Overview
One sentence introduction:
I am a Lecturer in Educational Data Science, meaning I look back on the past in order to look into the future.
Slightly longer:
Motivated by social problems such as unequal access to education and inequality in health and wellbeing, I am interested in educational programmes that unbar the gate without lowering the bar and, research projects that hold promise to help solve or re-solve the problems through improved data analysis and effective storytelling.
Even longer:
I worked in Durham University as a Research Data Scientist before joining the Graduate School of Education in Exeter. In Durham, I conducted follow-up analyses of increasingly big trial data from educational interventions funded by the Education Endowment Foundation (EEF) in England. All EEF projects are independently evaluated by research teams from universities and research organisations. The data from the interventions are then deposited in a data archive which aims to accumulate research findings and help answer research questions surrounding scientific evidence. As the EEF intends to award as much as £200 million by 2025, the archive may also help us better track longer-term impact of the interventions as results from national tests become available where this is possible.
Career
Research Data Scientist, Durham University, 2015 - 2018
Research group links
Research
Research interests
As an educational data scientist with an interest in data science education, I am primarily interested in data science for causal inference, evidence generation, synthesis, and communication. In particular, I am interested in counterfactual prediction, the reproducibility of research findings, and if the results researchers reported are sensitive to analytical choices.
I collaborate with colleagues from multiple disciplines and institutions in order to solve some methodological challenges associated with social and educational interventions, for instance, the unintended consequences of randomisation in educational interventions as a result of pupils being randomly allocated to a treatment arm that might be sub-optimal, if not harmful, to them.
To achieve the above goals, I have been working on predictive and personalised approaches to evidence, particularly, individualised treatment effect in clinical trials or Pupil Advantage Index in randomised controlled trials using both conventional regression methods and machine learning techniques.
Understanding that what we observe right in front of our eyes is often ethnographically invisible, I employ mixed methods in my research and draw inspirations from medical anthropology.
Prospective PhD candidates interested in any of the following topics are welcome to get in touch:
- Causal Inference
- Data Science Education
- Digital Education
- Educational Data Science
- Evidence-Informed Policy
- Individualised Treatment Effect
- Mixed Methods Research
- Randomised Controlled Trials (RCT)
Research grants
- 2019 Exeter University
SSIS strategic discretionary research fund to estimate individualised treatment effect for randomised controlled trials using conventional regression models and machine learning algorithms. (£3,500)
- 2019 Exeter University
With Katherin Barg and Lauren Stentiford: A scoping project to automatically extract data from recorded lectures at the University of Exeter. (£2,660)
- 2019 HEFCE
Data Science Education. (£30,000)
- 2019 Exeter University
Data and Science with R. (£2,000)
- 2019 Exeter University
With Alexandra Allan, Karen Knapp, Susan McAnulla, and Matthew Newcombe: Education Incubator Project on flexible blended learning for mature students in the Graduate School of Education and Exeter Medical School. (£9,638)
- 2019 National Institute for Health Research
With Obioha Ukoumunne and Michael Nunns from the Medical School: Cluster randomised trials and child psychiatric epidemiology studentship for three years at UK/EU level. (£58,227)
Links
Publications
Key publications | Publications by category | Publications by year
Key publications
Xiao Z, Hauser OP, Kirkwood C, Jones B, Li DZ, Mo J, Higgins S (2020). Uncovering Individualised Treatment Effect: Evidence from Educational Trials.
Abstract:
Uncovering Individualised Treatment Effect: Evidence from Educational Trials
The use of large-scale Randomised Controlled Trials (RCTs) are fast becoming "the gold standard" of testing the causal effects of policy, social, and educational interventions. RCTs are typically evaluated — and ultimately judged — by the economic, educational, and statistical significance of the Average Treatment Effect (ATE) in the study sample. However, many interventions have heterogeneous treatment effects (HTEs) across different individuals, not captured by the ATE. One option to identify HTEs is to conduct subgroup analyses, such as focusing on low-income Free School Meal pupils as required for projects funded by the Education Endowment Foundation (EEF) in England. These subgroup analyses, as we demonstrate in 48 EEF-funded RCTs involving over 200,000 students, are usually not standardised across studies and offer flexible degrees of freedom to researchers, potentially leading to mixed results. Here, we develop and deploy a machine-learning and regression-based framework for systematic estimation of Individualised Treatment Effects, which can show where a seemingly ineffective and uninformative intervention worked, for whom, and by how much. Our findings have implications for policy-makers in education, public health, and other government services.
