ObjectivesThe lack of an income question on the Census has meant the production of multivariate income by ethnicity statistics has not been possible in census outputs to date. Our ambition is to provide individual-level records for every member of the usually resident population of England and Wales using admin data.
MethodExciting progress has been made in the development of admin-based characteristics measures, including the ongoing feasibility research to produce admin-based datasets on ethnic group and income. Demonstrated improved results include the coverage of the population steadily increasing and methods to create these datasets gradually improving.
Access to these record-level administrative datasets has allowed us to combine admin-based income and ethnicity measures developed in previous research, linking individuals between the two. We review the coverage of the combined dataset and the feasibility of producing multivariate statistics at subnational levels in England and Wales for the first time.
ResultsThis presentation will showcase our innovative progress so far. By combining admin-based income and admin-based ethnicity datasets, we established an income and a stated ethnicity for 77.1% of people in England and 82.1% of people in Wales aged 16 years and over in the admin-based Statistical Population Dataset (our population base).
For the first time, we have produced income percentiles for ethnic groups at different levels of geography in England and Wales including national figures, regional figures and figures for local authorities and lower layer super output areas; although statistical disclosure control means that some of the figures have been suppressed.
We will highlight some of the challenges in using administrative data sources to produce these statistics and in assessing their statistical quality.
ConclusionOur research developing these statistics is truly novel and shows much promise. Future work will include research to improve the univariate admin-based measures that are used, to continue to explore the limitations of the combined dataset, to explore the data by occupied address, and explore methods to adjust for missingness.
{"title":"Social Statistics Transformation: Understanding the population through the production of income by ethnicity statistics from administrative data","authors":"Joanna Harkrader, Michelle Bellham, Samantha Pendleton, Alison Morgan, Joe Pearce, Emily Stennard","doi":"10.23889/ijpds.v8i2.2243","DOIUrl":"https://doi.org/10.23889/ijpds.v8i2.2243","url":null,"abstract":"ObjectivesThe lack of an income question on the Census has meant the production of multivariate income by ethnicity statistics has not been possible in census outputs to date. Our ambition is to provide individual-level records for every member of the usually resident population of England and Wales using admin data.
 MethodExciting progress has been made in the development of admin-based characteristics measures, including the ongoing feasibility research to produce admin-based datasets on ethnic group and income. Demonstrated improved results include the coverage of the population steadily increasing and methods to create these datasets gradually improving.
 Access to these record-level administrative datasets has allowed us to combine admin-based income and ethnicity measures developed in previous research, linking individuals between the two. We review the coverage of the combined dataset and the feasibility of producing multivariate statistics at subnational levels in England and Wales for the first time.
 ResultsThis presentation will showcase our innovative progress so far. By combining admin-based income and admin-based ethnicity datasets, we established an income and a stated ethnicity for 77.1% of people in England and 82.1% of people in Wales aged 16 years and over in the admin-based Statistical Population Dataset (our population base).
 For the first time, we have produced income percentiles for ethnic groups at different levels of geography in England and Wales including national figures, regional figures and figures for local authorities and lower layer super output areas; although statistical disclosure control means that some of the figures have been suppressed.
 We will highlight some of the challenges in using administrative data sources to produce these statistics and in assessing their statistical quality.
 ConclusionOur research developing these statistics is truly novel and shows much promise. Future work will include research to improve the univariate admin-based measures that are used, to continue to explore the limitations of the combined dataset, to explore the data by occupied address, and explore methods to adjust for missingness.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"144 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134913250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-14DOI: 10.23889/ijpds.v8i2.2197
Erin Early, Sarah Miller, Laura Dunne, John Moriarty
ObjectivesThis study examined the individual and collective impacts of socio-demographics and school-level factors on GCSE outcomes in Northern Ireland, using linked administrative data. A pupil’s sex, religious affiliation and socio-economic background (measured by eight measures) were examined, along with parental socio-economic background, attended school type (grammar/non-grammar) and school management structure.
MethodThis study used the first linked administrative dataset for education in Northern Ireland. The dataset linked the 2011 household Census, School Leavers Survey (2010-2014) and School Census (2010-2014) for the first time. Data were provided for three pupil cohorts who completed their GCSE examinations in consecutive academic years (2010/2011 – 2012/2013).
The study conducted multilevel models to understand the nested effects of pupil-, household- and school-level factors on GCSE attainment outcomes. Interaction models were also executed to examine the multiplicative effects of a pupil’s sex, religious affiliation and socio-economic background on their educational attainment.
