pith. sign in

arxiv: 2605.23048 · v1 · pith:XACBYEFQnew · submitted 2026-05-21 · 💻 cs.HC · cs.CY· stat.AP· stat.ME

StanBKT: Rethinking Parameter Estimation in Bayesian Knowledge Tracing

Siddhartha Pradhan , Yanping Pei , Morgan Lee , Puyuan Zhang , Erin Ottmar , Adam C. Sales This is my paper

Pith reviewed 2026-05-25 05:11 UTC · model grok-4.3

classification 💻 cs.HC cs.CYstat.APstat.ME
keywords Bayesian Knowledge Tracingparameter estimationeducational data miningBayesian inferencestudent modelinghidden Markov modelsuncertainty quantification
0
0 comments X

The pith

StanBKT replaces point estimates with Bayesian posteriors for parameters in student knowledge tracing models.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces StanBKT, a Python package that estimates Bayesian Knowledge Tracing models through probabilistic methods such as Hamiltonian Monte Carlo rather than optimization routines that return only single values. This setup keeps the original hidden Markov model intact while adding support for uncertainty in parameters, hierarchical structures across learners, and direct comparisons between experimental conditions. A reader would care because it turns learning, forgetting, guessing, and slipping rates into distributions instead of fixed numbers, allowing more careful interpretation of educational interventions. The authors demonstrate this on large datasets including ASSISTments 2020, where the new methods match traditional predictive accuracy but add posterior diagnostics.

Core claim

StanBKT supplies a unified framework for BKT estimation that supports Hamiltonian Monte Carlo, variational inference, Pathfinder, and optimization while preserving the hidden Markov structure, and it enables standard, grouped, and hierarchical model variants along with posterior predictive checks.

What carries the argument

StanBKT, a Python package that interfaces with Stan to perform Bayesian inference on BKT parameters with flexible priors and utilities for visualization.

If this is right

  • Posterior distributions allow direct, uncertainty-aware comparisons of learning and forgetting rates across experimental conditions.
  • Hierarchical modeling becomes feasible for sharing information across students while retaining individual parameter estimates.
  • Posterior predictive inference supports checks on model adequacy beyond point predictions.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar Bayesian wrappers could be written for other hidden Markov models used in educational data mining.
  • Adaptive tutoring systems might incorporate parameter uncertainty when deciding when to advance a student to new material.

Load-bearing premise

The inference methods keep the hidden Markov structure of classical BKT and produce posteriors that faithfully represent uncertainty in the parameters.

What would settle it

If posterior predictive performance on held-out data falls below that of expectation-maximization fits or if condition-specific parameter differences lose statistical support once uncertainty intervals are examined.

read the original abstract

Bayesian Knowledge Tracing (BKT) is a widely used and interpretable student modeling approach in intelligent tutoring systems and educational data mining. However, most implementations rely on expectation-maximization or related optimization methods that yield only point estimates, limiting uncertainty quantification and principled comparisons across learners and conditions. We introduce StanBKT, an open-source Python package for estimating BKT models using Bayesian inference in Stan. StanBKT provides a unified framework supporting Hamiltonian Monte Carlo, variational inference, Pathfinder, and optimization-based estimation while preserving the hidden Markov structure and interpretability of classical BKT. It supports standard, grouped, and hierarchical BKT models, flexible prior specification, posterior predictive inference, and utilities for visualization and diagnostics. We evaluate StanBKT on large-scale observational and controlled educational datasets. On the ASSISTments 2020 dataset, we show that supported inference methods achieve comparable predictive performance while differing in computational efficiency and posterior fidelity. We further demonstrate how posterior inference enables principled comparison of condition-specific parameters in an educational intervention involving perceptual cue manipulations. Results illustrate how uncertainty quantification facilitates more reliable interpretation of differences in learning, forgetting, guessing, and slipping parameters across experimental conditions. Overall, StanBKT extends BKT beyond point estimation by providing a flexible framework for probabilistic student modeling, uncertainty quantification, and hierarchical inference in educational data mining.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper introduces StanBKT, an open-source Python package implementing Bayesian inference (HMC, variational inference, Pathfinder, and optimization) for standard, grouped, and hierarchical BKT models in Stan while preserving the classical hidden-Markov structure. It supports flexible priors, posterior predictive checks, and visualization utilities. Evaluations on the ASSISTments 2020 dataset report comparable predictive performance across methods with differences in efficiency and posterior fidelity; a second analysis applies the framework to an educational intervention to compare condition-specific learning, forgetting, guessing, and slipping parameters via uncertainty quantification.

