Julius Kölzer
@juliusk.bsky.social
Doctoral Researcher in Political Science @dynamics.bsky.social & Research Associate @hertieschool.bsky.social | Sometimes working at Zeit Online | Interested in political geography, inequality and elections.
"causally identify yourself"
November 21, 2025 at 9:44 AM
"causally identify yourself"
Listening to this will fix all the problems in my R code
November 20, 2025 at 1:49 PM
Listening to this will fix all the problems in my R code
Normalster Tag in Norddeutschland
November 16, 2025 at 6:25 PM
Normalster Tag in Norddeutschland
Was für Post Apocalyptic, das is legit jede zweite deutsche uni zwischen november und mai
November 15, 2025 at 8:14 PM
Was für Post Apocalyptic, das is legit jede zweite deutsche uni zwischen november und mai
Getting promising results after spending months in the dark depths of data collection
November 15, 2025 at 10:48 AM
Getting promising results after spending months in the dark depths of data collection
live to fight another day
November 15, 2025 at 8:14 AM
live to fight another day
Weimar Constitution safeguarding German democracy in the early 1930s
November 14, 2025 at 7:58 AM
Weimar Constitution safeguarding German democracy in the early 1930s
Just saw this at the bus stop. Guess I am in the treated group?
November 11, 2025 at 6:34 PM
Just saw this at the bus stop. Guess I am in the treated group?
- Reisezug in Crottendorf kurz vor der Betriebseinstellung (1996) -
November 7, 2025 at 6:37 PM
- Reisezug in Crottendorf kurz vor der Betriebseinstellung (1996) -
regressions are haunting me in the grocery store
November 7, 2025 at 11:18 AM
regressions are haunting me in the grocery store
The patterns by neighborhood rent look quite different, and the relationship with education level is also more nuanced. A small 3% gap to Cuomo in precincts with incomes above $200,000 indicates that Mamdani performed relatively well in these areas.
November 6, 2025 at 8:25 AM
The patterns by neighborhood rent look quite different, and the relationship with education level is also more nuanced. A small 3% gap to Cuomo in precincts with incomes above $200,000 indicates that Mamdani performed relatively well in these areas.
Dass es in NYC republikanische Bürgermeister gab, ist, denke ich, einem weirden Parteiestablishment und absurd viel Geld geschuldet. In allen sonstigen Wahlen ist die Stadt natürlich deutlich eher links (auch bei Einstellungen I guess).
www.cambridge.org/core/journal...
www.cambridge.org/core/journal...
November 5, 2025 at 9:54 PM
Dass es in NYC republikanische Bürgermeister gab, ist, denke ich, einem weirden Parteiestablishment und absurd viel Geld geschuldet. In allen sonstigen Wahlen ist die Stadt natürlich deutlich eher links (auch bei Einstellungen I guess).
www.cambridge.org/core/journal...
www.cambridge.org/core/journal...
If the election were held today, these patterns for the Liberal Democrats look rather different. A similar image emerges for the Green Party, with largely no clear correlation between deprivation and vote share (again likely attributable to the forces of the UK’s electoral system).
November 3, 2025 at 8:41 AM
If the election were held today, these patterns for the Liberal Democrats look rather different. A similar image emerges for the Green Party, with largely no clear correlation between deprivation and vote share (again likely attributable to the forces of the UK’s electoral system).
The Lib Dems are generally more successful in areas with low deprivation, as shown by the outliers in the top right. However, the overall correlation is weak, as the party records low support in many seats, independent of their deprivation rank (reflecting the dynamics of the UK’s electoral system).
November 3, 2025 at 8:41 AM
The Lib Dems are generally more successful in areas with low deprivation, as shown by the outliers in the top right. However, the overall correlation is weak, as the party records low support in many seats, independent of their deprivation rank (reflecting the dynamics of the UK’s electoral system).
And indeed, just using basic measures of constituency-level education and qualification levels appears to be a stronger predictor of Reform UK’s 2024 vote share than more direct measures of socio-economic deprivation.
November 3, 2025 at 8:41 AM
And indeed, just using basic measures of constituency-level education and qualification levels appears to be a stronger predictor of Reform UK’s 2024 vote share than more direct measures of socio-economic deprivation.
While support for Reform UK is often noted to be higher in “left-behind” areas where deprivation is thought to be more pronounced, the correlation between their vote share and different deprivation measures remains rather weak. Only deprivation in education shows a somewhat stronger relationship.
November 3, 2025 at 8:41 AM
While support for Reform UK is often noted to be higher in “left-behind” areas where deprivation is thought to be more pronounced, the correlation between their vote share and different deprivation measures remains rather weak. Only deprivation in education shows a somewhat stronger relationship.
As expected, the geographical support patterns of the Conservatives show a near opposite image. Their vote share is significantly higher in areas with lower levels of income and employment deprivation. The same is true for deprivation in health risks and local exposure to crime.
November 3, 2025 at 8:41 AM
As expected, the geographical support patterns of the Conservatives show a near opposite image. Their vote share is significantly higher in areas with lower levels of income and employment deprivation. The same is true for deprivation in health risks and local exposure to crime.
Interestingly, the slope for local deprivation in educational attainment and skills is noticeably weaker than for income and employment deprivation.
November 3, 2025 at 8:41 AM
Interestingly, the slope for local deprivation in educational attainment and skills is noticeably weaker than for income and employment deprivation.
Even though Labour’s geographic support base is often described as shifting toward more urban, affluent, highly educated constituencies, its vote share remains clearly higher in more deprived areas. Interestingly, there is a slight positive correlation with fewer barriers to housing and services.
November 3, 2025 at 8:41 AM
Even though Labour’s geographic support base is often described as shifting toward more urban, affluent, highly educated constituencies, its vote share remains clearly higher in more deprived areas. Interestingly, there is a slight positive correlation with fewer barriers to housing and services.
The electoral outcome most strongly linked to deprivation is not any party’s vote share, but turnout. Across almost all indicators, turnout is markedly lower in more deprived areas, with only barriers to housing & services and quality in the living environment showing weaker correlations.
November 3, 2025 at 8:41 AM
The electoral outcome most strongly linked to deprivation is not any party’s vote share, but turnout. Across almost all indicators, turnout is markedly lower in more deprived areas, with only barriers to housing & services and quality in the living environment showing weaker correlations.
How do these various deprivation indices relate to voting behaviour in the last UK General Election? To explore this, I merged data constituency-level vote shares in 2024 with constituency-level ranks across the different deprivation indices. Again, a lower rank reflects greater deprivation levels.
November 3, 2025 at 8:41 AM
How do these various deprivation indices relate to voting behaviour in the last UK General Election? To explore this, I merged data constituency-level vote shares in 2024 with constituency-level ranks across the different deprivation indices. Again, a lower rank reflects greater deprivation levels.
Even though the various dimensions of deprivation are often strongly correlated, interesting differences remain. For example, some constituencies are more deprived in terms of income and employment (lower bars) but less deprived when it comes to barriers to housing and services (higher bars).
November 3, 2025 at 8:41 AM
Even though the various dimensions of deprivation are often strongly correlated, interesting differences remain. For example, some constituencies are more deprived in terms of income and employment (lower bars) but less deprived when it comes to barriers to housing and services (higher bars).