ResultsThe findings of this study highlight that the impact of socio-economic status is multidimensional, with some measures having a greater impact on GCSE attainment than others. For example, a mother’s education qualifications had the largest impact of socio-economic measures included in the multilevel models. The analysis also found that Free School Meal Eligibility remains an important predictor of attainment outcomes. When considering pupils’ sex, females had higher GCSE attainment scores than males. However, religious affiliation had a varied influence on GCSE outcomes, indicating the need for a more nuanced approach when considering this factor. The importance of interaction terms to gain an in-depth understanding of the multiplicative effect of factors on attainment outcomes was also highlighted in the analysis.
ConclusionThrough the use of linked administrative data, this study highlights the hierarchy of socio-economic effects on GCSE attainment outcomes in Northern Ireland. It also highlights the importance of collectively considering the factors that make up a pupil’s demographic profile to garner a holistic understanding of attainment trends in Northern Ireland.
{"title":"Understanding the disparity of educational attainment in Northern Ireland: The role of socio-demographic and school-level factors on GCSE attainment.","authors":"Erin Early, Sarah Miller, Laura Dunne, John Moriarty","doi":"10.23889/ijpds.v8i2.2197","DOIUrl":"https://doi.org/10.23889/ijpds.v8i2.2197","url":null,"abstract":"ObjectivesThis study examined the individual and collective impacts of socio-demographics and school-level factors on GCSE outcomes in Northern Ireland, using linked administrative data. A pupil’s sex, religious affiliation and socio-economic background (measured by eight measures) were examined, along with parental socio-economic background, attended school type (grammar/non-grammar) and school management structure.
 MethodThis study used the first linked administrative dataset for education in Northern Ireland. The dataset linked the 2011 household Census, School Leavers Survey (2010-2014) and School Census (2010-2014) for the first time. Data were provided for three pupil cohorts who completed their GCSE examinations in consecutive academic years (2010/2011 – 2012/2013).
 The study conducted multilevel models to understand the nested effects of pupil-, household- and school-level factors on GCSE attainment outcomes. Interaction models were also executed to examine the multiplicative effects of a pupil’s sex, religious affiliation and socio-economic background on their educational attainment.
 ResultsThe findings of this study highlight that the impact of socio-economic status is multidimensional, with some measures having a greater impact on GCSE attainment than others. For example, a mother’s education qualifications had the largest impact of socio-economic measures included in the multilevel models. The analysis also found that Free School Meal Eligibility remains an important predictor of attainment outcomes. When considering pupils’ sex, females had higher GCSE attainment scores than males. However, religious affiliation had a varied influence on GCSE outcomes, indicating the need for a more nuanced approach when considering this factor. The importance of interaction terms to gain an in-depth understanding of the multiplicative effect of factors on attainment outcomes was also highlighted in the analysis.
 ConclusionThrough the use of linked administrative data, this study highlights the hierarchy of socio-economic effects on GCSE attainment outcomes in Northern Ireland. It also highlights the importance of collectively considering the factors that make up a pupil’s demographic profile to garner a holistic understanding of attainment trends in Northern Ireland.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"215 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134913334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-14DOI: 10.23889/ijpds.v8i2.2214
Matt Dickson
ObjectivesThis paper estimates the effect of continuing in education post-16 on the probability of experiencing youth custody at ages 17 and 18, addressing the issue of non-random selection into continued participation to derive a causal estimate.
MethodsWe exploit the natural experiment created by the ‘raising of the participation age’ (RPA) in England. Unlike previous cohorts who could leave education aged 16, young people starting the final year of compulsory schooling in September 2012 were required to continue in education or training until the end of the school year in which they turned 17, and those starting the final year in September 2013 were required to continue until age 18. Using linked National Pupil Database and National Client Caseload Information System data we utilise the variation in participation between cohorts that the RPA induced to estimate the causal effect of continued participation on custody outcomes at ages 17 and 18.
ResultsThe effect of the law change was to increase the proportion of young people participating in education at age 17 by 1.7pp (1.2pp) for boys (girls), from a base of 82.1% (85.0%) prior to the reform. Despite this increase in participation, there was no effect on the probability of custody when aged 17 or 18. This suggests that the 0.64pp (0.04pp) reduction in probability of custody associated with continued participation for boys (girls) estimated without addressing the selection issue, is actually capturing the effect on custody probability of the unobservable characteristics of those who choose to continue in education beyond 16. Results are robust to different estimation methods and different treatment specifications.
ConclusionThe negative relationship between education and crime is well documented but the decision to remain in education beyond the compulsory age is not random. Evidence here suggests that the cross-sectional reduction in probability of custody associated with continued education is driven by the unobservable characteristics of those who voluntarily continue their education rather than reflecting a causal effect of education.