Significance. If the approximation methods preserve exact HMM marginal likelihoods and deliver calibrated posteriors, the package would provide a practical, reproducible tool for moving BKT beyond point estimates, enabling hierarchical modeling and principled cross-condition inference in educational data mining. The open-source release with diagnostics is a concrete strength for the field.

major comments (2)
  1. [Abstract and §4] Abstract and §4 (ASSISTments 2020 evaluation): the claim that methods differ in 'posterior fidelity' while achieving comparable predictive performance is not accompanied by any quantitative verification (e.g., posterior calibration on synthetic data with known parameters, or direct numerical comparison of the Stan marginal likelihood against the classical forward-algorithm likelihood). This check is load-bearing for the central claim that the framework supports reliable uncertainty quantification and hierarchical inference.
  2. [§3] §3 (model implementation): the statement that all supported inference methods 'preserve the hidden Markov structure' requires explicit confirmation that the discrete-state marginalization remains exact under VI and Pathfinder; without a derivation or numerical equivalence test against the forward algorithm, the fidelity advantage over EM remains unverified.
minor comments (2)
  1. The manuscript would benefit from an appendix containing the core Stan model blocks (or a link to the exact model files in the repository) to allow readers to verify the HMM implementation directly.
  2. Figure captions and axis labels in the intervention analysis should explicitly state the posterior intervals (e.g., 95% HDI) used for the condition comparisons.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments, which help clarify the presentation of our contributions. We address each major comment below.

read point-by-point responses

  1. Referee: [Abstract and §4] Abstract and §4 (ASSISTments 2020 evaluation): the claim that methods differ in 'posterior fidelity' while achieving comparable predictive performance is not accompanied by any quantitative verification (e.g., posterior calibration on synthetic data with known parameters, or direct numerical comparison of the Stan marginal likelihood against the classical forward-algorithm likelihood). This check is load-bearing for the central claim that the framework supports reliable uncertainty quantification and hierarchical inference.

    Authors: We agree that quantitative verification of posterior fidelity is needed to support the central claims. In the revised manuscript we will add synthetic-data experiments (with known ground-truth parameters) to §4, reporting posterior calibration metrics such as credible-interval coverage and a direct numerical comparison of Stan marginal likelihoods against the forward algorithm on small instances. revision: yes

  2. Referee: [§3] §3 (model implementation): the statement that all supported inference methods 'preserve the hidden Markov structure' requires explicit confirmation that the discrete-state marginalization remains exact under VI and Pathfinder; without a derivation or numerical equivalence test against the forward algorithm, the fidelity advantage over EM remains unverified.

    Authors: The Stan model code implements exact marginalization over discrete states for the likelihood; this specification is identical for HMC, VI, Pathfinder and optimization. We acknowledge that the manuscript currently lacks an explicit derivation and numerical test. In revision we will add both a short derivation in §3 and numerical equivalence checks (Stan log marginal likelihood vs. forward algorithm) on small models. revision: yes

Circularity Check

0 steps flagged

No circularity: StanBKT is a software implementation of existing BKT models

full rationale

The paper describes an open-source Python package implementing Bayesian inference (HMC, VI, Pathfinder) for standard BKT hidden-Markov models. No derivation chain, parameter fitting, or prediction step is presented that reduces to a quantity defined or fitted within the same paper. Claims rest on preservation of classical BKT structure and empirical evaluation on external datasets (ASSISTments 2020), with no self-definitional equations, fitted-input predictions, or load-bearing self-citations. This is a normal non-finding for an implementation paper.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only; no free parameters, axioms, or invented entities are described. The work implements existing BKT structure in a new inference engine rather than introducing new mathematical entities.

pith-pipeline@v0.9.0 · 5787 in / 1024 out tokens · 23347 ms · 2026-05-25T05:11:14.595595+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

