Stephen Burgess
@stevesphd.bsky.social
Medical statistician, work with genetic data to disentangle causation from correlation. Author of book on Mendelian randomization.
Thanks to Will Bickford Smith for co-leading this, and the people at Policy Exchange for commissioning this work. Great to see this work published! Policy document: policyexchange.org.uk/publication/..., Statistical report: policyexchange.org.uk/wp-content/u....
From School To The Skilled Workforce - Policy Exchange
Download Publication Independent Analysis Online Reader This new report by Policy Exchange makes the case that University Technical Colleges (UTCs) can play a vital role in addressing the UK’s profoun...
policyexchange.org.uk
November 14, 2025 at 10:30 AM
Thanks to Will Bickford Smith for co-leading this, and the people at Policy Exchange for commissioning this work. Great to see this work published! Policy document: policyexchange.org.uk/publication/..., Statistical report: policyexchange.org.uk/wp-content/u....
Based on student attainment data, UTC students perform less well at English at age 16, but equally well at maths (and potentially better at maths for disadvantaged students).
November 14, 2025 at 10:30 AM
Based on student attainment data, UTC students perform less well at English at age 16, but equally well at maths (and potentially better at maths for disadvantaged students).
Based on leaver data, UTC students were more likely to go into apprenticeships, potentially more likely to go into employment, and no more likely (and possibly less likely) to have no sustained outcome.
November 14, 2025 at 10:30 AM
Based on leaver data, UTC students were more likely to go into apprenticeships, potentially more likely to go into employment, and no more likely (and possibly less likely) to have no sustained outcome.
Differences in overall education participation were maintained across surveys, but differences in further education participation appeared to attenuate to zero over time.
November 14, 2025 at 10:30 AM
Differences in overall education participation were maintained across surveys, but differences in further education participation appeared to attenuate to zero over time.
Based on student destination data, UTC students were consistently less likely to be in sustained education, but more likely to be in sustained employment, and less likely to not have a sustained outcome compared with students at comparable schools.
November 14, 2025 at 10:30 AM
Based on student destination data, UTC students were consistently less likely to be in sustained education, but more likely to be in sustained employment, and less likely to not have a sustained outcome compared with students at comparable schools.
We performed a doubly-robust analysis, matching on these variables but also adjusting for them in a regression model. We also adjusted for proportion of pupils who are boys (%BOYS), but didn't match on this variable (as schools with unbalanced sex ratio are often atypical).
November 14, 2025 at 10:30 AM
We performed a doubly-robust analysis, matching on these variables but also adjusting for them in a regression model. We also adjusted for proportion of pupils who are boys (%BOYS), but didn't match on this variable (as schools with unbalanced sex ratio are often atypical).
We matched on three variables, proportion of pupils with English as an additional language (%EAL), proportion with special educational needs (%SEN), and proprtion with free school meals (%FSM).
November 14, 2025 at 10:30 AM
We matched on three variables, proportion of pupils with English as an additional language (%EAL), proportion with special educational needs (%SEN), and proprtion with free school meals (%FSM).
For most outcomes, we performed a matched analysis, matching each UTC with 5 similar schools in the same Local Educational Authority, and comparing outcomes within each matched set. This analysis uses less data than an analysis of the full dataset, but more relevant data.
November 14, 2025 at 10:30 AM
For most outcomes, we performed a matched analysis, matching each UTC with 5 similar schools in the same Local Educational Authority, and comparing outcomes within each matched set. This analysis uses less data than an analysis of the full dataset, but more relevant data.
We wanted to benchmark UTC performance for exam results and leaver outcomes. The analysis was challenging in a number of ways. How to conduct a like-with-like comparison of UTCs with similar secondary schools that are not UTCs using publicly-available data?
November 14, 2025 at 10:30 AM
We wanted to benchmark UTC performance for exam results and leaver outcomes. The analysis was challenging in a number of ways. How to conduct a like-with-like comparison of UTCs with similar secondary schools that are not UTCs using publicly-available data?
University Technical Colleges (UTCs) are non-selective state-funded “free” secondary schools in the UK (free = outside the control of the local education authority, as well as non-paying) that focus on science and technology.
November 14, 2025 at 10:30 AM
University Technical Colleges (UTCs) are non-selective state-funded “free” secondary schools in the UK (free = outside the control of the local education authority, as well as non-paying) that focus on science and technology.
Thanks to @hwang_seongwon for leading the project, to @jeffreypullin.bsky.social for performing code review, and to @chr1sw.bsky.social allace and John Whittaker for co-supervising - has been a fun project so far, and look forward to getting feedback from the community!
November 8, 2025 at 3:40 PM
Thanks to @hwang_seongwon for leading the project, to @jeffreypullin.bsky.social for performing code review, and to @chr1sw.bsky.social allace and John Whittaker for co-supervising - has been a fun project so far, and look forward to getting feedback from the community!
However, like all statistical methods, it has limitations, and results should not be thought of as unquestionable truth. It is likely that the differences between datasets in other applications are similar or stronger than those we considered here.
