research about foreign literature

Foreign Literatures in America (FLA) is a project devoted to the recovery and understanding of the significance of foreign authored literary works, as well as immigrant authored literary works, in the U.S. throughout U.S. history. Our principal mission is to challenge conceptions of “American literature” that turn upon the American citizenship of an author—when historically it is clear that foreign authored works, as well as works by immigrant authors who wrote in many languages and were not citizens of the United States, have long profoundly constituted an important part of the literatures and cultures of the U.S. This project thus seeks to offer many fresh opportunities to globalize the terms through which we understand American literature and American culture, both of these domains rediscovered as richly constituted and interpenetrated by global texts, concerns, contexts, voices. FLA pursues these goals by offering various means of studying the reception of foreign and immigrant authored literary works in the U.S., in interdisciplinary terms that encompass literature, culture, politics, history, and international relations. **Archival resources: **The project offers extensive archival resources of primary reception materials (i.e., accounts of “foreign” authors and works in newspapers, magazines, images, rarer archives, etc.) accessible in themselves, browsable in useful arrays, and searchable and subject to certain forms of quantitative analysis by nuanced means. Laboratories: The project develops laboratories based on cutting edge tools of machine learning drawn from recent digital humanities innovations in the areas of topics modeling and sentiment analysis; these laboratories allow users to mine large databases of “big data” already assembled for meaningful patterns and insights of literary reception. Book review pages: The project assembles its own smaller databases of book review pages from various U.S. newspapers and periodicals over time (beginning with the New York Times, The New Republic, and The Crisis), subject not only to the kind of searching and quantitative techniques of analysis found in the archival and laboratory sections, but also to comparative quantification of the most frequently mentioned authors in user-determined time frames and periodical ranges—(these book review pages thus become a powerful means of recovering forgotten literary and cultural history). Collective Forum: FLA is a committedly and internationally a collective forum for research, innovation, discussion, and collaboration, one in which blogging and various forms of collective interchange, suggestion, and crowd-sourced cooperation are facilitated—both as concerns all the research functions described above, and also toward innovation of further functions FLA could undertake. Beyond the general aims and specific outcomes noted above, there are two specific aims of this project that should be emphasized. First, in the shorter term, rather than do full comprehensive justice to any one of the functions described above, we are really trying to “open the door” with respect to them all, encouraging different teams of faculty and student researches both at the University of Maryland and around the country and world to develop dynamic possibilities for this project. Anyone interested in the kind of scholarly and analytic priorities foregrounded by the project is most warmly encouraged to contact us with your ideas and to join our project. The second point is more long term: though this project does aim to offer the means for a wholesale remapping of American literary studies (what this domain consists of, which voices and texts, why they are important), it is also a project of significance not only among university and academic research communities but also in larger social domains of education as well—including not only secondary schools and undergraduate pedagogy and also those interested in non-traditional education forms in our culture and society generally. This general goal of making a productive globalizing contribution to American education in the broadest possible terms is an ultimate aspiration for this project. The technological infrastructure for this project has been supported in part by a generous grant from Amazon Web Services .

Participants

FLA Project Website
Faculty Fellowships Open Up New Avenues for Research Collaboration August 29, 2011 MITH
Beginnings… December 7, 2011 Peter Mallios
Searching for the Quantum Dimension of Foreign Literature December 21, 2011 Rebecca Borden
Reinventing the Boundaries of American Literature January 9, 2012 Nicholas Slaughter
Telling the Story of Foreign Literatures in America January 23, 2012 Jennifer Wellman
Extremely Visible and Incredibly Close Reading of Logos February 7, 2012 Amanda Visconti
Open Water February 20, 2012 Peter Mallios
My Dissertation in the Year 2112 March 6, 2012 Rebecca Borden
Archive of Emotion April 2, 2012 Katherine Stanutz
Names of the Game April 16, 2012 Nicholas Slaughter
Progress Update on the Modern British Archive May 9, 2012 Jennifer Wellman
On Fish, FLA, and the Digital Humanities May 23, 2012 Peter Mallios
An Undergraduate View of Data Mining with WEKA November 5, 2012 Peter Mallios
Asking Questions of Lots of Text with Weka December 18, 2012 Peter Mallios
Amazon Web Services

Chester Fritz Library
Library of the Health Sciences
Thormodsgard Law Library

Foreign Language Literatures

How to use this research guide, more resources.

Finding Articles
Finding Books
Streaming Video
Guides for specific classes
Greek and Latin

Contact Your Subject Librarian

This guide is meant to assist you in searching for foreign language literatures. It has information on how to search our library for foreign language titles, as well as selected resources for various language literatures.

Use the green navigation tabs for more specific resources:

Article databases

Book Search and In-library locations

Foreign films on AVON streaming platform

For Languages Taught at UND, each specific language tab on the left contains:

Reference material

Selected works from each language translated into English

Guides for class assignments

Information on foreign language learning is contained on its own guide.

Information on Linguistics is contained on its own guide.

More foreign language film titles are on the DVDs and Streaming Video guide.

