Lightnews — Scholar-powered news

DB Krupp

@dbkrupp.bsky.social

930 followers 310 following 200 posts

Working on social evolution theory, kinship, inequality, and competition. Probably not the interdisciplinarian you’re looking for. https://www.saltlab.org

Posts Replies Media Videos

DB Krupp

@dbkrupp.bsky.social

Also willing to accept, given how often the words "class and "classy" show up, that RT doesn't have the best verification process for its Verified Audience reviews. So maybe also "today in falsification bias"? 🤷

Screenshot from Rotten Tomatoes of reviews from two "Verified" audience members. "Deborah T" gives five stars and writes "Excellent movie, well done throughly enjoyable. Such a pleasure to see real class return to the White House". Kelly S also gives five stars and writes "Classy movie. Loved every minute of it!"

February 1, 2026 at 5:08 PM

DB Krupp

@dbkrupp.bsky.social

Today in selection bias.

Screenshot from Rotten Tomatoes of the poster, Tomatometer score, and Popcornmeter score for the new "Melania" documentary. The poster, in black and white, shows Melania Trump seated in a chair and with her body facing to the viewer's left and her head facing the viewer. The Tomatometer, with 20 critics' reviews, has a score of 10%. The Popcornmeter, with 500+ Verified Ratings, has a score of 99%.

February 1, 2026 at 5:00 PM

DB Krupp

@dbkrupp.bsky.social

Any time I see a study about wings, I think:

Survivorship bias airplane image. Illustration of a plane with bullet holes, but the planes documented were only the ones that returned to base.

January 5, 2026 at 12:37 PM

DB Krupp

@dbkrupp.bsky.social

Two charts. On the left is a bar chart showing 44,100 civilian deaths and 8,900 militant deaths. On the right is a pie chart, showing 83% civilian deaths and 17% militant deaths. Source: internal Israeli intelligence database.

August 21, 2025 at 8:13 PM

DB Krupp

@dbkrupp.bsky.social

Rick & Morty, Season 8, Episode 2 (2025)

Donald Trump post on Truth Social: "There is no #JoeBiden - executed in 2020. #Biden clones doubles & robotic engineered soulless mindless entities are what you see.
>#Democrats dont know the difference."

June 2, 2025 at 5:58 PM

DB Krupp

@dbkrupp.bsky.social

You may want to check out Martin Daly's 2016 book, Killing the Competition, where he speaks to this argument in a few spots. Here are a few relevant pages

Text from Daly (2016), Killing the Competition: "Cohort Size and Homicide

"If there were no baby booms and busts—if birth cohorts didn’t vary
in size or if they steadily tracked the population growth rate—and if
age-specific mortality patterns were stable, too, then a fixed age struc-
ture would persist for decades. However, birth rates do in fact go up and
down, and age-specific mortality patterns do change. The result is that
the age structure of populations can change rapidly, substantially, and
consequentially. In 1960, 13.2 percent of the population of the United
States consisted of men between the ages of fifteen and thirty-four.
But as the post-war baby boomers became teenagers and young adults,
things shifted, and by 1980, that same demographic group of men 15–34
constituted 17.5 percent of the populace. Their proportionate share of
the total population had grown by over 30 percent in just twenty years!
And then it began to fall again. Figure 8.2 portrays these changes."

Below this text is a line graph of the percent US population males age 15–34 as a function of year, running from 1960 until 2015. The line rises from 1960 until 1980 from ~13% to ~17.5% and then declines thereafter, returning to ~14% in 2015.

Text continues: "Does this curve looks familiar? If so, that may be because it resem-
bles the trajectory of the US homicide rate over the same period, as
portrayed in figure 8.1. And could that be more than just a coincidence?
Of course it could! This is, after all, the demographic group that is
most likely to kill, so it would be strange if the gross homicide rate
did not depend, in part, on young men’s share of the total populace.
And in fact, there is a lot of evidence that the proportion consisting
of young men indeed makes a difference, both in cross-sectional and
in longitudinal analyses. In Chapter 2, for example, you saw that there
was a strong positive relationship between income inequality and the
homicide rate in 1990, comparing across the fifty US states, so strong
that the Gini index could account for half of the variance in homicide.
My colleagues and I then looked at a variety of state-level variables
to see if any of them could account for a significant proportion of the
remaining variance that had not been “explained” by inequality, and we
were able to identify only one: the proportion of a state’s population
that consisted of young men.

"There is, however, a sense in which I have cheated a little in how
I designed figure 8.2. By portraying the range of the young male per-
centage from only 12 through 18, I made the changes look bigger than
if I had instead drawn it with a “true zero.” With the range extending
from 0 to 18, the rise and fall would look much less impressive, and
would not be as visually reminiscent of the homicide curve. This issue
of alternative data portrayals may sound like a mere matter of optics
or aesthetics, but there is an important substantive question lurking
behind it. The US homicide rate doubled between 1963 and 1973,
bounced around for a while, and then fell back to that 1963 level fifty
years later. How much of that roller-coaster ride can be attributed to
change in the population’s age structure?"

