MétaCan
Menu
Back to cohort
Record W4384929382 · doi:10.31219/osf.io/dk6zv

Regularized cross-sectional network modeling with missing data: A comparison of methods

2023· preprint· en· W4384929382 on OpenAlexafffund
Carl F. Falk, Joshua P. Starr

Bibliographic record

Venuenot available
Typepreprint
Languageen
FieldPsychology
TopicMental Health Research Topics
Canadian institutionsMcGill University
FundersNatural Sciences and Engineering Research Council of Canada
KeywordsMissing dataExpectation–maximization algorithmStructural equation modelingCovarianceGaussianCross-validationComputer scienceData miningCovariance matrixMaximizationStatisticsMaximum likelihoodAlgorithmMathematicsArtificial intelligenceMachine learningMathematical optimization

Abstract

fetched live from OpenAlex

Network modeling has emerged as a popular way to explore relationships among self-report items in psychology. Many applications of this approach involve cross-sectional data of symptom prevalence for one or more psychological disorders, and analyses are often conducted using a regularized Gaussian graphical model (GGM). Despite the close relationship between the GGM and structural equation modeling (SEM), methodology for appropriate handling of missing data is underdeveloped for the regularized GGM. Use of listwise deletion is suboptimal and precludes the possibility of a planned missing data design to reduce participant fatigue. In this research, we compare three approaches to handling missing data for the regularized GGM. The first resembles a two-stage estimation approach whereby a saturated covariance matrix among the items is estimated (e.g., using the expectation-maximization algorithm) prior to estimating the desired regularized GGM with EBIC to select the tuning parameter. The second and third approaches estimate a regularized GGM using the expectation-maximization algorithm in a single stage and either use EBIC or cross-validation. We compared these approaches in a simulation study that presumed a planned missing data design with a variety of sample sizes, proportions of missing data, and network saturation. Results suggest that the EM algorithm with cross-validation performed better, but all methods appeared to be viable strategies under larger samples and with less missing data.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.111
metaresearch head score (Gemma)0.223
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch
Consensus categoriesnone
DomainCandidate signal: Methods · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.889
Threshold uncertainty score0.589

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.1110.223
Meta-epidemiology (narrow)0.0020.002
Meta-epidemiology (broad)0.0020.004
Bibliometrics0.0030.003
Science and technology studies0.0010.001
Scholarly communication0.0020.006
Open science0.0060.004
Research integrity0.0020.004
Insufficient payload (model declined to judge)0.0040.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.703
GPT teacher head0.660
Teacher spread0.043 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

Study designSimulation or modeling
DomainMethods
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations2
Published2023
Admission routes2
Has abstractyes

Explore more

Same topicMental Health Research TopicsFrench-language works237,207