Next: Finding Articles >>
Last Updated: Jan 4, 2024 1:23 PM
URL: https://libguides.und.edu/foreign-literatures

Foreign Language and Literature
East Tennessee State University
Subject Guides
Research Methods
Articles & Databases
Books & eBooks
News Sources
Reference Tools
Research Tutorials
Open Educational Resources

Search Research Methods

SAGE Research Methods contains research methods, cases, datasets, and videos. All of these interfaces can be individually searched:

SAGE Research Methods This link opens in a new window Research methodology at all levels, step by step more... less... SAGE Research Methods (SRM) has a wide array of tools for every step of the research process plus hundreds of qualitative and quantitative methods, including interactive tools such as the Methods Map and Project Planner.
SAGE Research Methods Cases This link opens in a new window Case studies showing how methods are applied in real research projects. more... less... SAGE Research Methods Cases include hundreds of case studies showing how methods are applied in real research projects in various disciplines.
SAGE Research Methods Datasets This link opens in a new window Collection of datasets that can be used to support the teaching and learning of quantitative and qualitative analytical methods used in the social sciences. more... less... These are datasets taken from real research projects, but edited and cleaned for teaching purposes. Each dataset will be accompanied by a short clear narrative description of the data and easy-to-follow instructions on how to apply the research method.
SAGE Research Methods Videos This link opens in a new window 480+ videos covering the research process and hundreds of qualitative, quantitative, and mixed methods. more... less... SAGE Research Methods Video includes hours of tutorials, interviews, video case studies, and mini-documentaries covering the entire research process.
<< Previous: Reference Tools
Next: Research Tutorials >>

Methods Map

Need to find a Method? Explore the Methods Map!

ETSU Research Resources

REDCap DB/Survey Tool
Qualtrics Survey Software
Institutional Review Board (IRB)
Office of Research and Sponsored Programs

Foreign Literature

Foreign Literature , founded in 1980, has the longest history and the biggest influence among professional academic publications in the foreign literature research in China. Editors of the journal are researchers from the Institute of Foreign Literature, Beijing Foreign Studies University, and the editorial board is comprised of renowned scholars from home and abroad.

The journal conducts anonymous peer review. Famous scholars including Wang Zuoliang and Hu Wenzhong have served as its editor-in-chief, and the current editor-in-chief is Professor Jin Li.

With foreign literature researchers and enthusiasts as its main readers, the journal pays extensive attention to the languages and literature of various nationalities around the world, introduces the trends of foreign writers and their works and researches and develops critical theory. It publishes the latest results of foreign literature research, and promotes academic dialogue among different countries, regions and cultures.

The journal advocates in-depth research and new explorations, adheres to its literary nature, pursues an open, accurate, and concise style of writing, and is committed to creating a flourishing, lively academic atmosphere.

It is now a bimonthly magazine with columns such as Review, Theory, Cultural Research, and Book Reviews. In addition, it is a Chinese core journal, a source journal of the Chinese Social Science Citation Index (CSSCI), and a level-A core journal of AMI Comprehensive Evaluation of Chinese Humanities and Social Sciences Journals. It hosts a national academic seminar every year.

International subscription is made through China International Book Trading Corporation, 35 West ChegongzhuangRoad, Haidian District, Beijing 100048, China.

Manuscripts must adhere to MLA Style. Submission: http://wgwxqk.cbpt.cnki.net

Editorial correspondence should be addressed to Foreign Literature, Beijing Foreign Studies University, Beijing100089, China. Tel.: 86-10-88816730; Email: [email protected]

East Campus: No 2 Xisanhuan North Road, Haidian District, Beijing, China West Campus: No 19 Xisanhuan North Road, Haidian District, Beijing, China

Chapter Four: Theory, Methodologies, Methods, and Evidence

Research Methods

You are viewing the first edition of this textbook. a second edition is available – please visit the latest edition for updated information..

This page discusses the following topics:

Research Goals

Research method types.

Before discussing research methods , we need to distinguish them from methodologies and research skills . Methodologies, linked to literary theories, are tools and lines of investigation: sets of practices and propositions about texts and the world. Researchers using Marxist literary criticism will adopt methodologies that look to material forces like labor, ownership, and technology to understand literature and its relationship to the world. They will also seek to understand authors not as inspired geniuses but as people whose lives and work are shaped by social forces.

Example: Critical Race Theory Methodologies

Critical Race Theory may use a variety of methodologies, including

Interest convergence: investigating whether marginalized groups only achieve progress when dominant groups benefit as well
Intersectional theory: investigating how multiple factors of advantage and disadvantage around race, gender, ethnicity, religion, etc. operate together in complex ways
Radical critique of the law: investigating how the law has historically been used to marginalize particular groups, such as black people, while recognizing that legal efforts are important to achieve emancipation and civil rights
Social constructivism: investigating how race is socially constructed (rather than biologically grounded)
Standpoint epistemology: investigating how knowledge relates to social position
Structural determinism: investigating how structures of thought and of organizations determine social outcomes

To identify appropriate methodologies, you will need to research your chosen theory and gather what methodologies are associated with it. For the most part, we can’t assume that there are “one size fits all” methodologies.

Research skills are about how you handle materials such as library search engines, citation management programs, special collections materials, and so on.

