Thumbnail for Raw data and analysis code for the study “Project 2025 as a technocratic blueprint: A corpus-based linguistic analysis of conservative governance discourse”

Raw data and analysis code for the study “Project 2025 as a technocratic blueprint: A corpus-based linguistic analysis of conservative governance discourse”

Name: Raw data and analysis code for the study “Project 2025 as a technocratic blueprint: A corpus-based linguistic analysis of conservative governance discourse”
Published: 2025-11-11
License: http://creativecommons.org/licenses/by/4.0

Open this dataset in the live repository

Important usage condition: Use and redistribution are governed by CC BY 4.0. Review and comply with the license before using the data.

Persistent identifier: doi:10.60507/FK2/BK14F8
Published version: 1.0
Publication date: 2025-11-11
License: CC BY 4.0

Description

This repository contains all data and scripts used for the study “Project 2025 as a Technocratic Blueprint: A Corpus-Based Linguistic Analysis of Conservative Governance Discourse” (Schilling & Fuchs, 2025). The study investigates the language of the Heritage Foundation’s Project 2025, a 900-page conservative policy blueprint, using methods from Corpus-Assisted Discourse Studies (CADS), Political Discourse Analysis (PDA), and psycholinguistic text analysis. The corpus includes Project 2025 and Democratic and Republican Party platforms (2016–2024). The dataset includes: Raw data (/data/raw_data/): full texts of Project 2025 and party platforms in CSV format. Processed data (/data/raw_data/Project2025_lemmaPOS.csv): tokenized, lemmatized, and POS-tagged text. Keyness results (/data/keyness/): unigram and bigram keyness calculations (log-likelihood, log-ratio). Collocation results (/data/collocations/): top 10 adjective, noun, and verb collocates per node. LIWC results (/data/liwc/): LIWC-22 category scores for each corpus. Analysis scripts (/code/): R Markdown file (analysis_project2025.Rmd) and two Python scripts for collocation and keyness analysis. All files are in UTF-8 plain-text format. The dataset contains no personal, sensitive, or proprietary data and derives entirely from publicly accessible political documents.

Creators

Schilling, Julia
Fuchs, Robert

Keywords

Project 2025, Heritage Foundation, political discourse, policy communication, corpus linguistics, discourse analysis, collocation analysis, keyness, LIWC

Files

Before downloading: Use and redistribution are governed by CC BY 4.0. Review and comply with the license before using the data.

File	Type	Bytes	Checksum
keyness_2gram_pos_manifesto_rep_overuse.csv	text/csv	13103889	MD5 fc2cee9dd32ff05bc68acaabc9a260a7
collocation_analysis.py	text/x-python-script	20533	MD5 bb3e93a961f46a78fe49671abfec56ee
P2025_colloc_ADJ_w7_top10_per_node.csv	text/csv	7007	MD5 35f5564343e279ba2f9687dc354fd63e
keyness_1gram_pos_manifesto_dem_overuse.csv	text/csv	1057511	MD5 c96c120d14dadcb655181244fbc64de3
P2025_colloc_VERB_w7_top10_per_node.csv	text/csv	7038	MD5 fec75c16cb53fca40d7c5ed5f807e9c4
keyness_1gram_pos_manifesto_rep_overuse.csv	text/csv	1058835	MD5 206e1a07b39133218c461ccf7a4082f4
keyness_analysis.py	text/x-python-script	9106	MD5 fdba36fc4bbe3e9d0daf93a0cd6fc599
keyness_2gram_pos_manifesto_dem_overuse.csv	text/csv	12827129	MD5 220625d115ab7262e2764383921ff8f6
P2025_colloc_NOUN_w7_top10_per_node.csv	text/csv	7110	MD5 ac2033a4ca681c8e26ca0ab893bd955c
Platforms_Democrats_LIWC_subset.csv	text/csv	916161	MD5 1a33450f5ce0ac9dd0e34d60ed0c350f
Project2025_LIWC_subset.csv	text/csv	2310611	MD5 75c059d442c8a215e95d4dded4b186c8
Platforms_Republicans_LIWC_subset.csv	text/csv	330359	MD5 641ccc8365b1afd101375b50a1a86a3a
analysis_project2025.Rmd	text/x-r-notebook	26326	MD5 136d84af88837928bbedb7225e6bf0bd
Project2025.csv	text/csv	2154423	MD5 4662cb4b2b1301d0d40c83d077959ea7
Platforms_Democrats.csv	text/csv	846946	MD5 aabcf833be06563a10bc99671e8f02f8
Platforms_Republicans.csv	text/csv	303972	MD5 a18aca13011427bccdb4c70b35e8608d
Project2025_lemmaPOS.csv	text/csv	4200254	MD5 24a16b6e1b93862c3a3975bca7d55e5d
README_Project2025.md	text/markdown	9188	MD5 2e5e8f4560c10cd97b8f04adcbe4e023

Citation

Schilling, Julia; Fuchs, Robert, 2025-11-11, Raw data and analysis code for the study “Project 2025 as a technocratic blueprint: A corpus-based linguistic analysis of conservative governance discourse”, doi:10.60507/FK2/BK14F8, V1.0

Additional Dataverse fields

Id	517
Dataset Type	dataset
Internal Version Number	31
Latest Version Publishing State	RELEASED
Release Time	2025-11-11T12:47:28Z
Create Time	2025-11-05T19:06:46Z
Citation Date	2025-11-11
File Access Request	True

Export metadata

Static metadata exports available for this published dataset version:

Complete Dataverse metadata

Citation Metadata

Title title

Raw data and analysis code for the study “Project 2025 as a technocratic blueprint: A corpus-based linguistic analysis of conservative governance discourse”

