@mannegerell.bsky.social
Associate Professor in Criminology & co-founder Centre for policing & prevention, Malmö Uni, Sweden. Research on police, the geography of crime & fear, & crime prevention.
Bluesky is the new science Twitter, new study by @whysharksmatter.bsky.social and Julia Wester concludes!
"Results show that for every reported professional benefit that scientists once gained from Twitter, scientists can now gain that benefit more effectively on Bluesky than on Twitter."
Emelie Stiernströmer has a scoping review of facial recognition technology in law enforcement out now. It is a small but growing field, 11 studies but the oldest from 2020. Most studies are on attitudes, and more research on its use and impact needed.
www.tandfonline.com/doi/full/10....
Overall, I think our results are:
1. Believable. I think deep down we all know we publish "too many" significant results.
2. Disastrous. Selecting on significance in this way biases results across our whole literature away from zero & it stops us from learning what doesn't work.
It must be very hard to publish null results Publication practices in the social sciences act as a filter that favors statistically significant results over null findings. While the problem of selection on significance (SoS) is well-known in theory, it has been difficult to measure its scope empirically, and it has been challenging to determine how selection varies across contexts. In this article, we use large language models to extract granular and validated data on about 100,000 articles published in over 150 political science journals from 2010 to 2024. We show that fewer than 2% of articles that rely on statistical methods report null-only findings in their abstracts, while over 90% of papers highlight significant results. To put these findings in perspective, we develop and calibrate a simple model of publication bias. Across a range of plausible assumptions, we find that statistically significant results are estimated to be one to two orders of magnitude more likely to enter the published record than null results. Leveraging metadata extracted from individual articles, we show that the pattern of strong SoS holds across subfields, journals, methods, and time periods. However, a few factors such as pre-registration and randomized experiments correlate with greater acceptance of null results. We conclude by discussing implications for the field and the potential of our new dataset for investigating other questions about political science.
I have a new paper. We look at ~all stats articles in political science post-2010 & show that 94% have abstracts that claim to reject a null. Only 2% present only null results. This is hard to explain unless the research process has a filter that only lets rejections through.
11.02.2026 17:00 — 👍 637 🔁 222 💬 30 📌 51
First post of the year, new paper out today: we present possibly the biggest case of systematic Measurement Schmeasurement in tech use. It seems that most studies on gaming (videogame) addiction/disorder haven't measured gaming after all. This research took years, so long 🧵 doi.org/10.1098/rsos...
28.01.2026 09:56 — 👍 193 🔁 100 💬 8 📌 20
Line chart showing the homicide rate per million people in England and Wales from 1900 to 2024. Light purple dots represent the annual rate, a thick purple line shows the trend, and black triangles mark spikes during the World Wars. A single black dot in 2003 highlights Harold Shipman’s murders, which were recorded that year despite occurring earlier. The rate generally declined from 1900 to the mid-1960s, then rose until the early 2000s before declining again. Data source: Home Office historical crime data. Base: all recorded homicides, 1900–2024. Vertical axis ranges from 0 to 20 homicides per million people.
A bar chart comparing homicide rates per million residents in 2023 across European countries. Each country is represented by a vertical blue bar, except for England and Wales, which is highlighted in dark purple. A dotted line indicates the EU average of 9.1 homicides per million. Malta has the lowest rate, while Latvia has the highest, exceeding 40 homicides per million. England and Wales sits at almost exactly the EU average. Countries are listed along the x-axis, and the y-axis shows homicide rates. Data sources are the Home Office for England and Wales, and Eurostat for other countries.
Bar chart showing homicide rates per million residents in countries with populations over 10 million, using the most recent available data (since 2021). Countries are sorted left to right by increasing homicide rate. Japan, Indonesia, and South Korea have the lowest rates, all well below the 2023 world average of 52 homicides per million (marked by a dotted line). England and Wales appears below average. The highest rates are seen in Brazil, Mexico, Colombia, South Africa, and Ecuador, with Ecuador exceeding 450 homicides per million. Source: Home Office (UK) and World Bank. Vertical axis shows homicide rate; horizontal axis lists countries.
We're consumed by stories of murder, but how much do you know about homicide in England and Wales?
Every day this week I'm going to post a thread summarising a different aspect of what data can tell us about the nature of homicide.
🧵
For the, likely tiny, subset of you who live in/are interested in western Sweden (north of Gothenburg) i will be doing local commentary for radio P4 Väst sometime after 1600.
22.01.2026 13:56 — 👍 0 🔁 0 💬 0 📌 0
For those in Sweden with Swedish language knowledge, I will be in radio P4 this afternoon discussing some of these changes and patterns. Tune in to P4 extra to hear about it.
www.sverigesradio.se/p4extra?flik...
A final tidbit:
Stockholm is the biggest city andhas the highest per capita crime rate in the country (~19k/100k pop).
Malmö 3rd biggest city, 3rd highest crime rate (16k/100k).
But. Gothenburg 2nd biggest city, but only 18th highest rate (14k).
Clearly Gothenburg is doing something right?
Both traffic (-3%, 2k) and narcotics (-4%, 5k) down. These crimes are typically detected by the police, so a reduction likely reflects changing police priorities.