Research methods are about where and how you get answers to your research questions. Are you conducting interviews? Visiting archives? Doing close readings? Reviewing scholarship? You will need to choose which methods are most appropriate to use in your research and you need to gain some knowledge about how to use these methods. In other words, you need to do some research into research methods!

Your choice of research method depends on the kind of questions you are asking. For example, if you want to understand how an author progressed through several drafts to arrive at a final manuscript, you may need to do archival research. If you want to understand why a particular literary work became a bestseller, you may need to do audience research. If you want to know why a contemporary author wrote a particular work, you may need to do interviews. Usually literary research involves a combination of methods such as archival research , discourse analysis , and qualitative research methods.

Literary research methods tend to differ from research methods in the hard sciences (such as physics and chemistry). Science research must present results that are reproducible, while literary research rarely does (though it must still present evidence for its claims). Literary research often deals with questions of meaning, social conventions, representations of lived experience, and aesthetic effects; these are questions that reward dialogue and different perspectives rather than one great experiment that settles the issue. In literary research, we might get many valuable answers even though they are quite different from one another. Also in literary research, we usually have some room to speculate about answers, but our claims have to be plausible (believable) and our argument comprehensive (meaning we don’t overlook evidence that would alter our argument significantly if it were known).

A literary researcher might select the following:

Theory: Critical Race Theory

Methodology: Social Constructivism

Method: Scholarly

Skills: Search engines, citation management

Wendy Belcher, in Writing Your Journal Article in 12 Weeks , identifies two main approaches to understanding literary works: looking at a text by itself (associated with New Criticism ) and looking at texts as they connect to society (associated with Cultural Studies ). The goal of New Criticism is to bring the reader further into the text. The goal of Cultural Studies is to bring the reader into the network of discourses that surround and pass through the text. Other approaches, such as Ecocriticism, relate literary texts to the Sciences (as well as to the Humanities).

The New Critics, starting in the 1940s, focused on meaning within the text itself, using a method they called “ close reading .” The text itself becomes e vidence for a particular reading. Using this approach, you should summarize the literary work briefly and q uote particularly meaningful passages, being sure to introduce quotes and then interpret them (never let them stand alone). Make connections within the work; a sk “why” and “how” the various parts of the text relate to each other.

Cultural Studies critics see all texts as connected to society; the critic therefore has to connect a text to at least one political or social issue. How and why does the text reproduce particular knowledge systems (known as discourses) and how do these knowledge systems relate to issues of power within the society? Who speaks and when? Answering these questions helps your reader understand the text in context. Cultural contexts can include the treatment of gender (Feminist, Queer), class (Marxist), nationality, race, religion, or any other area of human society.

Other approaches, such as psychoanalytic literary criticism , look at literary texts to better understand human psychology. A psychoanalytic reading can focus on a character, the author, the reader, or on society in general. Ecocriticism look at human understandings of nature in literary texts.

We select our research methods based on the kinds of things we want to know. For example, we may be studying the relationship between literature and society, between author and text, or the status of a work in the literary canon. We may want to know about a work’s form, genre, or thematics. We may want to know about the audience’s reading and reception, or about methods for teaching literature in schools.

Below are a few research methods and their descriptions. You may need to consult with your instructor about which ones are most appropriate for your project. The first list covers methods most students use in their work. The second list covers methods more commonly used by advanced researchers. Even if you will not be using methods from this second list in your research project, you may read about these research methods in the scholarship you find.

Most commonly used undergraduate research methods:

Scholarship Methods: Studies the body of scholarship written about a particular author, literary work, historical period, literary movement, genre, theme, theory, or method.
Textual Analysis Methods: Used for close readings of literary texts, these methods also rely on literary theory and background information to support the reading.
Biographical Methods: Used to study the life of the author to better understand their work and times, these methods involve reading biographies and autobiographies about the author, and may also include research into private papers, correspondence, and interviews.
Discourse Analysis Methods: Studies language patterns to reveal ideology and social relations of power. This research involves the study of institutions, social groups, and social movements to understand how people in various settings use language to represent the world to themselves and others. Literary works may present complex mixtures of discourses which the characters (and readers) have to navigate.
Creative Writing Methods: A literary re-working of another literary text, creative writing research is used to better understand a literary work by investigating its language, formal structures, composition methods, themes, and so on. For instance, a creative research project may retell a story from a minor character’s perspective to reveal an alternative reading of events. To qualify as research, a creative research project is usually combined with a piece of theoretical writing that explains and justifies the work.

Methods used more often by advanced researchers:

Archival Methods: Usually involves trips to special collections where original papers are kept. In these archives are many unpublished materials such as diaries, letters, photographs, ledgers, and so on. These materials can offer us invaluable insight into the life of an author, the development of a literary work, or the society in which the author lived. There are at least three major archives of James Baldwin’s papers: The Smithsonian , Yale , and The New York Public Library . Descriptions of such materials are often available online, but the materials themselves are typically stored in boxes at the archive.
Computational Methods: Used for statistical analysis of texts such as studies of the popularity and meaning of particular words in literature over time.
Ethnographic Methods: Studies groups of people and their interactions with literary works, for instance in educational institutions, in reading groups (such as book clubs), and in fan networks. This approach may involve interviews and visits to places (including online communities) where people interact with literary works. Note: before you begin such work, you must have Institutional Review Board (IRB) approval “to protect the rights and welfare of human participants involved in research.”
Visual Methods: Studies the visual qualities of literary works. Some literary works, such as illuminated manuscripts, children’s literature, and graphic novels, present a complex interplay of text and image. Even works without illustrations can be studied for their use of typography, layout, and other visual features.