Author author

Author Name
Schilling, Julia
Author Affiliation
Universität Bonn
Author Identifier Scheme
ORCID
Author Identifier
Type Name
authorIdentifier
Multiple
False
Type Class
primitive
Value
https://orcid.org/0009-0005-4105-1415
Expandedvalue
Person Name
Schilling, Julia
@id
https://orcid.org/0009-0005-4105-1415
Scheme
ORCID
@type
https://schema.org/Person
Author Name
Fuchs, Robert
Author Affiliation
Universität Bonn
Author Identifier Scheme
ORCID
Author Identifier
Type Name
authorIdentifier
Multiple
False
Type Class
primitive
Value
https://orcid.org/0000-0001-7694-062X
Expandedvalue
Person Name
Fuchs, Robert
@id
https://orcid.org/0000-0001-7694-062X
Scheme
ORCID
@type
https://schema.org/Person

Dataset Contact datasetContact

Dataset Contact Name
Schilling, Julia
Dataset Contact Affiliation
Universität Bonn
Dataset Contact Email
julia.schilling@uni-bonn.de

Ds Description dsDescription

Ds Description Value
This repository contains all data and scripts used for the study “Project 2025 as a Technocratic Blueprint: A Corpus-Based Linguistic Analysis of Conservative Governance Discourse” (Schilling & Fuchs, 2025). The study investigates the language of the Heritage Foundation’s Project 2025, a 900-page conservative policy blueprint, using methods from Corpus-Assisted Discourse Studies (CADS), Political Discourse Analysis (PDA), and psycholinguistic text analysis. The corpus includes Project 2025 and Democratic and Republican Party platforms (2016–2024). The dataset includes: Raw data (/data/raw_data/): full texts of Project 2025 and party platforms in CSV format. Processed data (/data/raw_data/Project2025_lemmaPOS.csv): tokenized, lemmatized, and POS-tagged text. Keyness results (/data/keyness/): unigram and bigram keyness calculations (log-likelihood, log-ratio). Collocation results (/data/collocations/): top 10 adjective, noun, and verb collocates per node. LIWC results (/data/liwc/): LIWC-22 category scores for each corpus. Analysis scripts (/code/): R Markdown file (analysis_project2025.Rmd) and two Python scripts for collocation and keyness analysis. All files are in UTF-8 plain-text format. The dataset contains no personal, sensitive, or proprietary data and derives entirely from publicly accessible political documents.

Subject subject

Arts and Humanities

Subject Refinement subjectRefinement

Subject Refinement Value
Linguistics: Applied Linguistics, Experimental Linguistics, Computational Linguistics [104-04]

Free Keyword freeKeyword

Free Keyword Value
Project 2025
Free Keyword Value
Heritage Foundation
Free Keyword Value
political discourse
Free Keyword Value
policy communication
Free Keyword Value
corpus linguistics
Free Keyword Value
discourse analysis
Free Keyword Value
collocation analysis
Free Keyword Value
keyness
Free Keyword Value
LIWC

Publication publication

Publication Relation Type
IsSupplementTo
Publication Citation
Schilling, Julia & Fuchs, Robert (2025). Project 2025 as a Technocratic Blueprint: A Corpus-Based Linguistic Analysis of Conservative Governance Discourse. Submitted manuscript, PLOS ONE.

Language language

English

Depositor depositor

Schilling, Julia

Date Of Deposit dateOfDeposit

2025-11-05

Software software

Software Name
spaCy
Software Version
3.7
Software Name
R
Software Version
4.4.1
Software Name
RStudio
Software Version
2024.09.0+375
Software Name
Python
Software Version
3.11.8
Software Name
LIWC
Software Version
22

Software Metadata (CodeMeta v2.0)

Code Version codeVersion

1.0

Development Status developmentStatus

Active

Application Category applicationCategory

Text and corpus analysis

Application Sub Category applicationSubCategory

Computational linguistics; political discourse analysis

Programming Language programmingLanguage

R; Python

Runtime Platform runtimePlatform

RStudio (≥2024.04) with R 4.4.1; Python 3.11 interpreter with spaCy 3.7.2

Operating System operatingSystem

macOS 14; Windows 10; Ubuntu 22.04 (tested)

Target Product targetProduct

Corpus-based linguistic analysis of Project 2025 (policy discourse)

Software Requirements Item softwareRequirementsItem

Software Requirements
spaCy 3.7.2
Software Requirements Info Url
https://spacy.io
Software Requirements
quanteda 4.1
Software Requirements Info Url
https://quanteda.io
Software Requirements
tidyverse 2.0
Software Requirements Info Url
https://www.tidyverse.org
Software Requirements
pandas 2.2
Software Requirements Info Url
https://pandas.pydata.org
Software Requirements
numpy 1.26
Software Requirements Info Url
https://numpy.org

Software Suggestions Item softwareSuggestionsItem

Software Suggestions
LIWC-22
Software Suggestions Info Url
https://www.liwc.app

Memory Requirements memoryRequirements

< 4 GB RAM

Processor Requirements processorRequirements

Standard multi-core CPU

Storage Requirements storageRequirements

< 500 MB (including raw text, collocations, and LIWC outputs)

Expected crawler behaviour

Use a stable, truthful User-Agent with product/version and a working contact URL. Across all IP addresses and HTTP connections used by one crawler identity, allow no more than 5 requests in flight and wait at least 20 seconds between request starts. Crawl URLs listed in the catalog sitemap, including file pages and download URLs when they are published, use conditional requests, honor Retry-After, and apply exponential backoff after errors.

The welcome page may link to the interactive repository for human navigation. Automated clients must not treat that human link as a catalog crawl target.

Read the live machine-readable crawler policy before and during a crawl. Stop crawling when it reports CPU or memory utilization at or above 80% and 80% respectively.