{"title":"The effect of education participation on youth custody: Causal evidence from England","authors":"Matt Dickson","doi":"10.23889/ijpds.v8i2.2214","DOIUrl":"https://doi.org/10.23889/ijpds.v8i2.2214","url":null,"abstract":"ObjectivesThis paper estimates the effect of continuing in education post-16 on the probability of experiencing youth custody at ages 17 and 18, addressing the issue of non-random selection into continued participation to derive a causal estimate.
 MethodsWe exploit the natural experiment created by the ‘raising of the participation age’ (RPA) in England. Unlike previous cohorts who could leave education aged 16, young people starting the final year of compulsory schooling in September 2012 were required to continue in education or training until the end of the school year in which they turned 17, and those starting the final year in September 2013 were required to continue until age 18. Using linked National Pupil Database and National Client Caseload Information System data we utilise the variation in participation between cohorts that the RPA induced to estimate the causal effect of continued participation on custody outcomes at ages 17 and 18.
 ResultsThe effect of the law change was to increase the proportion of young people participating in education at age 17 by 1.7pp (1.2pp) for boys (girls), from a base of 82.1% (85.0%) prior to the reform. Despite this increase in participation, there was no effect on the probability of custody when aged 17 or 18. This suggests that the 0.64pp (0.04pp) reduction in probability of custody associated with continued participation for boys (girls) estimated without addressing the selection issue, is actually capturing the effect on custody probability of the unobservable characteristics of those who choose to continue in education beyond 16. Results are robust to different estimation methods and different treatment specifications.
 ConclusionThe negative relationship between education and crime is well documented but the decision to remain in education beyond the compulsory age is not random. Evidence here suggests that the cross-sectional reduction in probability of custody associated with continued education is driven by the unobservable characteristics of those who voluntarily continue their education rather than reflecting a causal effect of education.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134913479","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-14DOI: 10.23889/ijpds.v8i2.2318
Leah Quinn, Rachel Shipsey
ObjectivesLinking large-scale datasets is challenging due to the computational power required. This research explores using Locality-Sensitive-Hashing (LSH) as a blocking method to reduce the computational complexity when linking large administrative datasets. LSH hashes similar data into ‘buckets’, thus reducing the search space and processing power required to find links.
MethodsA gold-standard linked dataset was used during method development. Test datasets were made using samples of gold-standard matches and non-matches, then blocked using LSH.
Various LSH parameters including shingle length, signature length, band size and number of matching bands were tested. Precision and recall were used to find optimal parameters for identifying good candidate pairs, with 100% recall and >20% precision being desirable.
Alternative formats for date of birth, postcode and gender variables were tested, with additional characters used to simulate agreement weighting.
ResultsResults as of spring 2023 are promising, with the caveat that currently only small datasets have been tested. The LSH method with optimal parameters creates ~9,000 candidate pairs whilst maintaining recall of 100% (i.e., all true matches are included in the candidate pairs) and precision of 27.6%. In contrast, our traditional deterministic blocking method using the same variables creates ~70,000 candidate pairs, and a cartesian product creates over 23.4 million candidate pairs. We have therefore shown that LSH can be used to create a significant reduction in the search-space size.
Furthermore, the method easily handles alternative names, postcodes, etc. that may be present in longitudinal data or composite datasets, with no need to account for different possible combinations of variables.
ConclusionCurrent research has shown that LSH can be used to drastically reduce the search space when blocking for data linkage. Using variable formatting to prioritise agreement for specific sections e.g., of postcode, has overcome a potential downside of LSH. Further research on variable formatting, parameter optimisation and testing of the method at scale is ongoing.
{"title":"Exploring locality sensitive hashing as a blocking method for large-scale administrative datasets","authors":"Leah Quinn, Rachel Shipsey","doi":"10.23889/ijpds.v8i2.2318","DOIUrl":"https://doi.org/10.23889/ijpds.v8i2.2318","url":null,"abstract":"ObjectivesLinking large-scale datasets is challenging due to the computational power required. This research explores using Locality-Sensitive-Hashing (LSH) as a blocking method to reduce the computational complexity when linking large administrative datasets. LSH hashes similar data into ‘buckets’, thus reducing the search space and processing power required to find links.
 MethodsA gold-standard linked dataset was used during method development. Test datasets were made using samples of gold-standard matches and non-matches, then blocked using LSH.
 Various LSH parameters including shingle length, signature length, band size and number of matching bands were tested. Precision and recall were used to find optimal parameters for identifying good candidate pairs, with 100% recall and >20% precision being desirable.