63 extracted references · 63 canonical work pages · 4 internal anchors

  1. [1]

    Stan Reference Manual , year =

    work page

  2. [2]

    and Corbett, Albert T

    Anderson, John R. and Corbett, Albert T. and Koedinger, Kenneth R. and Pelletier,. Cognitive. Journal of the Learning Sciences , volume =. doi:10.1207/s15327809jls0402_2 , urldate =

    work page doi:10.1207/s15327809jls0402_2

  3. [3]

    and Boyle, C

    Anderson, John R. and Boyle, C. Franklin and Reiser, Brian J. , year = 1985, month = apr, journal =. Intelligent. doi:10.1126/science.228.4698.456 , urldate =

    work page doi:10.1126/science.228.4698.456 1985

  4. [4]

    Optimizing

    Badrinath, Anirudhan and Pardos, Zachary , year = 2025, month = jan, journal =. Optimizing. doi:10.5281/zenodo.14707987 , urldate =

    work page doi:10.5281/zenodo.14707987 2025

  5. [5]

    Baker, Ryan S

    d. Baker, Ryan S. J. and Pardos, Zachary A. and Gowda, Sujith M. and Nooraei, Bahador B. and Heffernan, Neil T. , editor =. Ensembling. User. doi:10.1007/978-3-642-22362-4_2 , abstract =

    work page doi:10.1007/978-3-642-22362-4_2

  6. [6]

    Baker, Ryan S

    d. Baker, Ryan S. J. and Corbett, Albert T. and Aleven, Vincent , editor =. More. Intelligent. doi:10.1007/978-3-540-69132-7_44 , abstract =

    work page doi:10.1007/978-3-540-69132-7_44

  7. [7]

    and Chang, Kai-min and Mostow, Jack and Corbett, Albert , editor =

    Beck, Joseph E. and Chang, Kai-min and Mostow, Jack and Corbett, Albert , editor =. Does. Intelligent. doi:10.1007/978-3-540-69132-7_42 , abstract =

    work page doi:10.1007/978-3-540-69132-7_42

  8. [8]

    Desirable

    Bjork, Robert and Bjork, Elizabeth , year = 2020, month = dec, journal =. Desirable. doi:10.1016/j.jarmac.2020.09.003 , file =

    work page doi:10.1016/j.jarmac.2020.09.003 2020

  9. [9]

    Bulut, Okan and Shin, Jinnie and. An. Psych , volume =. doi:10.3390/psych5030050 , abstract =

    work page doi:10.3390/psych5030050

  10. [10]

    Chang, Kai-min and Beck, Joseph and Mostow, Jack and Corbett, Albert , editor =. A. Intelligent. doi:10.1007/11774303_11 , urldate =

    work page doi:10.1007/11774303_11

  11. [11]

    and Chan, Yun-Chen and Smith, Hannah and Ottmar, Erin R

    Closser, Avery H. and Chan, Yun-Chen and Smith, Hannah and Ottmar, Erin R. , year = 2022, month = sep, journal =. Perceptual Learning in Math:

    work page 2022

  12. [12]

    and Anderson, John R

    Corbett, Albert T. and Anderson, John R. , year = 1994, month = dec, journal =. Knowledge Tracing:. doi:10.1007/BF01099821 , urldate =

    work page doi:10.1007/bf01099821 1994

  13. [13]

    Analyzing

    Cui, Yang and Chu, Man-Wai and Chen, Fu , year = 2019, month = jun, journal =. Analyzing

    work page 2019

  14. [14]

    Gong, Yue and Rai, Dovan and Beck, Joseph E and Heffernan, Neil T , year = 2009, journal =. Does

    work page 2009

  15. [15]

    and Heffernan, Cristina Lindquist , year = 2014, month = dec, journal =

    Heffernan, Neil T. and Heffernan, Cristina Lindquist , year = 2014, month = dec, journal =. The. doi:10.1007/s40593-014-0024-x , urldate =

    work page doi:10.1007/s40593-014-0024-x 2014

  16. [16]

