方法对比

	Wordfish Scaling	Wordfish	Wordscores
领域	Political Science	心理测量学	心理测量学
方法族	Latent structure	Latent structure	Latent structure
起源年份	2008	2008	2003
提出者	Jonathan Slapin and Sven-Oliver Proksch	Jonathan Slapin, Svenja-Sophia Proksch	Michael Laver, Kenneth Benoit, John Garry
类型	Unsupervised latent-position model for word-count data	Generative text model for dimension reduction	Text analysis and dimension reduction
开创性文献	Slapin, J. B., & Proksch, S.-O. (2008). A Scaling Model for Estimating Time-Series Party Positions from Texts. American Journal of Political Science, 52(3), 705–722.	Slapin, J. B., & Proksch, S. O. (2008). A scaling model for estimating time-series party positions from texts. Journal of Politics, 70(3), 554-569.	Laver, M., Benoit, K., & Garry, J. (2003). Extracting policy positions from political texts using words as data. American Political Science Review, 97(2), 311-331.
别名	Wordfish text scaling, Poisson scaling of texts, Unsupervised text scaling, Wordfish position estimation	—	—
相关	4	5	5
摘要	Wordfish scaling is an unsupervised text-as-data method that estimates a single latent position for each political document — a party manifesto, a legislative speech, a press release — directly from its word frequencies, without any reference texts or hand coding. Introduced by Slapin and Proksch in 2008, it models word counts as draws from a Poisson distribution whose rate depends on a document position and word-specific parameters, recovering, for example, a left–right ordering of parties purely from how often each word appears in each text.	Wordfish is a statistical model for scaling documents on latent dimensions, developed by Slapin and Proksch (2008). Unlike reference-based methods like Wordscores, Wordfish uses a Poisson generative model to jointly estimate word frequencies and document positions without requiring reference texts or manual annotation. It is particularly useful for estimating time-series changes in policy positions and can scale documents from multiple languages simultaneously.	Wordscores is a text-based scaling method developed by Laver, Benoit, and Garry (2003) that estimates the policy positions of political actors based on word frequencies in their texts. By comparing word usage in reference texts of known positions with test texts, the method infers the latent political dimension of any document without requiring manual coding or training data.