

Parkinson's Link to Gut Bacteria Hints at Unexpectedly Simple Treatment Scientists have suspected for some time that the link between our gut and brain plays a role in the onset of Parkinson's disease.

When drawing links between health and the gut microbiome, the obvious explanation that should be ruled out first is always disease—>diet—>microbiome (or M—>diet and disease—>microbiome). Microbiome—>disease is the least likely explanation by a light year
www.sciencealert.com/parkinsons-l...