pith. sign in

arxiv: 2607.01464 · v1 · pith:V2RDYJTWnew · submitted 2026-07-01 · 💻 cs.CL

Comparing Architectures for Supervised Political Scaling

Anna Golub , Sebastian Pad\'o This is my paper

Pith reviewed 2026-07-03 21:00 UTC · model grok-4.3

classification 💻 cs.CL
keywords political text scalingideological positioningjoint predictionhybrid modelssupervised NLPclassification regressiontext analysis
0
0 comments X

The pith

Joint prediction of ideological scales outperforms individual predictions and creates a middle ground between classification and regression.

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

The paper consolidates supervised methods for positioning political actors on ideological scales from text by testing whether joint prediction across multiple scales improves results compared to handling each scale separately. It further examines whether hybrid models can effectively combine elements of classification and regression approaches. These comparisons aim to identify stronger architectures for reducing manual effort in political text analysis. A reader would care because more accurate automated scaling could support larger-scale studies of political discourse without proportional increases in human labeling.

Core claim

Predicting scales jointly rather than individually improves performance on political scaling tasks, and hybrid models occupy a useful middle ground between pure classification and regression methods.

What carries the argument

Joint multi-scale prediction architectures and hybrid classification-regression models applied to ideological text scaling.

If this is right

  • Joint models reduce error rates compared to separate scale predictions on the evaluated datasets.
  • Hybrid architectures achieve performance between classification and regression baselines.
  • The approach lowers the annotation burden for multi-dimensional ideological analysis.
  • Architecture choice matters more when the number of scales increases.

Where Pith is reading between the lines

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

  • The joint prediction benefit might extend to other multi-label or multi-dimensional NLP tasks beyond politics.
  • Testing these models on streaming social media could reveal whether they support dynamic, real-time scaling.
  • The hybrid middle ground invites exploration of loss functions that blend discrete and continuous targets in adjacent domains like sentiment or stance detection.

Load-bearing premise

Standard evaluation metrics and datasets for political scaling are adequate to detect meaningful differences between joint prediction, hybrid models, and baseline architectures.

What would settle it

A replication on multiple held-out political text datasets where joint models show no accuracy gain over separate predictions would falsify the main performance claim.

Figures

Figures reproduced from arXiv: 2607.01464 by Anna Golub, Sebastian Pad\'o.

Figure 1
Figure 1. Figure 1: Experimental setup: Computational ap￾proaches to scaling and research questions. views that complement RILE’s economic perspective. Still, to our knowledge, there is no work on predicting GAL-TAN positions. What is more, despite being orthogonal in theory, the empirical RILE and GAL-TAN scores are often found to be interdependent. Left-wing parties gravitate toward GAL social policies, and right-wing views… view at source ↗
Figure 2
Figure 2. Figure 2: Left: Confusion matrix heatmap: Label aggregation RILE individual prediction with SBERT. Right: Predicted vs. true scatterplot: Chunk-level RILE regression with BigBird (test set, random seed 7). ModernBERT underperforms BigBird somewhat, however this may be a consequence of our choice to use the maximum chunk size (twice as high for ModernBERT as for BigBird), cf. Section 4.3. Error Analysis. While the ma… view at source ↗
Figure 3
Figure 3. Figure 3: SBERT learning curves: joint multitask learn [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Joint chunk-level regression with Modern [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Classification-regression continuum: RILE true vs. predicted manifesto scores scatterplot (test set, random [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
read the original abstract

Text scaling, the task of positioning political actors on an ideological scale, is a fundamental task in political analysis. To ease the need for manual analysis, various NLP methods have been proposed for this task, including classification- and regression-based approaches, showing successes as well as limitations. The goal of our paper is to consolidate the state of the art in this area. We ask two questions: (a) Can the performance of scaling methods be improved by predicting scales not individually but jointly? (b) Is there a middle ground between classification and regression?

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 / 1 minor

Summary. The manuscript consolidates the state of the art in supervised text scaling for positioning political actors on ideological scales. It poses two empirical questions: (a) whether predicting scales jointly rather than individually improves performance, and (b) whether hybrid architectures can occupy a middle ground between classification-based and regression-based approaches.