Abstract.
Author URL.
Xiao Z (2019). Mobile phones as life and thought companions.
RESEARCH PAPERS IN EDUCATION Author URL.
Full text.
Xiao ZM (2019). ‘You Are Too Out!’: a mixed methods study of the ways in which digital divides articulate status and power in China.
Information DevelopmentAbstract:
‘You Are Too Out!’: a mixed methods study of the ways in which digital divides articulate status and power in China
© the Author(s) 2019. This study investigates the differences in adolescent engagement with Information and Communication Technologies (ICT), such as computers, the Internet, and mobile phones. Involving 698 second-year high school students from urban, rural, and ethnic Tibetan regions of China, it finds that patterns of access and use indicate status and power, and the meanings teenagers pour into the technologies articulate social and educational differences. On average, Tibetans are disadvantaged in access, and the return on parental education is greater for the mainstream Han than it is for Tibetans. However, state ‘preferential policies’ have mitigated Tibetans’ plight in use, which makes the least privileged to be Han students with parents having no more than six years of education.
Abstract.
Full text.
Xiao Z, Higgins S (2018). The power of noise and the art of prediction.
INTERNATIONAL JOURNAL OF EDUCATIONAL RESEARCH,
87, 36-46.
Author URL.
Full text.
Xiao ZM, Higgins S, Kasim A (2017). An Empirical Unraveling of Lord's Paradox.
Journal of Experimental Education,
87(1), 17-32.
Abstract:
An Empirical Unraveling of Lord's Paradox
© 2017, © 2017 Taylor. &. Francis Group, LLC. Lord's Paradox occurs when a continuous covariate is statistically controlled for and the relationship between a continuous outcome and group status indicator changes in both magnitude and direction. This phenomenon poses a challenge to the notion of evidence-based policy, where data are supposed to be self-evident. We examined 50 effect size estimates from 34 large-scale educational interventions and found that impact estimates are affected in magnitude, with or without reversal in sign, when there is substantial baseline imbalance. We also demonstrated that multilevel modeling can ameliorate the divergence in sign and/or magnitude of effect estimation, which, together with project specific knowledge, promises to help those who are presented with conflicting or confusing evidence in decision-making.
Abstract.
Full text.
Xiao Z, Kasim A, Higgins S (2016). Same difference? Understanding variation in the estimation of effect sizes from educational trials.
INTERNATIONAL JOURNAL OF EDUCATIONAL RESEARCH,
77, 1-14.
Author URL.
Full text.
Publications by category
Journal articles
Xiao Z (2019). Mobile phones as life and thought companions.
RESEARCH PAPERS IN EDUCATION Author URL.
Full text.
Xiao ZM (2019). ‘You Are Too Out!’: a mixed methods study of the ways in which digital divides articulate status and power in China.
Information DevelopmentAbstract:
‘You Are Too Out!’: a mixed methods study of the ways in which digital divides articulate status and power in China
© the Author(s) 2019. This study investigates the differences in adolescent engagement with Information and Communication Technologies (ICT), such as computers, the Internet, and mobile phones. Involving 698 second-year high school students from urban, rural, and ethnic Tibetan regions of China, it finds that patterns of access and use indicate status and power, and the meanings teenagers pour into the technologies articulate social and educational differences. On average, Tibetans are disadvantaged in access, and the return on parental education is greater for the mainstream Han than it is for Tibetans. However, state ‘preferential policies’ have mitigated Tibetans’ plight in use, which makes the least privileged to be Han students with parents having no more than six years of education.
Abstract.
Full text.
Xiao Z, Higgins S (2018). The power of noise and the art of prediction.
INTERNATIONAL JOURNAL OF EDUCATIONAL RESEARCH,
87, 36-46.
Author URL.
Full text.
Xiao ZM, Higgins S, Kasim A (2017). An Empirical Unraveling of Lord's Paradox.
Journal of Experimental Education,
87(1), 17-32.