 Alternative formats for date of birth, postcode and gender variables were tested, with additional characters used to simulate agreement weighting.
 ResultsResults as of spring 2023 are promising, with the caveat that currently only small datasets have been tested. The LSH method with optimal parameters creates ~9,000 candidate pairs whilst maintaining recall of 100% (i.e., all true matches are included in the candidate pairs) and precision of 27.6%. In contrast, our traditional deterministic blocking method using the same variables creates ~70,000 candidate pairs, and a cartesian product creates over 23.4 million candidate pairs. We have therefore shown that LSH can be used to create a significant reduction in the search-space size.
 Furthermore, the method easily handles alternative names, postcodes, etc. that may be present in longitudinal data or composite datasets, with no need to account for different possible combinations of variables.
 ConclusionCurrent research has shown that LSH can be used to drastically reduce the search space when blocking for data linkage. Using variable formatting to prioritise agreement for specific sections e.g., of postcode, has overcome a potential downside of LSH. Further research on variable formatting, parameter optimisation and testing of the method at scale is ongoing.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134913582","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-14DOI: 10.23889/ijpds.v8i2.2306
Emma Gorman, Dave Thomson, Peter Urwin, Zhang Min
We examine the post age-16 educational pathways taken by the 44% of young people who do not gain “good” grades in English and Maths at age 16 years. We then assess the causal effects of attending General Further Education (GFE) colleges on education and labour market outcomes for this group.
We use the Longitudinal Education Outcomes dataset, which comprises linked administrative education, employment and income records for the population of English school pupils aged 16 in 2011. To summarise complex post-16 education trajectories, we present Sankey charts stratified by indicators of disadvantage. We study the effects of attending GFE at age 17 on whether a pupil gains a Level 3 qualification by age 19, and their earnings and employment status at age 24. To estimate a causal impact, we use distance from home to the closest GFE college as an instrumental variable, controlling for a rich set of background characteristics.
Our graphical results highlight the complexity of post-16 educational pathways and transitions, which are differentiated by disadvantage. Over 50% have GFE as their first post-16 destination. Results from instrumental variable analyses show a positive association between attending GFE and gaining a Level 3 qualification by age 19, among pupils who do not gain a “good” pass in the General Certificate of Secondary Education (GCSE) in either English and/or Maths. Restricting analyses to the bottom of the distribution – those who gain an E, F or G grade in both English and Maths - we do not detect an impact of GFE on qualifications at age 19. Among both subgroups, we do not detect any impact of attending GFE on earnings and employment at age 24 years.
While the post-16 pathways taken by disadvantaged, lower-attaining pupils do increase qualification attainment for some, the value these have in the labour market appears limited. These results may indicate the importance of “soft-skills” and early employment experiences for this subgroup of lower-attainers.
{"title":"Education pathways to the labour market for 16-year-olds who struggle to achieve maths and English in General Certificate of Secondary Education","authors":"Emma Gorman, Dave Thomson, Peter Urwin, Zhang Min","doi":"10.23889/ijpds.v8i2.2306","DOIUrl":"https://doi.org/10.23889/ijpds.v8i2.2306","url":null,"abstract":"We examine the post age-16 educational pathways taken by the 44% of young people who do not gain “good” grades in English and Maths at age 16 years. We then assess the causal effects of attending General Further Education (GFE) colleges on education and labour market outcomes for this group.
 We use the Longitudinal Education Outcomes dataset, which comprises linked administrative education, employment and income records for the population of English school pupils aged 16 in 2011. To summarise complex post-16 education trajectories, we present Sankey charts stratified by indicators of disadvantage. We study the effects of attending GFE at age 17 on whether a pupil gains a Level 3 qualification by age 19, and their earnings and employment status at age 24. To estimate a causal impact, we use distance from home to the closest GFE college as an instrumental variable, controlling for a rich set of background characteristics.
 Our graphical results highlight the complexity of post-16 educational pathways and transitions, which are differentiated by disadvantage. Over 50% have GFE as their first post-16 destination. Results from instrumental variable analyses show a positive association between attending GFE and gaining a Level 3 qualification by age 19, among pupils who do not gain a “good” pass in the General Certificate of Secondary Education (GCSE) in either English and/or Maths. Restricting analyses to the bottom of the distribution – those who gain an E, F or G grade in both English and Maths - we do not detect an impact of GFE on qualifications at age 19. Among both subgroups, we do not detect any impact of attending GFE on earnings and employment at age 24 years.