Regardless of the method(s) you choose, you will need to learn how to apply them to your work and how to carry them out successfully. For example, you should know that many archives do not allow you to bring pens (you can use pencils) and you may not be allowed to bring bags into the archives. You will need to keep a record of which documents you consult and their location (box number, etc.) in the archives. If you are unsure how to use a particular method, please consult a book about it. [1] Also, ask for the advice of trained researchers such as your instructor or a research librarian.

What research method(s) will you be using for your paper? Why did you make this method selection over other methods? If you haven’t made a selection yet, which methods are you considering?
What specific methodological approaches are you most interested in exploring in relation to the chosen literary work?
What is your plan for researching your method(s) and its major approaches?
What was the most important lesson you learned from this page? What point was confusing or difficult to understand?

Write your answers in a webcourse discussion page.

Introduction to Research Methods: A Practical Guide for Anyone Undertaking a Research Project by Catherine, Dr. Dawson
Practical Research Methods: A User-Friendly Guide to Mastering Research Techniques and Projects by Catherine Dawson
Qualitative Inquiry and Research Design: Choosing Among Five Approaches by John W. Creswell Cheryl N. Poth
Qualitative Research Evaluation Methods: Integrating Theory and Practice by Michael Quinn Patton
Research Design: Qualitative, Quantitative, and Mixed Methods Approaches by John W. Creswell J. David Creswell
Research Methodology: A Step-by-Step Guide for Beginners by Ranjit Kumar
Research Methodology: Methods and Techniques by C.R. Kothari

Strategies for Conducting Literary Research Copyright © 2021 by Barry Mauer & John Venecek is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License , except where otherwise noted.

Share This Book

Harvard Library
Research Guides
Faculty of Arts & Sciences Libraries

Literature: A Research Guide for Graduate Students

Research dos & don'ts.

Get Started
Find a Database

DON'T reinvent the wheel

Many scholars have spent their entire careers in your field, watching its developments in print and in person. Learn from them! The library is full of specialized guides, companions, encyclopedias, dictionaries, bibliographies, histories and other "reference" sources that will help orient you to a new area of research. Similarly, every works cited list can be a gold mine of useful readings.

Techniques for finding where a particular publication is cited (reverse footnote-mining) [Harvard Library FAQ]
Top resources and search tips for locating scholarly companions and guides [general topic guide for literary research]
The literature section of the Loker Reading Room reference collection [HOLLIS browse]
James Harner's Literary Research Guide: an Annotated Listing of Reference Sources in English Literary Studies [HOLLIS record with ONLINE ACCESS]

DO get to know your field

Know Your Field , a module from Unabridged On Demand, offers tips, thought prompts, and links to resources for quickly learning about and staying current with an area of scholarly study.

DON'T treat every search box like Google or ChatGPT

Break free of the search habits that Google and generative AI have taught you! Learn to pay attention to how a search system operates and what is in it, and to adjust your search inputs accordingly.

Google and generative AI interfaces train you to type in your question as you would say it to another person. They give you the illusion of a search box that can read your thoughts and that access the entire internet. That's not what's actually happening, of course! Google is giving you the results others have clicked on most while generative AI is giving you the output that is most probable based on your input. Other search systems, like the library catalog, might be matching your search inputs to highly structured, human-curated data. They give the best results when you select specific keywords and make use of the database's specialized search tools.

Learn more about searching:

Database Search Tips from MIT: a great, concise introduction to Booleans, keywords v. subjects, and search fields
Improve Your Search , a module from our library research intensive, Unabridged On Demand

Search technique handouts

"Search Smarter" Bookmark Simple steps to improve your searching, plus a quick guide to the search commands HOLLIS uses
Decoding a database A two-page guide to the most effective ways to quickly familiarize yourself with a new system.
Optimize Your Search A 3-column review of the basic search-strategy differences between Google and systems like JSTOR or HOLLIS.

DO adjust your language

Searching often means thinking in someone else's language, whether it's the librarians who created HOLLIS's subject vocabularies, or the scholars whose works you want to find in JSTOR, or the people of another era whose ideas you're trying to find in historical newspapers. The Search Vocabulary page on the general topic guide for literary studies is a great place to start for subject vocabularies.

DON'T search in just one place

Judicious triangulation is the key to success. No search has everything. There's always one more site you could search. Strike a balance by always searching at least 3-4 ways.

DO SEARCH A VARIETY OF RESOURCES:

Your library catalog , HOLLIS
A subject-specific scholarly index , such as the MLA International Bibliography , LION (Literature Online) , or the IMB (International Medieval Bibliography)
A full-text collection of scholarship, such as JSTOR or ProjectMuse
One of Google's full-text searches, Google Scholar or Google Books

DO look beyond the library's collections

The library purchases and licenses materials for your use. There's plenty of other material that's freely available or that you would need to travel to see---please let me help you find it!