Significance. If the experiments demonstrate clear gains from joint prediction and identify effective hybrids, the work would offer actionable architecture recommendations for computational political analysis, building on existing classification and regression baselines in NLP for ideology detection.

major comments (2)
  1. [Abstract] Abstract and introduction: the two research questions are posed but the manuscript provides no quantitative results, error analysis, or dataset details in the supplied abstract; without the methods/results sections it is impossible to evaluate whether joint prediction or hybrids outperform baselines on standard political scaling metrics.
  2. The weakest assumption noted in the reader report—that standard evaluation metrics and datasets suffice to detect meaningful differences—remains untestable here; if the full paper relies on the same metrics without ablation on metric sensitivity, this would be load-bearing for claims about improvement.
minor comments (1)
  1. Notation for 'scales' and 'joint prediction' should be defined early with reference to prior work on multi-task vs. single-task scaling.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their review. We address the major comments point by point below, drawing on the full manuscript content.

read point-by-point responses

  1. Referee: [Abstract] Abstract and introduction: the two research questions are posed but the manuscript provides no quantitative results, error analysis, or dataset details in the supplied abstract; without the methods/results sections it is impossible to evaluate whether joint prediction or hybrids outperform baselines on standard political scaling metrics.

    Authors: The abstract is a concise overview of the research questions only. The full manuscript contains a methods section detailing the datasets, architectures, and standard political scaling metrics, along with a results section that reports quantitative performance comparisons, including error analysis for joint prediction versus individual scaling and for hybrid models versus pure classification or regression baselines. revision: partial

  2. Referee: The weakest assumption noted in the reader report—that standard evaluation metrics and datasets suffice to detect meaningful differences—remains untestable here; if the full paper relies on the same metrics without ablation on metric sensitivity, this would be load-bearing for claims about improvement.

    Authors: The manuscript applies the same standard metrics and datasets used in prior political scaling work to all models, enabling direct relative comparisons. No explicit ablation on metric sensitivity is present; we can add a limitations discussion on this point in revision. revision: partial

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper poses two open empirical questions about joint vs. individual scale prediction and the existence of a middle ground between classification and regression. No derivations, equations, fitted parameters renamed as predictions, or self-citation chains are present in the provided abstract or described structure. The work is a comparative evaluation on standard datasets and metrics, with no load-bearing step that reduces to its own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review reveals no free parameters, non-standard axioms, or invented entities; relies on standard supervised NLP assumptions for political text.

pith-pipeline@v0.9.1-grok · 5604 in / 875 out tokens · 24133 ms · 2026-07-03T21:00:19.009138+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

68 extracted references · 27 canonical work pages · 3 internal anchors

  1. [1]

    Smith , year=

    Anna Brigevich and William B. Smith , year=. Proceedings of the European Union Studies Association Biannual Conference , address=

  2. [2]

    West European Politics , volume =

    Anna-Sophie Kurella and Milena Rapp , title =. West European Politics , volume =. 2026 , publisher =

    2026

  3. [3]

    Proceedings of the ACM Conference on Fairness, Accountability, and Transparency , year = 2024, pages =

    Liesenfeld, Andreas and Dingemanse, Mark , title =. Proceedings of the ACM Conference on Fairness, Accountability, and Transparency , year = 2024, pages =. doi:10.1145/3630106.3659005 , numpages = 14, address =

    work page doi:10.1145/3630106.3659005 2024

  4. [4]

    WIREs Data Mining and Knowledge Discovery , volume =

    Alva Principe, Renzo and Chiarini, Nicola and Viviani, Marco , title =. WIREs Data Mining and Knowledge Discovery , volume =. doi:https://doi.org/10.1002/widm.70019 , url =. https://wires.onlinelibrary.wiley.com/doi/pdf/10.1002/widm.70019 , note =

    work page doi:10.1002/widm.70019

  5. [5]

    , title =

    Koedam, Jelle and Binding, Garret and Steenbergen, Marco R. , title =. British Journal of Political Science , volume =. 2025 , publisher =. doi:10.1017/S0007123424000474 , url =

    work page doi:10.1017/s0007123424000474 2025

  6. [6]