 While the post-16 pathways taken by disadvantaged, lower-attaining pupils do increase qualification attainment for some, the value these have in the labour market appears limited. These results may indicate the importance of “soft-skills” and early employment experiences for this subgroup of lower-attainers.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134913831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-14DOI: 10.23889/ijpds.v8i2.2249
Kirsteen Campbell, Rebecca Whitehorn, Simon Browning, Rebecca Harmston, Ray Harris, Szu-Chia Huang, Dianna Moylan, Karen Williams, Jacqueline Oakley, Katharine Evans, Stela McLachlan, Richard Thomas, Emma Turner, Robin Flaig, Andy Boyd
ObjectivesWe created a panel with members of the public and longitudinal study participants who review our data access requests. This panel forms an integral part of our data access application process, giving the public a say who can access the data for research.
MethodsWe advertised our lay member vacancies using social media, newsletters, word of mouth and the internet. We appointed six people to the public panel. Our panel includes study participants, NHS service users, parents, carers, and people with experience of disability, neurodiversity, and long-term health conditions.
The Panel Terms of Reference were created with help from stakeholders and study teams involved in longitudinal studies that involve the public in data access applications. This ensured that the purpose of the panel was clear. The panel reviews lay summaries and makes sure that researchers have adequate public involvement in their project.
ResultsPanel members have reviewed 28 applications. Researchers present their research at an online meeting with the panel then answer questions from the panel members. We publish meeting minutes on our website for transparency.
A 6-month review was overwhelmingly positive - all panel members indicated they felt valued. They felt able to challenge and question researchers as part of the data access application process. This provides a level of public scrutiny to our work.
“I feel there’s a real value in the panel. You get a real sense that this has got such potential to make a contribution.” (panel member)
We are further developing the Panel Terms of Reference with panel members. We will consider additional areas of responsibility, for example, public benefit review.
ConclusionWe regularly review how to improve public involvement in our work. The panel has proven its value during our application process. Therefore we are exploring with the panel a new approach to assess the public benefit of applications and what is meant by ‘public benefit research’.
{"title":"A public panel reviews applications and questions applicants: Team member and public contributor discuss a transparent and inclusive approach to data access reviews","authors":"Kirsteen Campbell, Rebecca Whitehorn, Simon Browning, Rebecca Harmston, Ray Harris, Szu-Chia Huang, Dianna Moylan, Karen Williams, Jacqueline Oakley, Katharine Evans, Stela McLachlan, Richard Thomas, Emma Turner, Robin Flaig, Andy Boyd","doi":"10.23889/ijpds.v8i2.2249","DOIUrl":"https://doi.org/10.23889/ijpds.v8i2.2249","url":null,"abstract":"ObjectivesWe created a panel with members of the public and longitudinal study participants who review our data access requests. This panel forms an integral part of our data access application process, giving the public a say who can access the data for research.
 MethodsWe advertised our lay member vacancies using social media, newsletters, word of mouth and the internet. We appointed six people to the public panel. Our panel includes study participants, NHS service users, parents, carers, and people with experience of disability, neurodiversity, and long-term health conditions.
 The Panel Terms of Reference were created with help from stakeholders and study teams involved in longitudinal studies that involve the public in data access applications. This ensured that the purpose of the panel was clear. The panel reviews lay summaries and makes sure that researchers have adequate public involvement in their project.
 ResultsPanel members have reviewed 28 applications. Researchers present their research at an online meeting with the panel then answer questions from the panel members. We publish meeting minutes on our website for transparency.
 A 6-month review was overwhelmingly positive - all panel members indicated they felt valued. They felt able to challenge and question researchers as part of the data access application process. This provides a level of public scrutiny to our work.
 “I feel there’s a real value in the panel. You get a real sense that this has got such potential to make a contribution.” (panel member)
 We are further developing the Panel Terms of Reference with panel members. We will consider additional areas of responsibility, for example, public benefit review.
 ConclusionWe regularly review how to improve public involvement in our work. The panel has proven its value during our application process. Therefore we are exploring with the panel a new approach to assess the public benefit of applications and what is meant by ‘public benefit research’.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134913928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-14DOI: 10.23889/ijpds.v8i2.2218
Isobel Ward, Katie Finning, Daniel Ayoubkhani, Katie Hendry, Emma Sharland, Louis Appleby, Vahé Nafilyan
ObjectivesWith suicide a major public health concern, it is vital research identifies predictors of suicide to support vulnerable groups who should be targeted for intervention. We use a novel linkage of 2011 Census and population level mortality data to assess which risk factors are important predictors of suicide.