<< Previous: Find a Database

Except where otherwise noted, this work is subject to a Creative Commons Attribution 4.0 International License , which allows anyone to share and adapt our material as long as proper attribution is given. For details and exceptions, see the Harvard Library Copyright Policy ©2021 Presidents and Fellows of Harvard College.

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

We're Hiring!
Help Center

CHAPTER 2 Review of Related Literature and Studies Foreign Literature Student Performance Galiher

Related Papers

Ruban Vasanth

Partly Cloudy

Chika Albert

Justine Martin

Muhammad Daniyal

Eumi Cuntapay

Determinants of students ' performance have been the subject of ongoing debate among educators, academics, and policy makers. There have been many studies that sought to examine this issue and their findings point out to hard work, previous schooling, parents ’ education, family income and self motivation as factors that have a significant effect on the students GPA. Most of those studies have focused on students ' performance in the U.S. and Europe. However, since cultural differences may play a role in shaping the factors that affect students' performance, it is very important to examine those relevant factors to the UAE society. The aim of this study is to investigate the socio-economic characteristics of students of the College of Business and Economics-UAEU in relation to these students ' performance and taking into account variables pertaining to the UAE Society. Using a sample of 864 CBE student and

Nova Degorio , Katsikas Elias

JOSEPHAT NCHUNGO

SYSTEMATIC REVIEW article

Foreign language anxiety research in system between 2004 and 2023: looking back and looking forward.

Faculty of Humanities and Foreign Languages, Xi’an University of Technology, Xi’an, Shaanxi, China

With the deepening of the research on emotional factors, foreign language anxiety (FLA) has become the focus of researchers in the field of foreign language learning (FLL) and teaching. This paper aims to provide an overview of the historical trajectory of FLA research that has been published in System between 2004 and 2023. While examining the retrieved 49 studies, focus has been laid on the methodologies including research instruments, methods, participants, major themes and key findings of FLA research. Although almost all of the studies employed quantitative and mix-methods methodologies, questionnaires and semi-structured interviews were the most preferred research methods. FL learners from 21 countries/regions were represented, but a significant number of the studies came from China, Japan and Iran. And an overwhelming majority of the studies focused on FLA among the learners learning English as a foreign language (EFL). The review concluded with some research lacunae and possible directions for future research on FLA.

Introduction

FLA, prevalent among foreign language (FL) learners ( Dewaele and Macintyre, 2014 ; Li, 2020 ), is a very special and complex psychological phenomenon during the process of FLL ( Gardner, 1985 ; Macintyre and Gardner, 1994 ). FLA is regarded as the biggest emotional obstacle during the process of FLL ( Arnold and Brown, 1999 ), which may undermine students’ confidence and motivation in FLL ( Macintyre, 2017 ). Horwitz (2010) considered FLA as one of the strongest predictors of success or failure in FLL. Previously, anxiety in FFL, as an auxiliary variable in FLL research, had only drawn scarcity of attention from researchers ( Chastain, 1975 ; Dewaele and Li, 2021 ). It was not until 1986 that Horwitz et al. (1986) , for the first time, proposed the concept of FLA, reckoning that FLA is a unique synthesis of self-perception, belief, emotion and behavior associated with FLL. Meanwhile, Horwitz et al. (1986) developed the Foreign Language Classroom Anxiety Scale (FLCAS), which has become the most widely accepted FLA scale. Since then, researchers have conducted a plethora of studies on the connotations ( Macintyre and Gardner, 1994 ; Oxford, 1999 ), categorization ( Horwitz et al., 1986 ; Ellis, 1994 ; Macintyre and Gardner, 1994 ), impacts ( Steinberg and Horwitz, 1986 ; MacIntyre and Charos, 1996 ), sources ( Young, 1991 ; Macintyre, 2017 ), and measurement tools ( Macintyre and Gardner, 1994 ; Satio et al., 1999 ; Kim, 2000 ; Elkhafaifi, 2005 ; Woodrow, 2006 ; Cheng, 2017 ) of FLA.

System , one of the most influential and prestigious international journals devoted to FL teaching and learning, has stayed abreast of the development of FLA research. The articles having been published on FLA in System represent to a large extent the development trajectory of FLA research. Therefore, this review paper chooses System as the material to provide the historical trajectory of FLA research and suggest some under-researched topics and future directions of FLA research.

Foreign language anxiety

FLA, a principal learner emotional factor in foreign language learning (FLL), has become one of the significant research focuses in FLL since the 1970s. Originating from psychology, anxiety is defined as “an unpleasant state of mind that is characterized by individual perceived feelings like nervous, fear, and worry, and is activated by the autonomic nervousness system” ( Spielberger, 1972 ). FLA is a unique form of anxiety in the specific context of foreign language learning ( Horwitz et al., 1986 ; MacIntyre, 1995 ). Horwitz et al. (1986) conceptualized FLA as “a distinct complex of self-perceptions, beliefs, feelings, and behaviors related to classroom language learning arising from the uniqueness of the language learning process.”