    The 2024 Chapel Hill Expert Survey on political party positioning in Europe: Twenty-five years of party positional data , journal =

    Jan Rovny and Jonathan Polk and Ryan Bakker and Liesbet Hooghe and Seth Jolly and Gary Marks and Marco Steenbergen and Milada Anna Vachudova , keywords =. The 2024 Chapel Hill Expert Survey on political party positioning in Europe: Twenty-five years of party positional data , journal =. 2025 , issn =. doi:https://doi.org/10.1016/j.electstud.2025.102981 , url =

    work page doi:10.1016/j.electstud.2025.102981 2024

  7. [7]

    Party Policy in Modern Democracies , isbn =

    Benoit, Kenneth and Laver, Michael , year =. Party Policy in Modern Democracies , isbn =. doi:10.4324/9780203028179 , address=

    work page doi:10.4324/9780203028179

  8. [8]

    2013 , url =

    Budge, Ian , title =. 2013 , url =

    2013

  9. [9]

    and Best, Robin and Franzmann, Simon , isbn =

    Volkens, Andrea and Bara, Judith and Budge, Ian and McDonald, Michael D. and Best, Robin and Franzmann, Simon , isbn =. Understanding and Validating the Left-Right Scale (RILE) , booktitle =. 2013 , month =. doi:10.1093/acprof:oso/9780199640041.003.0006 , url =

    work page doi:10.1093/acprof:oso/9780199640041.003.0006 2013

  10. [10]

    2025 , note =

    Yu, Jun and Liu, Xiaokang and Luo, Chongliang and Zhou, Rong and Liu, Yixin and Hu, Jie and Chen, Jianmin and Zhang, Ke and Zhang, Dazheng and Shen, Yishan and Adhikarla, Eashan and Dai, Yutong and Zhang, Kai and Kong, Zhaoming and Ye, Wenxuan and Yin, Yilong and Namboodiri, Vinod and Davison, Brian and Moore, Jason and Chen, Yong , journal =. 2025 , note =

    2025

  11. [11]

    Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

    Facenet: A unified embedding for face recognition and clustering , author=. Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

  12. [12]

    BERT : Pre-training of Deep Bidirectional Transformers for Language Understanding

    Devlin, Jacob and Chang, Ming-Wei and Lee, Kenton and Toutanova, Kristina. BERT : Pre-training of Deep Bidirectional Transformers for Language Understanding. Proceedings of the 2019 Conference of the North A merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). 2019. doi:10.18653/v...

    work page doi:10.18653/v1/n19-1423 2019

  13. [13]

    European Journal of Political Research , volume =

    Dassonneville, Ruth and Hooghe, Liesbet and Marks, Gary , title =. European Journal of Political Research , volume =. doi:https://doi.org/10.1111/1475-6765.12590 , url =. https://ejpr.onlinelibrary.wiley.com/doi/pdf/10.1111/1475-6765.12590 , year =

    work page doi:10.1111/1475-6765.12590

  14. [14]

    Multilingual estimation of political-party positioning: From label aggregation to long-input Transformers

    Nikolaev, Dmitry and Ceron, Tanise and Pad \'o , Sebastian. Multilingual estimation of political-party positioning: From label aggregation to long-input Transformers. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 2023. doi:10.18653/v1/2023.emnlp-main.591

    work page doi:10.18653/v1/2023.emnlp-main.591 2023

  15. [15]

    American political science review , volume=

    Extracting policy positions from political texts using words as data , author=. American political science review , volume=. 2003 , publisher=

    2003

  16. [16]

    American Journal of Political Science , volume=

    A scaling model for estimating time-series party positions from texts , author=. American Journal of Political Science , volume=. 2008 , publisher=

    2008

  17. [17]

    Additive manifesto decomposition: A policy domain aware method for understanding party positioning

    Ceron, Tanise and Nikolaev, Dmitry and Pad \'o , Sebastian. Additive manifesto decomposition: A policy domain aware method for understanding party positioning. Findings of the Association for Computational Linguistics: ACL 2023. 2023. doi:10.18653/v1/2023.findings-acl.499

    work page doi:10.18653/v1/2023.findings-acl.499 2023

  18. [18]

    Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference

    Warner, Benjamin and Chaffin, Antoine and Clavi. Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025. doi:10.18653/v1/2025.acl-long.127

    work page doi:10.18653/v1/2025.acl-long.127 2025

  19. [19]

    Advances in neural information processing systems , volume=

    Big bird: Transformers for longer sequences , author=. Advances in neural information processing systems , volume=

  20. [20]

    Sentence- BERT : Sentence Embeddings using S iamese BERT -Networks

    Reimers, Nils and Gurevych, Iryna. Sentence- BERT : Sentence Embeddings using S iamese BERT -Networks. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019. doi:10.18653/v1/D19-1410

    work page doi:10.18653/v1/d19-1410 2019

  21. [21]

    Large Language Models: A Survey

    Shervin Minaee and Tomas Mikolov and Narjes Nikzad and Meysam Chenaghlu and Richard Socher and Xavier Amatriain and Jianfeng Gao , year=. 2402.06196 , archivePrefix=

    work page internal anchor Pith review Pith/arXiv arXiv

  22. [22]

    Unsupervised Cross-Lingual Scaling of Political Texts

    Glava s , Goran and Nanni, Federico and Ponzetto, Simone Paolo. Unsupervised Cross-Lingual Scaling of Political Texts. Proceedings of the 15th Conference of the E uropean Chapter of the Association for Computational Linguistics: Volume 2, Short Papers. 2017

    2017

  23. [23]

    Political Analysis , author=

    Word Embeddings for the Analysis of Ideological Placement in Parliamentary Corpora , volume=. Political Analysis , author=. 2020 , pages=. doi:10.1017/pan.2019.26 , number=

    work page doi:10.1017/pan.2019.26 2020

  24. [24]

    Political Text Scaling Meets Computational Semantics , year =

    Nanni, Federico and Glava. Political Text Scaling Meets Computational Semantics , year =. ACM/IMS Trans. Data Sci. , month = may, articleno =. doi:10.1145/3485666 , abstract =

    work page doi:10.1145/3485666

  25. [25]

    Optimizing text representations to capture (dis)similarity between political parties

    Ceron, Tanise and Blokker, Nico and Pad \'o , Sebastian. Optimizing text representations to capture (dis)similarity between political parties. Proceedings of the 26th Conference on Computational Natural Language Learning (CoNLL). 2022. doi:10.18653/v1/2022.conll-1.22

    work page doi:10.18653/v1/2022.conll-1.22 2022

  26. [26]

    and Zhang, Jiawei , title =

    Zhang, Haopeng and Yu, Philip S. and Zhang, Jiawei , title =. ACM Comput. Surv. , month = jun, articleno =. 2025 , issue_date =. doi:10.1145/3731445 , abstract =

    work page doi:10.1145/3731445 2025

  27. [27]

    ACM Comput

    Yu, Fei and Zhang, Hongbo and Tiwari, Prayag and Wang, Benyou , title =. ACM Comput. Surv. , month = oct, articleno =. 2024 , issue_date =. doi:10.1145/3664194 , abstract =

    work page doi:10.1145/3664194 2024

  28. [28]

    2021 , eprint=

    Whitening Sentence Representations for Better Semantics and Faster Retrieval , author=. 2021 , eprint=

    2021

  29. [29]

    International Conference on Learning Representations , year=

    Decoupled Weight Decay Regularization , author=. International Conference on Learning Representations , year=

  30. [30]

    How Contextual are Contextualized Word Representations? C omparing the Geometry of BERT , ELM o, and GPT -2 Embeddings

    Ethayarajh, Kawin. How Contextual are Contextualized Word Representations? C omparing the Geometry of BERT , ELM o, and GPT -2 Embeddings. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019. doi:10.18653/v1/D19-1006

    work page doi:10.18653/v1/d19-1006 2019

  31. [31]