MethodsExposures of interest were identified from Census 2011 and were sex, age, ethnicity, marital status, day-to-day impairments, religion, region, National Statistics Socio-economic Classification. Our study population consisted of 35,136,917 people aged 18-to-74; there were 35,928 suicides in our study period (28/03/2011-31/12/2021), with 73.9% occurring in men. We fitted generalised linear models with a Poisson link function, with suicide being the outcome of interest. The natural logarithm of exposure time was included as an offset term. To estimate rates of suicide per 100,000 people for each level of our exposure, by sex for the average age, we calculated marginal means.
ResultsThe groups with the highest rates of suicide were those who reported an impairment affecting their day-to-day activities, those who were long term unemployed or never had worked, or those who were single or separated. Comparison of minimally adjusted models with models accounting for all other characteristics identified predictors which remain important risk factors after accounting for other characteristics; day-to-day impairments were still found to increase the incidence of suicide relative to those whose activities were not impaired after adjusting for employment status. Additionally, the estimated rates of suicide remained lowest in London compared to other regions in our fully adjusted estimates. Overall, rates of suicide were higher in men compared to females across all ages, with the highest rates in 40- to 50-year-olds.
ConclusionThe findings of this work provide novel population level insights into the risk of suicide by sociodemographic characteristics, this work should pave the way for further research exploring the interaction of factors which lead to suicide and drive policy change for targeted intervention.
{"title":"Sociodemographic inequalities of suicide: A population-based cohort study of adults in England and Wales 2011-2021","authors":"Isobel Ward, Katie Finning, Daniel Ayoubkhani, Katie Hendry, Emma Sharland, Louis Appleby, Vahé Nafilyan","doi":"10.23889/ijpds.v8i2.2218","DOIUrl":"https://doi.org/10.23889/ijpds.v8i2.2218","url":null,"abstract":"ObjectivesWith suicide a major public health concern, it is vital research identifies predictors of suicide to support vulnerable groups who should be targeted for intervention. We use a novel linkage of 2011 Census and population level mortality data to assess which risk factors are important predictors of suicide.
 MethodsExposures of interest were identified from Census 2011 and were sex, age, ethnicity, marital status, day-to-day impairments, religion, region, National Statistics Socio-economic Classification. Our study population consisted of 35,136,917 people aged 18-to-74; there were 35,928 suicides in our study period (28/03/2011-31/12/2021), with 73.9% occurring in men. We fitted generalised linear models with a Poisson link function, with suicide being the outcome of interest. The natural logarithm of exposure time was included as an offset term. To estimate rates of suicide per 100,000 people for each level of our exposure, by sex for the average age, we calculated marginal means.
 ResultsThe groups with the highest rates of suicide were those who reported an impairment affecting their day-to-day activities, those who were long term unemployed or never had worked, or those who were single or separated. Comparison of minimally adjusted models with models accounting for all other characteristics identified predictors which remain important risk factors after accounting for other characteristics; day-to-day impairments were still found to increase the incidence of suicide relative to those whose activities were not impaired after adjusting for employment status. Additionally, the estimated rates of suicide remained lowest in London compared to other regions in our fully adjusted estimates. Overall, rates of suicide were higher in men compared to females across all ages, with the highest rates in 40- to 50-year-olds.
 ConclusionThe findings of this work provide novel population level insights into the risk of suicide by sociodemographic characteristics, this work should pave the way for further research exploring the interaction of factors which lead to suicide and drive policy change for targeted intervention.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134913933","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-14DOI: 10.23889/ijpds.v8i2.2314
Catriona Connell, Richard Kjellgren, Jan Savinc
ObjectivesTo compare the use of NHS services for mental health and substance use (MH/SU) between people released from prison and the general population. This paper describes the data linkage and analytical process, discusses policy implications and highlights the methodological contributions for future administrative data research in public health and justice.
MethodsRetrospective cohort study using linked Scottish health data and the Scottish Prison Service (SPS), involving all individuals released from prison in 2015 (n = 14,000), and a random general population sample (n = 70,000), matched on index date, age, sex, and postcode. Analysis will include descriptive comparison of service use between the two cohorts. Multiple regression models will be fitted to examine the influence of confounding variables in service use, and multilevel models will specifically assess cross-level geographical variation where feasible.
ResultsResults of the data linkage and analysis to date will be presented. This research will contribute to understanding the complex range of contacts people have with health services for MH/SU following imprisonment. It is the first research in Scotland to provide a national-level description of access to health services for MH/SU among people released from prison and offer a comparison to the general population. It also explores within-group differences in service access for people released from prison.
ConclusionThe linking and analysis of multiple justice and health-related datasets will provide crucial evidence to inform future healthcare delivery for justice-experienced populations. This research also advances our understanding of public health approaches and administrative data research in justice-related contexts.