Horwitz et al. (1986) first studied FLA as an independent phenomenon. In order to resolve the deficiency and insufficiency of traditional research tools in respect of FLA, Horwitz et al. (1986) framed the Foreign Language Classroom Anxiety Scale (FLCAS), putting an end to the history of FLA study having no standardized measurement tools ( Guo and Xu, 2014 ), foreboding that FLA research entered a period of relative maturity when researchers began to focus on the overall performance of FLA and its relationship with a variety of variables ( Young, 1986 , 1992 ; Aida, 1994 ), as well as the relationship between FLA and some basic language skills like listening, speaking, reading and writing ( Gungle and Taylor, 1989 ; Vogely, 1998 ; Sellers, 2000 ).

Simply put, FLA is the feeling of tension, fear and nervousness in self-consciousness, emotions, beliefs, and behaviors ( Aida, 1994 ) associated with a context which requires an individual to use a foreign language he or she is not proficient with ( MacIntyre and Gardner, 1991 ).

Research design

In order to present a systematic analysis of FLA research published in System , a narrative approach of systematic review was adopted. Systematic review involves “a clearly formulated question” and adopts “systematic and explicit methods to identify, select, and critically appraise relevant research, and to collect and analyze data from the studies that are included in the review” ( Cochrane Collaboration, 2003 ). A narrative approach relies “primarily on the use of words and text to summarize and explain the findings,” and is considered helpful to systematically review topics that have been studied differently researchers ( Popay et al., 2006 ), highlight the strengths and limitations of studies being reviewed ( Wong et al., 2013 ).

The review aims to provide a systematic analysis of FLA research during the past two decades between 2004 and 2023 by answering the following questions:

Question 1: What is the overall trend in FLA research published in System during the past two decades? Question 2: What are the major themes and the key findings of FLA research? Question 3: What are the existent gaps in the current research and the potential directions for future research?

Data collection

Following the PRISMA guidelines ( Moher et al., 2009 ), an extensive literature search was conducted to ensure a comprehensive analysis of the current FLA research published in System . The data selection criteria and collection process are summarized in Figure 1 .

Figure 1 . PRISMA flow chart.

Firstly, relevant studies published until and including December 15, 2023 were searched in the database of Elsevier ScienceDirect. The author conducted advanced searches in the database with the following searching parameters: In this journal or book title = ( System ) AND Title, abstract, or author-specific keywords = (anxiety). Overall, the database returned 185 publications, among which 95 were published in journals other than System and therefore were deleted. Then, 2 book reviews and 1 review article were deleted. The remaining 87 publications were evaluated for the eligibility by reading and analyzing the titles, abstracts and full texts, and 38 publications were excluded based on the following criterion that the studies focused on topics other than FLA.

Data analysis

This review first conducted a bibliometric analysis of the retrieved records. A coding analysis was then performed through iterative reading with the highlights on the following categories that guided the data analysis: year of publication, characteristics of samples, research methodologies, and key findings.

Descriptive characteristics of studies

Publishing years.

There is a dynamic upward trend in the number of studies on FLA over the past two decades (see Figure 2 ). 2021 witnessed a surge in the number of publications, reaching an all-time peak of 8 papers. There is a gradual downward trend in the following 2 years, but compared with the average of about 2 papers per year, there is still an increase in the number of papers in 2022 and 2023.

Figure 2 . Trend of FLA research in System (2004–2023).

Countries/regions of research

There was diversity of countries/ regions where the studies took place, with 21 countries/regions represented. Twenty-one papers (42.86%) came from China, followed by 6 papers (12.24%) from Japan and 4 papers (8.16%) from Iran. Three papers were conducted in Korea and USA each, followed by 2 papers from Saudi Arabia, and 1 paper from Canada, Australia, Indonesia, Austria, Germany, Switzerland, Italy, Slovakia, Macau, Chile, Thailand, Turkey, and UK each.

Research participants

The overwhelming majority of the studies ( n = 43, 87.76%) focused on FLA among university students, with 3 papers focused on primary school students and 5 on secondary school students. It is noteworthy to point out that there were 3 studies which focused on PhD students, adult students, and vocational high school students, respectively.

Foreign languages studied

Since the status of English as a universal language is beyond doubt, almost all the studies examined FLA in the context of EFL learning. Among the 49 papers, there were only 4 papers focusing on FLA in the context of FLL other than EFL learning. These foreign languages included Korean, Arabic, German and Chinese. There was 1 paper comparing the possible FLA differences between the contexts of German as L1 learning and English as LX learning.

Research methodologies

Of the retrieved records, 38 were cross-sectional studies and 11 were longitudinal studies. The average length of time for the longitudinal studies was approximately 11.36 weeks, ranging from the shortest length of 1 week to the longest length of 18 weeks. The studies adopting quantitative methods ( n = 24, 48.98%) and the studies using mix-methods ( n = 24, 48.98%) markedly dwarfed the only one study using qualitative methods (2.04%).

