A Model of Text for Experimentation in the Social Sciences

Statistical models of text have become increasingly popular in statistics and computer science as a method of exploring large document collections. Social scientists often want to move beyond exploration, to measurement and experimentation, and make inference about social and political processes tha...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of the American Statistical Association Jg. 111; H. 515; S. 988 - 1003
Hauptverfasser: Roberts, Margaret E., Stewart, Brandon M., Airoldi, Edoardo M.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Alexandria Taylor & Francis 01.09.2016
Taylor & Francis Group,LLC
Taylor & Francis Ltd
Schlagworte:
Applications and Case Studies
Causal inference
China
Computer science
data collection
Design of experiments
Discourse
Discourse analysis
equations
experimental design
Experimentation
Experiments
Generalized linear models
High-dimensional inference
Inference
Linear analysis
linear models
Measurement
Measurement techniques
News media
Parameter estimation
Political discourse
Politics
Research design
Social sciences
Social scientists
Statistical methods
Statistics
Text analysis
Variational approximation
China
ISSN:0162-1459, 1537-274X, 1537-274X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Statistical models of text have become increasingly popular in statistics and computer science as a method of exploring large document collections. Social scientists often want to move beyond exploration, to measurement and experimentation, and make inference about social and political processes that drive discourse and content. In this article, we develop a model of text data that supports this type of substantive research. Our approach is to posit a hierarchical mixed membership model for analyzing topical content of documents, in which mixing weights are parameterized by observed covariates. In this model, topical prevalence and topical content are specified as a simple generalized linear model on an arbitrary number of document-level covariates, such as news source and time of release, enabling researchers to introduce elements of the experimental design that informed document collection into the model, within a generally applicable framework. We demonstrate the proposed methodology by analyzing a collection of news reports about China, where we allow the prevalence of topics to evolve over time and vary across newswire services. Our methods quantify the effect of news wire source on both the frequency and nature of topic coverage. Supplementary materials for this article are available online.
Bibliographie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ObjectType-Article-1
ObjectType-Feature-2
content type line 23
ISSN:0162-1459
1537-274X
1537-274X
DOI:10.1080/01621459.2016.1141684