{"title":"Novel linkage of health and prison data in Scotland: Investigating access to services for mental health and substance use following release from prison","authors":"Catriona Connell, Richard Kjellgren, Jan Savinc","doi":"10.23889/ijpds.v8i2.2314","DOIUrl":"https://doi.org/10.23889/ijpds.v8i2.2314","url":null,"abstract":"ObjectivesTo compare the use of NHS services for mental health and substance use (MH/SU) between people released from prison and the general population. This paper describes the data linkage and analytical process, discusses policy implications and highlights the methodological contributions for future administrative data research in public health and justice.
 MethodsRetrospective cohort study using linked Scottish health data and the Scottish Prison Service (SPS), involving all individuals released from prison in 2015 (n = 14,000), and a random general population sample (n = 70,000), matched on index date, age, sex, and postcode. Analysis will include descriptive comparison of service use between the two cohorts. Multiple regression models will be fitted to examine the influence of confounding variables in service use, and multilevel models will specifically assess cross-level geographical variation where feasible.
 ResultsResults of the data linkage and analysis to date will be presented. This research will contribute to understanding the complex range of contacts people have with health services for MH/SU following imprisonment. It is the first research in Scotland to provide a national-level description of access to health services for MH/SU among people released from prison and offer a comparison to the general population. It also explores within-group differences in service access for people released from prison.
 ConclusionThe linking and analysis of multiple justice and health-related datasets will provide crucial evidence to inform future healthcare delivery for justice-experienced populations. This research also advances our understanding of public health approaches and administrative data research in justice-related contexts.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134914023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-14DOI: 10.23889/ijpds.v8i2.2323
Megan Munro, Patrycja Delong-Smith, Pete Matthews, Loes Charlton, Megan Lyons, James Tucker
There is an appetite to improve the reproducibility of quantitative analysis undertaken across government. Our team supports conversion of regular publications into Reproducible Analytical Pipelines (RAP). To achieve this, we developed a roadmap to work with analysts in our organisation and help them transform their pipelines and build their skills.
An audit of the current RAP status of all regular pipelines assisted in resource allocation and planning. The maturity of a RAP is evaluated on 7 criteria required to reach a minimum viable product (MVP) and 7 additional advanced criteria as outlined by the Analysis Function (AF). We use a combination of hands-on pair-coding with analysis teams, regular and ad hoc code reviews, and training sessions to convert existing pipelines into RAPs, while simultaneously upskilling the analysts. We have also developed guidance and training documentation to share internally and externally.
Currently, out of 73 regular publications, 12 have reached the MVP, with an average score of 4.36 out of 7. This scoring is reassessed monthly, allowing us to track the progress in real-time. Self-assessment of technical skills increased by between 43% and 89% and 97% said their understanding of RAP principals improved because of the training and 77% said they are now able to implement best practice into their work. By working with the pipeline owners instead of just refactoring the code directly, we are ensuring business resilience. The in-depth knowledge of the pipeline and skills required to maintain it are present within the analysis team. Publishing our methods, documentation and tools facilitates adoption of RAP for those without a dedicated RAP team.
We are on track to convert all our regular publications into RAPs and move to “RAP by default”, in line with the AF RAP Strategy. This will improve the reproducibility, quality, efficiency, transparency, and trustworthiness of analysis within government. We hope other organisations can learn from our methods.
{"title":"Introducing best practice for reproducibility in government","authors":"Megan Munro, Patrycja Delong-Smith, Pete Matthews, Loes Charlton, Megan Lyons, James Tucker","doi":"10.23889/ijpds.v8i2.2323","DOIUrl":"https://doi.org/10.23889/ijpds.v8i2.2323","url":null,"abstract":"There is an appetite to improve the reproducibility of quantitative analysis undertaken across government. Our team supports conversion of regular publications into Reproducible Analytical Pipelines (RAP). To achieve this, we developed a roadmap to work with analysts in our organisation and help them transform their pipelines and build their skills.
 An audit of the current RAP status of all regular pipelines assisted in resource allocation and planning. The maturity of a RAP is evaluated on 7 criteria required to reach a minimum viable product (MVP) and 7 additional advanced criteria as outlined by the Analysis Function (AF). We use a combination of hands-on pair-coding with analysis teams, regular and ad hoc code reviews, and training sessions to convert existing pipelines into RAPs, while simultaneously upskilling the analysts. We have also developed guidance and training documentation to share internally and externally.