    Expert Systems with Applications , volume =

    Student Modeling and Assessment in Intelligent Tutoring of Software Patterns , author =. Expert Systems with Applications , volume =. doi:10.1016/j.eswa.2011.07.010 , urldate =

    work page doi:10.1016/j.eswa.2011.07.010 2011

  17. [17]

    Kass, Robert , editor =. Student. User. doi:10.1007/978-3-642-83230-7_14 , urldate =

    work page doi:10.1007/978-3-642-83230-7_14

  18. [18]

    Kobsa, Alfred , year = 1989, series =. User

    work page 1989

  19. [19]

    Koedinger, Kenneth R and Pavlik, Phillip and McLaren, Bruce M and Aleven, Vincent , year = 2008, abstract =. Is It

    work page 2008

  20. [20]

    Quarterly Journal of Experimental Psychology , volume =

    Proximity and Precedence in Arithmetic , author =. Quarterly Journal of Experimental Psychology , volume =. doi:10.1080/17470211003787619 , abstract =

    work page doi:10.1080/17470211003787619

  21. [21]

    Lin, Chien-Chang and Huang, Anna Y. Q. and Lu, Owen H. T. , year = 2023, month = aug, journal =. Artificial Intelligence in Intelligent Tutoring Systems toward Sustainable Education: A Systematic Review , shorttitle =. doi:10.1186/s40561-023-00260-y , urldate =

    work page doi:10.1186/s40561-023-00260-y 2023

  22. [22]

    Lu, Yu and Chen, Chen and Chen, Penghe and Chen, Xiyang and Zhuang, Zijun , editor =. Smart. Artificial. doi:10.1007/978-3-319-93846-2_84 , abstract =

    work page doi:10.1007/978-3-319-93846-2_84

  23. [23]

    Predicting Students' Performance on Intelligent Tutoring System -

    Nedungadi, Prema and Remya, M.s , year = 2015, month = feb, volume =. Predicting Students' Performance on Intelligent Tutoring System -. doi:10.1109/FIE.2014.7044200 , abstract =

    work page doi:10.1109/fie.2014.7044200 2015

  24. [24]

    and Heffernan, Neil T

    Pardos, Zachary A. and Heffernan, Neil T. , editor =. User. doi:10.1007/978-3-642-22362-4_21 , abstract =

    work page doi:10.1007/978-3-642-22362-4_21

  25. [25]

    and Heffernan, Neil T

    Pardos, Zachary A. and Heffernan, Neil T. , editor =. Modeling. User. doi:10.1007/978-3-642-13470-8_24 , abstract =

    work page doi:10.1007/978-3-642-13470-8_24

  26. [26]

    and Tang, Matthew and Anastasopoulos, Ioannis and Sheel, Shreya K

    Pardos, Zachary A. and Tang, Matthew and Anastasopoulos, Ioannis and Sheel, Shreya K. and Zhang, Ethan , year = 2023, month = apr, series =. Proceedings of the 2023. doi:10.1145/3544548.3581574 , urldate =

    work page doi:10.1145/3544548.3581574 2023

  27. [27]

    Incorporating

    San Pedro, Michael and Baker, Ryan and Gobert, Janice , year = 2013, month = jan, abstract =. Incorporating. Proceedings of the 6th

    work page 2013

  28. [28]

    Adaptive

    Pel. Adaptive. International Journal of Artificial Intelligence in Education , volume =. doi:10.1007/s40593-024-00400-6 , urldate =

    work page doi:10.1007/s40593-024-00400-6

  29. [29]

    Bayesian Knowledge Tracing, Logistic Models, and beyond: An Overview of Learner Modeling Techniques , shorttitle =

    Pel. Bayesian Knowledge Tracing, Logistic Models, and beyond: An Overview of Learner Modeling Techniques , shorttitle =. User Modeling and User-Adapted Interaction , volume =. doi:10.1007/s11257-017-9193-2 , urldate =

    work page doi:10.1007/s11257-017-9193-2

  30. [30]

    Conceptual

    Pel. Conceptual. Artificial. doi:10.1007/978-3-319-93843-1_33 , abstract =

    work page doi:10.1007/978-3-319-93843-1_33

  31. [31]

