Specifying the random effect structure in linear mixed effect models for analyzing psycholinguistic data
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Linear mixed-effect models (LMEM) have become a popular method for analyzing nested experimental data, which are often encountered in psycholinguistics and other fields.This approach allows for experimental results to be generalized to the greater population of both subjects and experimental stimuli.The difference between LMEM and basic multiple regression is the inclusion of random effects, which model the variance attributable to random sampling.In an influential paper, Barr et al. (2013) recommend specifying the maximal random effect structure allowed by the experimental design, which means including random intercepts and random slopes for all within-subjects and within-items experimental factors, as well as allowing correlations between within-unit random effects.But anecdotally, maximal LMEMs are prone to model non-convergence, which is the failure of the model estimation algorithm to reach a solution.The goal of this thesis is thus to formally investigate the occurrence of model non-convergence in LMEMs with different random effect structures and different numbers of predictors through a simulation study.We also evaluate the use of information criteria (IC), specifically AIC and BIC, in selecting the best-fitting model.The results show that complex models (i.e., with more parameters) lead to a dramatic increase in the non-convergence rate, but only in cases of model overfitting.Furthermore, AIC and BIC were found to select the true model from four candidate models in the majority of cases, although selection accuracy varied by LMEM random effect structure.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.041 | 0.411 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.003 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.009 | 0.001 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it