Abstract:
An Empirical Unraveling of Lord's Paradox
© 2017, © 2017 Taylor. &. Francis Group, LLC. Lord's Paradox occurs when a continuous covariate is statistically controlled for and the relationship between a continuous outcome and group status indicator changes in both magnitude and direction. This phenomenon poses a challenge to the notion of evidence-based policy, where data are supposed to be self-evident. We examined 50 effect size estimates from 34 large-scale educational interventions and found that impact estimates are affected in magnitude, with or without reversal in sign, when there is substantial baseline imbalance. We also demonstrated that multilevel modeling can ameliorate the divergence in sign and/or magnitude of effect estimation, which, together with project specific knowledge, promises to help those who are presented with conflicting or confusing evidence in decision-making.
Abstract.
Full text.
Xiao Z, Kasim A, Higgins S (2016). Same difference? Understanding variation in the estimation of effect sizes from educational trials.
INTERNATIONAL JOURNAL OF EDUCATIONAL RESEARCH,
77, 1-14.
Author URL.
Full text.
Chapters
Xiao Z, Higgins S (2014). When English Meets Chinese in Tibetan Schools: Towards an Understanding of Multilingual Education in Tibet. In Feng A, Adamson B (Eds.)
Trilingualism in Education in China: Models and Challenges, Dordrecht: Springer, 117-140.
Abstract:
When English Meets Chinese in Tibetan Schools: Towards an Understanding of Multilingual Education in Tibet
Abstract.
Author URL.
Reports
Kasim A, Xiao Z, Higgings S, De Troyer E (2017).
eefAnalytics: Analysing Education Trials., CRAN.
Abstract:
eefAnalytics: Analysing Education Trials
Abstract.
Author URL.
Full text.
Hutton C, Lukes S, Abdulkader MS, Choudhury R, Gulaid A, Sadia S, Xiao Z, Zere A (2015).
Trusting the dice: immigration advice in Tower Hamlets., Toynbee Hall.
Author URL.
Higgins S, Xiao Z, Katsipataki M (2012).
The impact of digital technology on learning: a summary for the Education Endowment Foundation., Education Endowment Foundation.
Author URL.
Publications by year
2020
Xiao Z, Hauser OP, Kirkwood C, Jones B, Li DZ, Mo J, Higgins S (2020). Uncovering Individualised Treatment Effect: Evidence from Educational Trials.
Abstract:
Uncovering Individualised Treatment Effect: Evidence from Educational Trials
The use of large-scale Randomised Controlled Trials (RCTs) are fast becoming "the gold standard" of testing the causal effects of policy, social, and educational interventions. RCTs are typically evaluated — and ultimately judged — by the economic, educational, and statistical significance of the Average Treatment Effect (ATE) in the study sample. However, many interventions have heterogeneous treatment effects (HTEs) across different individuals, not captured by the ATE. One option to identify HTEs is to conduct subgroup analyses, such as focusing on low-income Free School Meal pupils as required for projects funded by the Education Endowment Foundation (EEF) in England. These subgroup analyses, as we demonstrate in 48 EEF-funded RCTs involving over 200,000 students, are usually not standardised across studies and offer flexible degrees of freedom to researchers, potentially leading to mixed results. Here, we develop and deploy a machine-learning and regression-based framework for systematic estimation of Individualised Treatment Effects, which can show where a seemingly ineffective and uninformative intervention worked, for whom, and by how much. Our findings have implications for policy-makers in education, public health, and other government services.
Abstract.
Author URL.
2019
Xiao Z (2019). Mobile phones as life and thought companions.
RESEARCH PAPERS IN EDUCATION Author URL.
Full text.
Xiao ZM (2019). ‘You Are Too Out!’: a mixed methods study of the ways in which digital divides articulate status and power in China.
Information DevelopmentAbstract:
‘You Are Too Out!’: a mixed methods study of the ways in which digital divides articulate status and power in China
© the Author(s) 2019. This study investigates the differences in adolescent engagement with Information and Communication Technologies (ICT), such as computers, the Internet, and mobile phones. Involving 698 second-year high school students from urban, rural, and ethnic Tibetan regions of China, it finds that patterns of access and use indicate status and power, and the meanings teenagers pour into the technologies articulate social and educational differences. On average, Tibetans are disadvantaged in access, and the return on parental education is greater for the mainstream Han than it is for Tibetans. However, state ‘preferential policies’ have mitigated Tibetans’ plight in use, which makes the least privileged to be Han students with parents having no more than six years of education.
Abstract.
Full text.
2018
Xiao Z, Higgins S (2018). The power of noise and the art of prediction.
INTERNATIONAL JOURNAL OF EDUCATIONAL RESEARCH,
87, 36-46.