    User Modeling and User-Adapted Interaction , volume =

    Leveraging Machine-Learned Detectors of Systematic Inquiry Behavior to Estimate and Predict Transfer of Inquiry Skill , author =. User Modeling and User-Adapted Interaction , volume =. doi:10.1007/s11257-011-9101-0 , urldate =

    work page doi:10.1007/s11257-011-9101-0

  32. [32]

    User Modeling and User-Adapted Interaction , volume =

    Twenty-Five Years of. User Modeling and User-Adapted Interaction , volume =. doi:10.1007/s11257-023-09389-4 , urldate =

    work page doi:10.1007/s11257-023-09389-4

  33. [33]

    Adaptive

    Schodde, Thorsten and Bergmann, Kirsten and Kopp, Stefan , year = 2017, month = mar, series =. Adaptive. Proceedings of the 2017. doi:10.1145/2909824.3020222 , urldate =

    work page doi:10.1145/2909824.3020222 2017

  34. [34]

    Learning Analytics and Educational Data Mining:

    Siemens, George and Baker, Ryan , year = 2012, month = apr, journal =. Learning Analytics and Educational Data Mining:. doi:10.1145/2330601.2330661 , abstract =

    work page doi:10.1145/2330601.2330661 2012

  35. [35]

    Takami, Kyosuke and Flanagan, Brendan and Dai, Yiling , year = 2021, month = nov, journal =. Toward

    work page 2021

  36. [36]

    Evaluating the

    Takami, Kyosuke and Flanagan, Brendan and Dai, Yiling and Ogata, Hiroaki , year = 2024, month = feb, journal =. Evaluating the. doi:10.4018/ijdet.337600 , abstract =

    work page doi:10.4018/ijdet.337600 2024

  37. [37]

    Examining the Applications of Intelligent Tutoring Systems in Real Educational Contexts:

    Wang, Huanhuan and Tlili, Ahmed and Huang, Ronghuai and Cai, Zhenyu and Li, Min and Cheng, Zui and Yang, Dong and Li, Mengti and Zhu, Xixian and Fei, Cheng , year = 2023, month = jan, journal =. Examining the Applications of Intelligent Tutoring Systems in Real Educational Contexts:. doi:10.1007/s10639-022-11555-x , urldate =

    work page doi:10.1007/s10639-022-11555-x 2023

  38. [38]

    Xu, Sheng and Sun, Manfang and Fang, Weili and Chen, Ke and Luo, Hanbin and Zou, Patrick X. W. , year = 2023, month = mar, journal =. A. doi:10.1016/j.dibe.2022.100111 , urldate =

    work page doi:10.1016/j.dibe.2022.100111 2023

  39. [39]

    Xu, Yanbo and Mostow, Jack , year = 2013, abstract =. Using

    work page 2013

  40. [40]

    Yang, Chunsheng and Chiang, Feng-Kuang and Cheng, Qiangqiang and Ji, Jun , year = 2021, month = oct, journal =. Machine. doi:10.1177/0735633120986256 , urldate =

    work page doi:10.1177/0735633120986256 2021

  41. [41]

    and Koedinger, Kenneth R

    Yudelson, Michael V. and Koedinger, Kenneth R. and Gordon, Geoffrey J. , editor =. Individualized. Artificial. doi:10.1007/978-3-642-39112-5_18 , abstract =

    work page doi:10.1007/978-3-642-39112-5_18

  42. [42]

    , year = 2024, month = mar, series =

    Zambrano, Andres Felipe and Zhang, Jiayi and Baker, Ryan S. , year = 2024, month = mar, series =. Investigating. Proceedings of the 14th. doi:10.1145/3636555.3636890 , urldate =

    work page doi:10.1145/3636555.3636890 2024

  43. [43]

    Journal of Statistical Software , author =

    Stan:. Journal of Statistical Software , author =. 2017 , keywords =. doi:10.18637/jss.v076.i01 , abstract =

    work page doi:10.18637/jss.v076.i01 2017

  44. [44]

    Journal of the American statistical Association , volume=

    Variational inference: A review for statisticians , author=. Journal of the American statistical Association , volume=. 2017 , publisher=

    work page 2017

  45. [45]

