Macartan Humphreys
@macartan.bsky.social
WZB Berlin, Humboldt University, TCD
https://macartan.github.io/
Politics, conflict, inequality, political economy of development, causal inference
https://macartan.github.io/
Politics, conflict, inequality, political economy of development, causal inference
Reposted by Macartan Humphreys
▶️ Comparative Politics of Development
👉🏽 Section chairs:
@macartan.bsky.social
&
@pavisuri.bsky.social
📢 Our section invites high-quality work on any topic in the political economy of development or comparative politics in low and middle income countries.
2/
👉🏽 Section chairs:
@macartan.bsky.social
&
@pavisuri.bsky.social
📢 Our section invites high-quality work on any topic in the political economy of development or comparative politics in low and middle income countries.
2/
October 23, 2025 at 11:14 AM
▶️ Comparative Politics of Development
👉🏽 Section chairs:
@macartan.bsky.social
&
@pavisuri.bsky.social
📢 Our section invites high-quality work on any topic in the political economy of development or comparative politics in low and middle income countries.
2/
👉🏽 Section chairs:
@macartan.bsky.social
&
@pavisuri.bsky.social
📢 Our section invites high-quality work on any topic in the political economy of development or comparative politics in low and middle income countries.
2/
Not sure it’s quite what you want but making the point that big data doesn’t guard against bias
declaredesign.org/blog/posts/w...
declaredesign.org/blog/posts/w...
With great power comes great responsibility – DeclareDesign
declaredesign.org
October 13, 2025 at 5:24 PM
Not sure it’s quite what you want but making the point that big data doesn’t guard against bias
declaredesign.org/blog/posts/w...
declaredesign.org/blog/posts/w...
Not bubbles exactly but this sounds like a job for geom_smooth
ggplot2.tidyverse.org/reference/ge...
See figs at bottom with bands
FWIW my original remarks weren’t about smoothing but about sequences that are themselves smooth; thought experiment imagine an infinity of measures all fall on a line
ggplot2.tidyverse.org/reference/ge...
See figs at bottom with bands
FWIW my original remarks weren’t about smoothing but about sequences that are themselves smooth; thought experiment imagine an infinity of measures all fall on a line
Smoothed conditional means — geom_smooth
Aids the eye in seeing patterns in the presence of overplotting.
geom_smooth() and stat_smooth() are effectively aliases: they
both use the same arguments. Use stat_smooth() if you want to
display the...
ggplot2.tidyverse.org
September 28, 2025 at 11:46 AM
Not bubbles exactly but this sounds like a job for geom_smooth
ggplot2.tidyverse.org/reference/ge...
See figs at bottom with bands
FWIW my original remarks weren’t about smoothing but about sequences that are themselves smooth; thought experiment imagine an infinity of measures all fall on a line
ggplot2.tidyverse.org/reference/ge...
See figs at bottom with bands
FWIW my original remarks weren’t about smoothing but about sequences that are themselves smooth; thought experiment imagine an infinity of measures all fall on a line
noise from trends if we believe smooth processes happen in nature but noisy measurement invariably produces series that are not smooth. Just a musing though.
September 28, 2025 at 9:28 AM
noise from trends if we believe smooth processes happen in nature but noisy measurement invariably produces series that are not smooth. Just a musing though.
To me the message from the graph seems clear and yes terrifying
I was trying to think why I wasn’t bothered by the lack of error bars and I think it’s
1. The trends are clear even if we aren’t sure about the points and
2. What I was musing about: seems to me we can infer something about …
I was trying to think why I wasn’t bothered by the lack of error bars and I think it’s
1. The trends are clear even if we aren’t sure about the points and
2. What I was musing about: seems to me we can infer something about …
September 28, 2025 at 9:28 AM
To me the message from the graph seems clear and yes terrifying
I was trying to think why I wasn’t bothered by the lack of error bars and I think it’s
1. The trends are clear even if we aren’t sure about the points and
2. What I was musing about: seems to me we can infer something about …
I was trying to think why I wasn’t bothered by the lack of error bars and I think it’s
1. The trends are clear even if we aren’t sure about the points and
2. What I was musing about: seems to me we can infer something about …
These lines are fairly straight. But imagine they were very very close to perfectly straight. Would you infer anything about the second moments?
September 28, 2025 at 8:38 AM
These lines are fairly straight. But imagine they were very very close to perfectly straight. Would you infer anything about the second moments?
Reposted by Macartan Humphreys
And we are delighted to announce that @valentimvicente.bsky.social will be honored with the A.SK Bright Mind Award for his work on how social norms influence behavior. 🎉Congratulations!
September 17, 2025 at 8:09 AM
And we are delighted to announce that @valentimvicente.bsky.social will be honored with the A.SK Bright Mind Award for his work on how social norms influence behavior. 🎉Congratulations!