    MPNet: Masked and Permuted Pre-training for Language Understanding , url =

    Song, Kaitao and Tan, Xu and Qin, Tao and Lu, Jianfeng and Liu, Tie-Yan , booktitle =. MPNet: Masked and Permuted Pre-training for Language Understanding , url =

  32. [32]

    Why Comparing Single Performance Scores Does Not Allow to Draw Conclusions About Machine Learning Approaches

    Why comparing single performance scores does not allow to draw conclusions about machine learning approaches , author=. arXiv preprint arXiv:1803.09578 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  33. [33]

    International conference on machine learning , pages=

    Learning transferable visual models from natural language supervision , author=. International conference on machine learning , pages=. 2021 , organization=

    2021

  34. [34]

    Advances in neural information processing systems , volume=

    Attention is all you need , author=. Advances in neural information processing systems , volume=

  35. [35]

    Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) , pages=

    Know what you don’t know: Unanswerable questions for SQuAD , author=. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) , pages=

  36. [36]

    Proceedings of the 2018 EMNLP workshop BlackboxNLP: Analyzing and interpreting neural networks for NLP , pages=

    GLUE: A multi-task benchmark and analysis platform for natural language understanding , author=. Proceedings of the 2018 EMNLP workshop BlackboxNLP: Analyzing and interpreting neural networks for NLP , pages=

    2018

  37. [37]

    Learning Thematic Similarity Metric from Article Sections Using Triplet Networks

    Dor, Liat Ein and Mass, Yosi and Halfon, Alon and Venezian, Elad and Shnayderman, Ilya and Aharonov, Ranit and Slonim, Noam. Learning Thematic Similarity Metric from Article Sections Using Triplet Networks. Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). 2018. doi:10.18653/v1/P18-2009

    work page doi:10.18653/v1/p18-2009 2018

  38. [38]

    A Robustly Optimized BERT Pre-training Approach with Post-training

    Zhuang, Liu and Wayne, Lin and Ya, Shi and Jun, Zhao. A Robustly Optimized BERT Pre-training Approach with Post-training. Proceedings of the 20th Chinese National Conference on Computational Linguistics. 2021

    2021

  39. [39]

    Pengcheng He and Jianfeng Gao and Weizhu Chen , booktitle=. De. 2023 , url=

    2023

  40. [40]

    mGTE : Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval

    Zhang, Xin and Zhang, Yanzhao and Long, Dingkun and Xie, Wen and Dai, Ziqi and Tang, Jialong and Lin, Huan and Yang, Baosong and Xie, Pengjun and Huang, Fei and Zhang, Meishan and Li, Wenjie and Zhang, Min. mGTE : Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval. Proceedings of the 2024 Conference on Empiri...

    work page doi:10.18653/v1/2024.emnlp-industry.103 2024

  41. [41]

    Improving Neural Political Statement Classification with Class Hierarchical Information

    Dayanik, Erenay and Blessing, Andre and Blokker, Nico and Haunss, Sebastian and Kuhn, Jonas and Lapesa, Gabriella and Pado, Sebastian. Improving Neural Political Statement Classification with Class Hierarchical Information. Findings of the Association for Computational Linguistics: ACL 2022. 2022. doi:10.18653/v1/2022.findings-acl.186

    work page doi:10.18653/v1/2022.findings-acl.186 2022

  42. [42]

    2025 , booktitle =

    Wu, Xinyi and Wang, Yifei and Jegelka, Stefanie and Jadbabaie, Ali , title =. 2025 , booktitle =

    2025

  43. [43]

    American Political Science Review , author=

    A Pairwise Comparison Framework for Fast, Flexible, and Reliable Human Coding of Political Texts , volume=. American Political Science Review , author=. 2017 , pages=. doi:10.1017/S0003055417000302 , number=

    work page doi:10.1017/s0003055417000302 2017

  44. [44]

    Advances in neural information processing systems , volume=

    Flashattention: Fast and memory-efficient exact attention with io-awareness , author=. Advances in neural information processing systems , volume=

  45. [45]

    Qbaibi, Samia , title =

  46. [46]

    Party Politics , volume=

    Scaling hand-coded political texts to learn more about left-right policy content , author=. Party Politics , volume=. 2022 , publisher=