    Journal of the royal statistical society: series B (methodological) , volume=

    Maximum likelihood from incomplete data via the EM algorithm , author=. Journal of the royal statistical society: series B (methodological) , volume=. 1977 , publisher=

    work page 1977

  46. [46]

    A Conceptual Introduction to Hamiltonian Monte Carlo

    A conceptual introduction to Hamiltonian Monte Carlo , author=. arXiv preprint arXiv:1701.02434 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  47. [47]

    A Gelman, G Jones XL Meng , volume=

    Mcmc using hamiltonian dynamics (handbook of markov chain monte carlo) ed s brooks et al , author=. A Gelman, G Jones XL Meng , volume=

    work page

  48. [48]

    Advances in neural information processing systems , author=

    Deep knowledge tracing. Advances in neural information processing systems , author=. Association for Computing Machinery , pages=

    work page

  49. [49]

    Journal of Machine Learning Research , volume=

    Pathfinder: Parallel quasi-Newton variational inference , author=. Journal of Machine Learning Research , volume=

    work page

  50. [50]

    Statistical and Computational Guarantees for the Baum-Welch Algorithm

    Yang, Fanny and Balakrishnan, Sivaraman and Wainwright, Martin J. , month = dec, year =. Statistical and. doi:10.48550/arXiv.1512.08269 , abstract =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1512.08269

  51. [51]

    Proceedings of the 14th

    Badrinath, Anirudhan and Wang, Frederic and Pardos, Zach , month = jun, year =. Proceedings of the 14th

    work page

  52. [52]

    2026 , keywords =

    Behavior Research Methods , author =. 2026 , keywords =. doi:10.3758/s13428-026-02955-9 , abstract =

    work page doi:10.3758/s13428-026-02955-9 2026

  53. [53]

    User Modeling and User-Adapted Interaction , author =

    A multifactor approach to student model evaluation , volume =. User Modeling and User-Adapted Interaction , author =. 2008 , keywords =. doi:10.1007/s11257-007-9046-5 , abstract =

    work page doi:10.1007/s11257-007-9046-5 2008

  54. [54]

    Pandey, Shalini and Karypis, George , month = jul, year =. A. doi:10.48550/arXiv.1907.06837 , abstract =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1907.06837 1907

  55. [55]

    The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo

    Hoffman, Matthew D. and Gelman, Andrew , month = nov, year =. The. doi:10.48550/arXiv.1111.4246 , abstract =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1111.4246

  56. [56]

    Journal of Machine Learning Research , author =

    Automatic. Journal of Machine Learning Research , author =. 2017 , pages =

    work page 2017

  57. [57]

    Prihar, Ethan and Botelho, Anthony and Yuen, Joseph and Corace, Mike and Shanaj, Andrew and Dai, Zekun and Heffernan, Neil , month = apr, year =. Student. Companion

    work page

  58. [58]

    Montero, Shirly and Arora, Akshit and Kelly, Sean and Milne, Brent and Mozer, Michael , month = jul, year =. Does

    work page

  59. [59]

    ArXiv , author =

    How. ArXiv , author =

    work page

  60. [60]

    and Abril-Pla, Oriol and Deklerk, Jordan and Axen, Seth D

    Martin, Osvaldo A. and Abril-Pla, Oriol and Deklerk, Jordan and Axen, Seth D. and Carroll, Colin and Hartikainen, Ari and Vehtari, Aki , year =. Journal of Open Source Software , publisher =. doi:10.21105/joss.09889 , number =

    work page doi:10.21105/joss.09889

  61. [61]

    Journal of Open Psychology Data , author =

    Data from the. Journal of Open Psychology Data , author =. doi:10.5334/jopd.139 , abstract =

    work page doi:10.5334/jopd.139

  62. [62]

    Incorporating scaffolding and tutor context into bayesian knowledge tracing to predict inquiry skill acquisition , booktitle =

    Sao Pedro, Michael and Baker, Ryan and Gobert, Janice , year =. Incorporating scaffolding and tutor context into bayesian knowledge tracing to predict inquiry skill acquisition , booktitle =

    work page

  63. [63]

    Journal of Educational Data Mining , author =

    Using. Journal of Educational Data Mining , author =. 2026 , note =

    work page 2026