November 8, 2025 at 3:40 PM
However, like all statistical methods, it has limitations, and results should not be thought of as unquestionable truth. It is likely that the differences between datasets in other applications are similar or stronger than those we considered here.
In conclusion, while all methods were well-calibrated in the baseline scenario, they struggled to declare colocalization to different degrees when the datasets varied in terms of platform and population. Colocalization can be a valuable tool for triaging and prioritizing.
November 8, 2025 at 3:40 PM
In conclusion, while all methods were well-calibrated in the baseline scenario, they struggled to declare colocalization to different degrees when the datasets varied in terms of platform and population. Colocalization can be a valuable tool for triaging and prioritizing.
This was not intended to be a fair comparison - fairness is impossible to achieve. For example, coloc-SuSiE was judged to support colocalization if there was high PP.H4 for any pair of credible sets. Rather, we wanted to compare methods as they would typically be used.
November 8, 2025 at 3:40 PM
This was not intended to be a fair comparison - fairness is impossible to achieve. For example, coloc-SuSiE was judged to support colocalization if there was high PP.H4 for any pair of credible sets. Rather, we wanted to compare methods as they would typically be used.
We acknowledge that there are many legitimate reasons why we may observe non-colocalization for the same protein when using estimates from different platforms / populations. Also, we acknowledge that different methods use different standards of evidence.
November 8, 2025 at 3:40 PM
We acknowledge that there are many legitimate reasons why we may observe non-colocalization for the same protein when using estimates from different platforms / populations. Also, we acknowledge that different methods use different standards of evidence.
Enumeration methods tended to outperform proportional methods in most scenarios. However, no single approach dominated in all scenarios, with coloc-SuSiE reporting the highest rate of colocalization in Case 1, Case 2B, and Case 4; colocPropTest in Case 2F; and coloc in Case 3.
November 8, 2025 at 3:40 PM
Enumeration methods tended to outperform proportional methods in most scenarios. However, no single approach dominated in all scenarios, with coloc-SuSiE reporting the highest rate of colocalization in Case 1, Case 2B, and Case 4; colocPropTest in Case 2F; and coloc in Case 3.
In these cases, results were more mixed. We observed frequent disagreement between methods as to whether there was colocalization, non-colocalization, or insufficient evidence. In the worst-case scenario, colocalization was only agreed by all four methods for 20% of proteins.
November 8, 2025 at 3:40 PM
In these cases, results were more mixed. We observed frequent disagreement between methods as to whether there was colocalization, non-colocalization, or insufficient evidence. In the worst-case scenario, colocalization was only agreed by all four methods for 20% of proteins.
We then consider associations with the same protein, but measured on different platforms (Olink vs SomaLogic in British [Case 2B] and Finnish [Case 2F] populations), and measured in different populations (British vs Finnish for Olink [Case 3] and SomaLogic [Case 4]).
November 8, 2025 at 3:40 PM
We then consider associations with the same protein, but measured on different platforms (Olink vs SomaLogic in British [Case 2B] and Finnish [Case 2F] populations), and measured in different populations (British vs Finnish for Olink [Case 3] and SomaLogic [Case 4]).
In the baseline context, we split the UK Biobank Pharma Proteomics Project in two at random, and tested associations for the same protein in one half of the data versus the other half of the data (Case 1). Unsurprisingly, all methods performed well in this context.
November 8, 2025 at 3:40 PM
In the baseline context, we split the UK Biobank Pharma Proteomics Project in two at random, and tested associations for the same protein in one half of the data versus the other half of the data (Case 1). Unsurprisingly, all methods performed well in this context.
We perform colocalization for protein-coding gene regions with ≥1 pQTL across four datasets using four colocalization methods: coloc, coloc-SuSiE, prop.coloc, and colocPropTest in a range of contexts.
November 8, 2025 at 3:40 PM
We perform colocalization for protein-coding gene regions with ≥1 pQTL across four datasets using four colocalization methods: coloc, coloc-SuSiE, prop.coloc, and colocPropTest in a range of contexts.
Big thanks to all co-authors for contributing to this: @amymariemason.bsky.social, @VerenaZuber, @explodecomputer, Elena, @IamYuXu, Amanda, @BarWoolf, @eliasallara, @dpsg108, and @OpeSoremekun. Feedback would be very welcome!
October 27, 2025 at 8:43 AM
Big thanks to all co-authors for contributing to this: @amymariemason.bsky.social, @VerenaZuber, @explodecomputer, Elena, @IamYuXu, Amanda, @BarWoolf, @eliasallara, @dpsg108, and @OpeSoremekun. Feedback would be very welcome!
Critical is what we can assume is shared between populations, and what is different - are we clear what we are assuming can be borrowed? And is it reasonable to borrow that information?
October 27, 2025 at 8:43 AM
Critical is what we can assume is shared between populations, and what is different - are we clear what we are assuming can be borrowed? And is it reasonable to borrow that information?
When analysing non-European data, there is often a compromise between only including the most relevant data to the target population, and including all available data from any population - we describe some approaches to this taken in the literature.
October 27, 2025 at 8:43 AM
When analysing non-European data, there is often a compromise between only including the most relevant data to the target population, and including all available data from any population - we describe some approaches to this taken in the literature.