 Currently, out of 73 regular publications, 12 have reached the MVP, with an average score of 4.36 out of 7. This scoring is reassessed monthly, allowing us to track the progress in real-time. Self-assessment of technical skills increased by between 43% and 89% and 97% said their understanding of RAP principals improved because of the training and 77% said they are now able to implement best practice into their work. By working with the pipeline owners instead of just refactoring the code directly, we are ensuring business resilience. The in-depth knowledge of the pipeline and skills required to maintain it are present within the analysis team. Publishing our methods, documentation and tools facilitates adoption of RAP for those without a dedicated RAP team.
 We are on track to convert all our regular publications into RAPs and move to “RAP by default”, in line with the AF RAP Strategy. This will improve the reproducibility, quality, efficiency, transparency, and trustworthiness of analysis within government. We hope other organisations can learn from our methods.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134914027","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-14DOI: 10.23889/ijpds.v8i2.2351
Lisa Bunting, Nicole Gleghorne, Aideen Maguire, Sarah McKenna, Dermot O'Reilly
ObjectivesThis study uses longitudinal administrative data to investigate the relationship between area level deprivation and the 1) referral, 2) investigation, 3) registration and 4) looked-after stages of children’s contact with child and family social work in Northern Ireland (NI) from 2010-2017 (stages 1-3) and 2010-2020 (stage 4).
MethodsChildren’s social care data (SOSCARE database) for the years 2010 to 2020 were obtained from the Honest Broker Service in NI. The data were linked with the 2017 NI Multiple Deprivation Measure through the family of origin postcode. Cross-tabulations of year and deprivation decile were used to produce frequencies of children who experienced the four levels of intervention within each of the study years. These were then used to calculate various measures of absolute and relative inequality including the Slope Index of Inequality (SII), the Relative Ratio of Inequality (RRI) and the Relative Index of Inequality (RII).
ResultsThere was a clear and increasing social gradient in child welfare interventions over time. Children referred to children’s social care during 2010-2017 were 4-5 times more likely to come from the most deprived areas compared to the least deprived. Despite fairly stable levels of referral inequality, the ratio of children subject to child protection investigations rose from 3 in 2010 to 6 in 2017, the ratio of children subject to child protection plans rose from 4.5 in 2010 to 8 in 2017 and the ratio of children looked after rose from 4 in 2010 to 9 in 2020. This widening inequality was largely driven by the increasing involvement of younger children from the most deprived areas in child protection and looked-after processes.
ConclusionIn an environment of economic austerity and reduced spending, we are intervening in the lives of children and families living in the most deprived areas of NI at disproportionate rates. The current independent review of children’s social care offers an opportunity to reconfigure current provision with a clear inequalities focus.
{"title":"Changing trends in child welfare inequalities in Northern Ireland","authors":"Lisa Bunting, Nicole Gleghorne, Aideen Maguire, Sarah McKenna, Dermot O'Reilly","doi":"10.23889/ijpds.v8i2.2351","DOIUrl":"https://doi.org/10.23889/ijpds.v8i2.2351","url":null,"abstract":"ObjectivesThis study uses longitudinal administrative data to investigate the relationship between area level deprivation and the 1) referral, 2) investigation, 3) registration and 4) looked-after stages of children’s contact with child and family social work in Northern Ireland (NI) from 2010-2017 (stages 1-3) and 2010-2020 (stage 4).
 MethodsChildren’s social care data (SOSCARE database) for the years 2010 to 2020 were obtained from the Honest Broker Service in NI. The data were linked with the 2017 NI Multiple Deprivation Measure through the family of origin postcode. Cross-tabulations of year and deprivation decile were used to produce frequencies of children who experienced the four levels of intervention within each of the study years. These were then used to calculate various measures of absolute and relative inequality including the Slope Index of Inequality (SII), the Relative Ratio of Inequality (RRI) and the Relative Index of Inequality (RII).
 ResultsThere was a clear and increasing social gradient in child welfare interventions over time. Children referred to children’s social care during 2010-2017 were 4-5 times more likely to come from the most deprived areas compared to the least deprived. Despite fairly stable levels of referral inequality, the ratio of children subject to child protection investigations rose from 3 in 2010 to 6 in 2017, the ratio of children subject to child protection plans rose from 4.5 in 2010 to 8 in 2017 and the ratio of children looked after rose from 4 in 2010 to 9 in 2020. This widening inequality was largely driven by the increasing involvement of younger children from the most deprived areas in child protection and looked-after processes.
 ConclusionIn an environment of economic austerity and reduced spending, we are intervening in the lives of children and families living in the most deprived areas of NI at disproportionate rates. The current independent review of children’s social care offers an opportunity to reconfigure current provision with a clear inequalities focus.","PeriodicalId":132937,"journal":{"name":"International Journal for Population Data Science","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134913014","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：