Maintained individual data distributed likelihood estimation (MIDDLE)

Steven M. Boker, Timothy Raymond Brick, Joshua N. Pritikin, Yang Wang, Timo von Oertzen, Donald Brown, John Lach, Ryne Estabrook, Michael D. Hunter, Hermine H. Maes, Michael C. Neale

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Maintained Individual Data Distributed Likelihood Estimation (MIDDLE) is a novel paradigm for research in the behavioral, social, and health sciences. The MIDDLE approach is based on the seemingly impossible idea that data can be privately maintained by participants and never revealed to researchers, while still enabling statistical models to be fit and scientific hypotheses tested. MIDDLE rests on the assumption that participant data should belong to, be controlled by, and remain in the possession of the participants themselves. Distributed likelihood estimation refers to fitting statistical models by sending an objective function and vector of parameters to each participant’s personal device (e.g., smartphone, tablet, computer), where the likelihood of that individual’s data is calculated locally. Only the likelihood value is returned to the central optimizer. The optimizer aggregates likelihood values from responding participants and chooses new vectors of parameters until the model converges. A MIDDLE study provides significantly greater privacy for participants, automatic management of opt-in and opt-out consent, lower cost for the researcher and funding institute, and faster determination of results. Furthermore, if a participant opts into several studies simultaneously and opts into data sharing, these studies automatically have access to individual-level longitudinal data linked across all studies.

Original languageEnglish (US)
Pages (from-to)706-720
Number of pages15
JournalMultivariate Behavioral Research
Volume50
Issue number6
DOIs
StatePublished - Jan 1 2015

Fingerprint

Statistical Models
Likelihood
Research Personnel
Handheld Computers
Behavioral Research
Information Dissemination
Privacy
Costs and Cost Analysis
Equipment and Supplies
Health
Statistical Model
Data Sharing
Longitudinal Data
Objective function
Choose
Paradigm
Converge
Smartphone

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Experimental and Cognitive Psychology
  • Arts and Humanities (miscellaneous)

Cite this

Boker, S. M., Brick, T. R., Pritikin, J. N., Wang, Y., von Oertzen, T., Brown, D., ... Neale, M. C. (2015). Maintained individual data distributed likelihood estimation (MIDDLE). Multivariate Behavioral Research, 50(6), 706-720. https://doi.org/10.1080/00273171.2015.1094387
Boker, Steven M. ; Brick, Timothy Raymond ; Pritikin, Joshua N. ; Wang, Yang ; von Oertzen, Timo ; Brown, Donald ; Lach, John ; Estabrook, Ryne ; Hunter, Michael D. ; Maes, Hermine H. ; Neale, Michael C. / Maintained individual data distributed likelihood estimation (MIDDLE). In: Multivariate Behavioral Research. 2015 ; Vol. 50, No. 6. pp. 706-720.
@article{0dd8e414fef84aedba506f156cb28007,
title = "Maintained individual data distributed likelihood estimation (MIDDLE)",
abstract = "Maintained Individual Data Distributed Likelihood Estimation (MIDDLE) is a novel paradigm for research in the behavioral, social, and health sciences. The MIDDLE approach is based on the seemingly impossible idea that data can be privately maintained by participants and never revealed to researchers, while still enabling statistical models to be fit and scientific hypotheses tested. MIDDLE rests on the assumption that participant data should belong to, be controlled by, and remain in the possession of the participants themselves. Distributed likelihood estimation refers to fitting statistical models by sending an objective function and vector of parameters to each participant’s personal device (e.g., smartphone, tablet, computer), where the likelihood of that individual’s data is calculated locally. Only the likelihood value is returned to the central optimizer. The optimizer aggregates likelihood values from responding participants and chooses new vectors of parameters until the model converges. A MIDDLE study provides significantly greater privacy for participants, automatic management of opt-in and opt-out consent, lower cost for the researcher and funding institute, and faster determination of results. Furthermore, if a participant opts into several studies simultaneously and opts into data sharing, these studies automatically have access to individual-level longitudinal data linked across all studies.",
author = "Boker, {Steven M.} and Brick, {Timothy Raymond} and Pritikin, {Joshua N.} and Yang Wang and {von Oertzen}, Timo and Donald Brown and John Lach and Ryne Estabrook and Hunter, {Michael D.} and Maes, {Hermine H.} and Neale, {Michael C.}",
year = "2015",
month = "1",
day = "1",
doi = "10.1080/00273171.2015.1094387",
language = "English (US)",
volume = "50",
pages = "706--720",
journal = "Multivariate Behavioral Research",
issn = "0027-3171",
publisher = "Psychology Press Ltd",
number = "6",