Questionnaires were the most common research method in both quantitative studies and mix-methods studies. The FLCAS was the most frequently-used scale ( n = 22, 48.83%), which a significant number of studies used directly ( n = 7, 14.58%), adapted ( n = 4, 8.33%), modified ( n = 6, 12.50%) or translated ( n = 5, 10.42%). Besides, a number of studies used questionnaires that adopted, modified or translated other scales such as the FLRAS ( Satio et al., 1999 ), the FLLAS ( Elkhafaifi, 2005 ), and the S-FLCAS ( Dewaele and Macintyre, 2014 ) among many others. Notwithstanding, some researchers devised targeted questionnaires ( Hurd, 2007 ; Woodrow, 2011 ; Lee, 2016 ; Li, 2018 ; Wang H. et al., 2021 ; Alrabai, 2022 ).

With regard to qualitative methodology, the research methods frequently used to measure FLA embraced interviews ( n = 15, 62.50%), classroom observations ( n = 5, 20.83%), students’ reflective journals ( n = 5, 20.83%), open-ended questions ( n = 3, 12.50%). Of note, Hurd (2007) employed audio-recorded think-aloud protocols combined with questionnaires and one-to-one semi-structured telephone interviews to explore FLA in a distance learning environment. Dryden et al. (2021) used linguistic ethnography to investigate how four migrant EFL learners in Australia experienced FLA.

Research themes and key findings

Level of fla.

Twelve papers (24.49%) were found to investigate FLA level of FL learners, however, no consensus has been reached on the level of FLA among FL learners, possibly due to the fact that the participants of the retrieved studies were different. For example, Jiang and Dewaele (2020) found 1,031 university freshmen in China experienced a moderate level of FLA. Zuniga and Simard (2022) and Lee et al. (2023) had similar findings. However, Jiang and Dewaele (2019) found a higher level of FLA among 564 EFL university freshmen in China than the counterpart participants in the study of Dewaele and MacIntyre (2014) . Similarly, Bekleyen (2009) found the language teacher candidates in Turkey experienced a high level of FL listening anxiety.

Dynamicity of FLA drew attention from some scholars. Koga (2010) investigated the dynamicity of FLA among 88 first-year university students in Japan and found FLA decreased significantly at the end of the 15-week English courses. Veenstra and Weaver (2022) investigated 341 students from two private universities in Japan and a continuum of FL speaking anxiety showed that the participants’ overall level of FL speaking anxiety decreased after finishing an English presentation course lasting 15 weeks.

Some studies explored some potential differences of FLA among different participants or among the same participants in different contexts. For example, Chen et al. (2022) found Chinese undergraduates had a higher level of EFL reading anxiety than Spanish undergraduates. Resnik and Dewaele (2020) found the participants experienced a higher level of FLA in English (LX) classes than in German (L1) classes.

Sources of FLA

Nine studies (18.37%) explored sources or causes of FLA. Bekleyen (2009) revealed some major sources of FL listening anxiety, including low priority of listening in previous FLL, and failure to recognize the spoken form of word, phrase or sentence. Jiang and Dewaele (2019) uncovered a number of factors contributing to FL class anxiety, including exams and quizzes, speaking in front of the class without preparation, challenging classroom activities, and teacher questioning. Bashori et al. (2021) identified insufficient vocabulary knowledge as one of the factors provoking FL speaking anxiety. Besides, speaking strategies, willingness to communicate, speaking self-efficacy and speaking proficiency were found to have positive direct effects on FLA ( Sun and Teng, 2021 ). Of note, Zare et al. (2022) focused on FLA outside the traditional face-to-face classroom and found that autonomous learning was the source of the participants’ anxiety during the data-driven FFL.

Correlation of FLA with other variables

Some studies ( n = 5, 10.20%) explored the correlation of FLA with demographic variables of the participants. Park and French (2013) found female students had significantly higher levels of FLA than male students. However, Jiang and Dewaele (2020) found gender and ethnic affiliation were not correlated with FLA while geographical background and experience in traveling abroad had a weak correlation with FLA. Similarly, Matsuda and Gobel (2004) found EFL learners with overseas experience experienced lower anxiety when speaking English and gender did not have a significant effect on FLA. However, Yim (2014) found gender had a significant effect of FLA. The discrepancies in the correlation with demographic variables may be attributed to the different samples or the possibility that male learners are not inclined to willingly admit anxiety than female learners ( Williams, 1996 ; Pappamihiel, 2002 ).

A number of studies ( n = 9, 18.37%) explored the correlation of FLA with academic performance/ achievement. For example, Pyun et al. (2014) found that oral achievement of the participants was negatively correlated with FLA. However, Tsang and Lee (2023) found FL speaking anxiety was not significantly related to speaking proficiency. Hamada and Takaki (2021) found FL reading anxiety had significantly direct effects on course achievement. Woodrow (2011) and Li et al. (2023) found FL writing anxiety was significantly negatively correlated with writing performance, but FLA did not have a significant prediction on writing achievement ( Li et al., 2023 ). Besides, In’nami (2006) found that test anxiety did not affect FL listening test performance.