    2022

  47. [47]

    Political Science Research and Methods , author=

    What’s in a buzzword? A systematic review of the state of populism research in political science , volume=. Political Science Research and Methods , author=. 2022 , pages=. doi:10.1017/psrm.2021.44 , number=

    work page doi:10.1017/psrm.2021.44 2022

  48. [48]

    Efficient Estimation of Word Representations in Vector Space

    Efficient estimation of word representations in vector space , author=. arXiv preprint arXiv:1301.3781 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  49. [49]

    International conference on machine learning , pages=

    Distributed representations of sentences and documents , author=. International conference on machine learning , pages=. 2014 , organization=

    2014

  50. [50]

    Unsupervised Cross-lingual Representation Learning at Scale

    Conneau, Alexis and Khandelwal, Kartikay and Goyal, Naman and Chaudhary, Vishrav and Wenzek, Guillaume and Guzm \'a n, Francisco and Grave, Edouard and Ott, Myle and Zettlemoyer, Luke and Stoyanov, Veselin. Unsupervised Cross-lingual Representation Learning at Scale. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. ...

    work page doi:10.18653/v1/2020.acl-main.747 2020

  51. [51]

    2020 , eprint=

    Longformer: The Long-Document Transformer , author=. 2020 , eprint=

    2020

  52. [52]

    European security , volume=

    The party politics of the EU’s relations with the USA: evidence from the European Parliament , author=. European security , volume=. 2021 , publisher=

    2021

  53. [53]

    2021 , publisher=

    Political Ideologies: An Introduction , author=. 2021 , publisher=

    2021

  54. [54]

    2004 , publisher=

    European integration and political conflict , author=. 2004 , publisher=

    2004

  55. [55]

    American Journal of Political Science , pages=

    Putting parties in their place: Inferring party left-right ideological positions from party manifestos data , author=. American Journal of Political Science , pages=. 2000 , publisher=

    2000

  56. [56]

    Party politics , volume=

    Locating political parties in policy space: A reanalysis of party manifesto data , author=. Party politics , volume=. 2006 , publisher=

    2006

  57. [57]

    Party Politics , volume=

    Conceptualizing Left and Right in comparative politics: Towards a deductive approach , author=. Party Politics , volume=. 2011 , publisher=

    2011

  58. [58]

    Party Politics , volume=

    The multidimensional nature of party competition , author=. Party Politics , volume=. 2010 , publisher=

    2010

  59. [59]

    Comparative European Politics , volume=

    The changing relevance and meaning of left and right in 34 party systems from 1945 to 2020 , author=. Comparative European Politics , volume=. 2023 , publisher=

    1945

  60. [60]

    American Journal of Political Science , volume=

    Treating words as data with error: Uncertainty in text statements of policy positions , author=. American Journal of Political Science , volume=. 2009 , publisher=

    2009

  61. [61]

    Estimating the Policy Position of Political Actors , pages=

    Estimating policy positions from the computer coding of political texts: results from Italy, the Netherlands and Ireland , author=. Estimating the Policy Position of Political Actors , pages=. 2003 , publisher=

    2003

  62. [62]

    1994 , publisher=

    An introduction to the bootstrap , author=. 1994 , publisher=

    1994

  63. [63]

    American Journal of Political Science , year=

    Using large language models to analyze political texts through natural language understanding , author=. American Journal of Political Science , year=

  64. [64]

    Political Science Research and Methods , volume=

    How to train your stochastic parrot: Large language models for political texts , author=. Political Science Research and Methods , volume=. 2025 , publisher=

    2025

  65. [65]

    Political Analysis , volume=

    Positioning political texts with large language models by asking and averaging , author=. Political Analysis , volume=. 2025 , publisher=

    2025

  66. [66]

    2026 , eprint=

    Olmo 3 , author=. 2026 , eprint=

    2026

  67. [67]

    1992 , publisher=

    Party policy and government coalitions , author=. 1992 , publisher=

    1992

  68. [68]

    Political analysis , volume=

    Coder reliability and misclassification in the human coding of party manifestos , author=. Political analysis , volume=. 2012 , publisher=

    2012