Author URL.
Full text.
2017
Xiao ZM, Higgins S, Kasim A (2017). An Empirical Unraveling of Lord's Paradox.
Journal of Experimental Education,
87(1), 17-32.
Abstract:
An Empirical Unraveling of Lord's Paradox
© 2017, © 2017 Taylor. &. Francis Group, LLC. Lord's Paradox occurs when a continuous covariate is statistically controlled for and the relationship between a continuous outcome and group status indicator changes in both magnitude and direction. This phenomenon poses a challenge to the notion of evidence-based policy, where data are supposed to be self-evident. We examined 50 effect size estimates from 34 large-scale educational interventions and found that impact estimates are affected in magnitude, with or without reversal in sign, when there is substantial baseline imbalance. We also demonstrated that multilevel modeling can ameliorate the divergence in sign and/or magnitude of effect estimation, which, together with project specific knowledge, promises to help those who are presented with conflicting or confusing evidence in decision-making.
Abstract.
Full text.
Kasim A, Xiao Z, Higgings S, De Troyer E (2017).
eefAnalytics: Analysing Education Trials., CRAN.
Abstract:
eefAnalytics: Analysing Education Trials
Abstract.
Author URL.
Full text.
2016
Xiao Z, Kasim A, Higgins S (2016). Same difference? Understanding variation in the estimation of effect sizes from educational trials.
INTERNATIONAL JOURNAL OF EDUCATIONAL RESEARCH,
77, 1-14.
Author URL.
Full text.
2015
Hutton C, Lukes S, Abdulkader MS, Choudhury R, Gulaid A, Sadia S, Xiao Z, Zere A (2015).
Trusting the dice: immigration advice in Tower Hamlets., Toynbee Hall.
Author URL.
2014
Xiao Z, Higgins S (2014). When English Meets Chinese in Tibetan Schools: Towards an Understanding of Multilingual Education in Tibet. In Feng A, Adamson B (Eds.)
Trilingualism in Education in China: Models and Challenges, Dordrecht: Springer, 117-140.
Abstract:
When English Meets Chinese in Tibetan Schools: Towards an Understanding of Multilingual Education in Tibet
Abstract.
Author URL.
2012
Higgins S, Xiao Z, Katsipataki M (2012).
The impact of digital technology on learning: a summary for the Education Endowment Foundation., Education Endowment Foundation.
Author URL.
zhimin_xiao Details from cache as at 2020-02-22 00:47:29
Refresh publications
External Engagement and Impact
Committee/panel activities
Emerging Applications Section of the Royal Statistical Society: Committee member.
Invited lectures
Uncovering Individualised Treatment Effect: Evidence from Educational Trials, Department of Computer Science, University of Bath, March 3rd, 2020.
Latest working paper presented at the Health Statistics Group, College of Medicine and Health, University of Exeter, January 7th, 2020.
The case against perfection in the mean: Why it is time for an individualised approach to evidence for education (Q-Step Centre, University of Exeter), November 8th, 2019.
The Case Against Perfection in the Mean: From Average to Individualised Treatment Effect in Randomised Controlled Trials for Education (Royal Statistical Society), September 23rd, 2019.
Pupil Advantage Index as an Alternative to Subgroup Analysis in RCTs for Education (Royal Statistical Society South West Local Group), February 13th, 2019.
Journal and book series Editorships and Editorial board membership
Associate Editor for the Digital Impacts Section of Frontiers in Human Dynamics
Teaching
I contribute to teaching and supervision of the following modules:
2019 - 2020:
(EFPM308) Preparing for Educational Research and Dissertation.
(EFPM913) Debating the Big Questions in Education.
(SSI2001) Learning from Work Experience (Undergraduate).
(EFPM005Z1) Preparing for Educational Enquiry.
2018 - 2019:
(EFPM308) Preparing for Educational Research and Dissertation.
(ERPM005) Designing and Communicating Research.
(SSI2001) Learning from Work Experience (Undergraduate).
(EFPM329a) Preparing for TESOL Enquiry and Dissertation.
(ERPM100) MSc Programme Dissertation Supervision.
Data and Science with R.
Supervision / Group
Postgraduate researchers
- Amal Alammar
- Abdallah Alharbi
- Thana Aljumaah
- Norah Alosayl
- Angela Short
- Julio Cesar Torres Rocha
- Yiwen Wang