}

Boker, SM, Brick, TR, Pritikin, JN, Wang, Y, von Oertzen, T, Brown, D, Lach, J, Estabrook, R, Hunter, MD, Maes, HH & Neale, MC 2015, 'Maintained individual data distributed likelihood estimation (MIDDLE)', Multivariate Behavioral Research, vol. 50, no. 6, pp. 706-720. https://doi.org/10.1080/00273171.2015.1094387

Maintained individual data distributed likelihood estimation (MIDDLE). / Boker, Steven M.; Brick, Timothy Raymond; Pritikin, Joshua N.; Wang, Yang; von Oertzen, Timo; Brown, Donald; Lach, John; Estabrook, Ryne; Hunter, Michael D.; Maes, Hermine H.; Neale, Michael C.

In: Multivariate Behavioral Research, Vol. 50, No. 6, 01.01.2015, p. 706-720.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Maintained individual data distributed likelihood estimation (MIDDLE)

AU - Boker, Steven M.

AU - Brick, Timothy Raymond

AU - Pritikin, Joshua N.

AU - Wang, Yang

AU - von Oertzen, Timo

AU - Brown, Donald

AU - Lach, John

AU - Estabrook, Ryne

AU - Hunter, Michael D.

AU - Maes, Hermine H.

AU - Neale, Michael C.

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Maintained Individual Data Distributed Likelihood Estimation (MIDDLE) is a novel paradigm for research in the behavioral, social, and health sciences. The MIDDLE approach is based on the seemingly impossible idea that data can be privately maintained by participants and never revealed to researchers, while still enabling statistical models to be fit and scientific hypotheses tested. MIDDLE rests on the assumption that participant data should belong to, be controlled by, and remain in the possession of the participants themselves. Distributed likelihood estimation refers to fitting statistical models by sending an objective function and vector of parameters to each participant’s personal device (e.g., smartphone, tablet, computer), where the likelihood of that individual’s data is calculated locally. Only the likelihood value is returned to the central optimizer. The optimizer aggregates likelihood values from responding participants and chooses new vectors of parameters until the model converges. A MIDDLE study provides significantly greater privacy for participants, automatic management of opt-in and opt-out consent, lower cost for the researcher and funding institute, and faster determination of results. Furthermore, if a participant opts into several studies simultaneously and opts into data sharing, these studies automatically have access to individual-level longitudinal data linked across all studies.

AB - Maintained Individual Data Distributed Likelihood Estimation (MIDDLE) is a novel paradigm for research in the behavioral, social, and health sciences. The MIDDLE approach is based on the seemingly impossible idea that data can be privately maintained by participants and never revealed to researchers, while still enabling statistical models to be fit and scientific hypotheses tested. MIDDLE rests on the assumption that participant data should belong to, be controlled by, and remain in the possession of the participants themselves. Distributed likelihood estimation refers to fitting statistical models by sending an objective function and vector of parameters to each participant’s personal device (e.g., smartphone, tablet, computer), where the likelihood of that individual’s data is calculated locally. Only the likelihood value is returned to the central optimizer. The optimizer aggregates likelihood values from responding participants and chooses new vectors of parameters until the model converges. A MIDDLE study provides significantly greater privacy for participants, automatic management of opt-in and opt-out consent, lower cost for the researcher and funding institute, and faster determination of results. Furthermore, if a participant opts into several studies simultaneously and opts into data sharing, these studies automatically have access to individual-level longitudinal data linked across all studies.

UR - http://www.scopus.com/inward/record.url?scp=84964039174&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84964039174&partnerID=8YFLogxK

U2 - 10.1080/00273171.2015.1094387

DO - 10.1080/00273171.2015.1094387

M3 - Article

VL - 50

SP - 706

EP - 720

JO - Multivariate Behavioral Research

JF - Multivariate Behavioral Research

SN - 0027-3171

IS - 6

ER -