This could be considered bad (less local police presence!) or good (focusing on serious violence!) depending on ones preferences.
Regarding assault, which was up 4%, it is also interesting to note that more women (33k) than men (32k) report being assaulted to the police. Over the past decade assault against women +15% whereas against men -11%.
Women predominantly assaulted by someone they know (82%), less so for men (45%).
Another notable change regarding property crime is that we no longer have seasonality. It uses to be true that residential burglary went up in the winter as we got longer nights and more darkness, but there is barely a hint of that left. No idea why, any suggestions?
22.01.2026 11:17 — 👍 2 🔁 0 💬 1 📌 0
The theft crime drop is quite general. Personal robberies -21%, bicycle theft -15%, theft from motor vehicle -17%.
Residential burglary down 15% (-1.6k) continuing a long run trend which has seen it drop 60% in a decade. Robberies down 52% since 2016 (Figure 1).
Preliminary 2025 crime stats for Sweden were released today. Over all a 3% (50k) drop in police reported crimes. This is driven by large drops in theft (-12%, -43k) and vandalism (-13%, -26k). Assaults up 4% (+3k) and fraud up 1% (+3k).
(Link in Swedish)
bra.se/statistik/st...
You've heard about the @MarkJCarney speech. You should listen to it all. It is one of great speeches of all time. We will talk of this for decades. www.youtube.com/watch?v=CQOr...
21.01.2026 11:36 — 👍 21 🔁 6 💬 1 📌 0
Mark Carney's speech really is terrific: full text is here and very much worth your time.
20.01.2026 17:52 — 👍 1979 🔁 688 💬 74 📌 174
Donald J. Trump @realDonaldTrump Everyone is asking about the Autopen? What was done is totally illegal, and anything signed that way is of "no further force or effect." The person who "worked" the Autopen had no idea whether or not Biden approved of what he was doing. There was no ORDER in writing, and it was an absolutely illegal act perpetrated by the Radical Left Insurrectionists who illegally ran the Biden Administration. Every one of them should be arrested for what they have done to our Country. They didn't win the Presidency but, when you think of it, neither did Joe Biden. The whole thing was RIGGED. There must be a price to pay, and it has got to be a BIG ONE! PRESIDENT DONALD J. TRUMP
Spiralling.
17.01.2026 20:20 — 👍 837 🔁 143 💬 127 📌 44
I’ve never seen a US president hated this broadly worldwide. Trump’s name comes up in bars,coffee shops far from America, and it’s never complimentary.
That’s before you factor in the US itself.
Man, Du Bois had some bangers, but for me, it doesn't get much better than his scathing indictment of the "Christianity" of the United States.
Certainly this was all true of the nation in his lifetime. But in my lifetime, his criticisms have never felt so true of the nation right now.
It is part of the over all picture. Using might to achieve goals, both internally and externally. Democracy and rule of law out of the question.
17.01.2026 18:32 — 👍 0 🔁 0 💬 1 📌 0This is just a personal opinion. I do not work on these issues at all.
But where are the campaigns to boycott stuff? Why are we still treating the US as a partner? I do not get it.
It is really time to cut all ties with the US. It is an autocracy that uses force of different means to achieve more power. I do not understand why the European response is so muted.
17.01.2026 17:43 — 👍 17 🔁 2 💬 2 📌 0Europe.
Is moving military assets.
To deter an invasion.
By the United States of America.
Varifrån kommer uppskattningen om 75 mord?
06.01.2026 13:19 — 👍 0 🔁 0 💬 0 📌 0Bara ett år har sett färre mord sedan 60-talet.
31.12.2025 19:50 — 👍 114 🔁 67 💬 7 📌 7
Shootings continue to drop in Sweden, in some categories dramatically. Since 2022, shootings (incl. incidents w/no injuries) down by 65% & shooting injuries down by 60%. 43 deaths in 2025, but 2025 included worst mass shooting in Swedish history w/10 killed. Without that, deaths down 50% since 2022.
06.01.2026 11:57 — 👍 16 🔁 6 💬 1 📌 0One example of how Grok is being used to target women. Swedish Deputy Prime Minister Ebba Busch being sexualised, degraded, and humiliated step-by-step by Grok. All the images accurately reflect the prompts provided.
05.01.2026 17:37 — 👍 4161 🔁 1709 💬 296 📌 599The thing that’s so dumb about this is that there’s nothing legitimate about the strategic importance of Greenland that the US doesn’t get through existing alliances. Denmark isn’t impeding US strategic interests. It’s about them wanting to increase US land area by 20%.
06.01.2026 12:48 — 👍 583 🔁 106 💬 31 📌 11Jan. 6 was widely condemned at the time, recognized as a violent attack on America. Trump clearly caused it by lying about the election to attempt a coup.
Within a year, Trump, GOP politicians and right-wing media flipped Republicans to neutral/positive, rendering it another muddy partisan dispute.
When the original version of this "Procedural Justice Training Works" study came out in PNAS a couple of years ago I posted the thread below on twitter. I'm going to re-post here because as far as I can tell the same questions apply to this new study.
16.12.2025 15:58 — 👍 1 🔁 1 💬 1 📌 0