Text continues: "A popular answer is “some, but not much.” In 2004, Steven Levitt,
the economist whose best-selling book Freakonomics made him a pop
guru, wrote a professional article in the Journal of Economic Perspectives
entitled “Understanding why crime fell in the 1990s: four factors that
explain the decline and six that do not.” Changing demographics, Levitt
concluded, was one of the “six that do not,” although he conceded that
“overall changes in the age distribution may have reduced homicide and
violent crime by a few percent and property crime by as much as 5–6
percent.” This is an old argument, and although Levitt’s version is more
sophisticated than some earlier ones, it is a specious argument, predi-
cated on the assumption that a “demographic explanation” is necessarily"

Text continues: "one that attributes fixed tendencies to each demographic group and then formulaically generates an expected magnitude of change from the various groups’ new proportionate representations. The real world is more subtle than that. As Margo Wilson and I complained in 1988, with respect to an essentially similar dismissal of demographic change:
'Violence is not simply a “trait” of individuals, but a product of social
interactions. The interpersonal violence of young men is a manifesta-
tion of competitiveness; if the proportional representation of young
men in a population increases, might not the level of competition
experienced by all men increase? Moreover, if the proportion of
men who are in a violence-prone age-class doubles (for example),
then even by a simplistic random model, the proportion of all dyadic
encounters that will involve two such men will quadruple; might we
not expect a greater than linear increase in violence with each incre-
ment in the numbers of violence-prone persons? And how does the
number of men in other age classes affect the perceived marital and
economic prospects of young men, and hence their risk-prone, violent
tendencies? Our point is simply that “demography alone” actually
“explains” nothing without some individual-level specification of
how demographic and other effects upon violence are mediated.'

"Much as we might wish to do so, we simply cannot finesse our way
around the big social psychological and developmental issues: What
aspects of their social environments are people sensitive to, exactly, and
how exactly do they respond, both immediately and developmentally?
A complete understanding of the effects of cohorts and population age
structures will require answers to these questions, as will a complete
understanding of the effects of inequality."

June 1, 2025 at 12:48 AM

DB Krupp

@dbkrupp.bsky.social

I think you’ve been spoiling for this fight for a while now 😆

Skeet from @stephenjwild.bsky.social from September 21, 2023: “Bluesky seems like a good place for synthetic Marx.”

May 28, 2025 at 1:14 PM

DB Krupp

@dbkrupp.bsky.social

I was also unlucky, twice over. SSHRC gave me a postdoctoral fellowship and then a grant. But, because one cannot be paid out of and simultaneously hold federal funds, they made me pay back 5 months of my salary. I also entered the job market in the immediate aftermath of the Great Recession. 5/9

A selection of text from the post, which reads: “I led research projects of diverse kinds, collaborating with biologists, linguists, psychologists and anthropologists. This fuelled promotions in rank and salary, a change of institutions and more funding, in particular a Future Research Leaders Grant from UKRI (UK government's main funding body). But even here I made a critical mistake: to be unlucky with my timing. I won that award the only year UKRI did not compel the hosting university to offer a permanent post. So I had a good thing for a few more years but nothing more.”

May 25, 2025 at 8:13 PM

DB Krupp

@dbkrupp.bsky.social

I had wonderful supervisors, but I also encountered terrible bullies. One tried to extort me to publish more favorable results, so I dumped the paper. Two others tore up a proposal to keep me on contract, even though I brought in more money in funding than it would have cost to have me stay. 4/9

A selection of text from the post, which reads: “The years after a PhD are demoralisingly unstable. You have invested years in training for no job security. There are far, far fewer good jobs than good candidates. You are on the wrong side of power asymmetries and some senior people are exploitative (one example). Applications for grants and fellowships take many months to be assessed and have low success rates. It is effectively compulsory to change locations for temporary posts of a year or three”

May 25, 2025 at 8:13 PM

DB Krupp

@dbkrupp.bsky.social

Like Thom, I also won awards for my work and found recognition from colleagues whom I admired. But still no jobs arrived. 3/9

A selection of text from the post, which reads: “I had direction and soon after I gained recognition too. I won prizes for my PhD work: first from the language evolution community and then from the British Psychological Society, for the best PhD that year. Soon after the European Human Behaviour & Evolution Association gave me their New Investigator Award, and the British Humanist Association asked me to give their Darwin Day Annual Lecture. Most rewarding of all, people whose work I found inspiring started to show interest in mine.”

May 25, 2025 at 8:13 PM

DB Krupp

@dbkrupp.bsky.social

Although I was trained in the experimental tradition, I was, like Thom, drawn closer to theory and have spent the better part of the last 20 years trying to add something of value to our understanding of social evolution. Getting a job doing this work is exceedingly hard. 2/9