Many studies ( n = 19, 38.78%) focused on the correlation of FLA with other student-specific variables, including learning motivation ( Tsai and Liao, 2021 ), willingness to communicate ( Lee and Hsieh, 2019 ; Wang H. et al., 2021 ), language proficiency ( Jiang and Dewaele, 2020 ) and trait emotional intelligence ( Resnik and Dewaele, 2020 ; Li et al., 2021 ) among many others. Several studies ( n = 5, 10.20%) focused on the correlation of FLA with teacher-specific variables, such as teachers’ oral corrective feedback ( Lee, 2016 ), perceived teacher emotional support ( Jin and Dewaele, 2018 ), and teaching styles ( Briesmaster and Briesmaster-Paredes, 2015 ).

Ways to relieve FLA

Ways to relieve FLA was also a topic of immense interest to researchers. Ten studies (20.41%) explored how to relieve or alleviate FLA. Jin et al. (2021) and Alrabai (2022) applied positive psychology intervention to reduce leaners’ FLA. Alrabai (2022) revealed that the integration of positive and negative emotions in FLL could result in alleviation of FLA among Saudi EFL learners. Jin et al. (2021) uncovered that reminiscing about language achievements significantly mitigated the levels of FLA among Chinese EFL learners. Similarly, Lee et al. (2023) found that constructing learners’ growth language mindset relieved their FLA.

Besides, Tsai and Liao (2021) found using machine translation systems had a positive effect on lowering FLA among EFL learners in Taiwan. Bashori et al. (2021) investigated the potential effects of Automatic Speech Recognition-based websites on EFL learners’ vocabulary, FLA and FLE. Other studies found that self-regulatory strategies ( Guo et al., 2018 ), recasts ( Li, 2018 ), and translanguaging ( Dryden et al., 2021 ) had a significant effect on mitigating the levels of FLA among EFL learners. Of note, Kralova et al. (2017) employed psycho-social training as a strategy to alleviate FLA among 68 Slovak EFL learners.

During the past two decades between 2004 and 2023, System has been an ardent supporter of FLA research, committed to probing into and resolving FLA-related problems of foreign language teaching and learning. However, based on the review, some research lacunae are discerned concerning samples, methodologies and themes of FLA research, and some possible directions for future FLA research are also suggested.

Research samples

Notwithstanding the FLA studies in System involved a variety of FL learners as the participants, there was a serious polarization phenomenon concerning the diversity of the research samples. An overwhelmingly large number of the studies focused on FLA among the FL learners in school and few studies focused on FLA among non-school FL learners. Moreover, a majority of the studies explored FLA among undergraduate students, especially the non-English-major university students, and there is a scarcity of studies investigating FLA among students in primary schools, secondary schools, vocational colleges as well as postgraduate students. In terms of geographical distribution of the research samples, most studies focused on FL learners from Asian countries including China, Japan and Iran among many others, and less attention was paid to FL learners from Europe, North America and South America. And no studies on FLA involving FL learners in Africa have been found. Meanwhile, most participants were from urban places, and only a couple of studies explored FLA among rural FL learners ( Hamada and Takaki, 2021 ; Li et al., 2023 ). Last but not the least, with regard to the types of FL, a plethora of studies concentrated on English as a FL. Of the 49 retrieved studies, only 4 studies focused on FLA among the participants learning Korean, Arabic, German and Chinese as a FL, respectively.

Future research should diversify the research objects and focus increasing attention on the FLA research among primary school students, secondary school students, vocational college students and non-school FL adult learners, and moderate attention should be paid to the FLA research among preschool children and postgraduate students, so as to avoid the polarization of research samples. Besides, the dominance of English as a lingua franca has made English the FL taught in schools around the globe ( Rose et al., 2020 ), facilitating FLA studies among EFL learners. However, recent years has witnessed frequent calls for conducting research on teaching and learning of foreign languages other than English ( Zhang et al., 2019 ; Guo et al., 2021 ). Future studies can also focus on FLA among learners of foreign languages other than English as well as FL learners in countries and regions outside Asia.

Notwithstanding an increasing number of studies combined quantitative methods and qualitative methods in recent years, questionnaires were still the staple tool for quantitative data collection, and semi-structured interviews for qualitative data collection. A few mix-methods studies used classroom observation, student journals, field investigation and case studies for qualitative analysis. In addition, the FLCAS was the most popular scale for quantitative data collection and analysis, and only a few studies devised target questionnaires for their research. Moreover, cross-sectional studies far exceeded longitudinal studies, and the average length of time for longitudinal studies were relatively short, lasing about 10 weeks. Finally, there were only three comparative studies on FLA, probing into FLA differences among the participants ( Resnik and Dewaele, 2020 ; Hamada and Takaki, 2021 ; Chen et al., 2022 ).

Future FLA research should adopt mix-methods studies with qualitative research not just being confined to semi-structured interviews, but embracing a variety of methods, such as classroom observation, video recording, student journals, field investigation, case study and particularly audio-recorded think-aloud protocols. And path analysis and structural equation modeling analysis should be increasingly employed to analyze the quantitative data. Meanwhile, some advanced techniques such as Event-related Potentials (ERP), Positron Emission Tomography (PET) and functional Magnetic Resonance Imaging (fMRI) can be used in future research to analyze FLA from the perspective of neural mechanism by measuring the electromagnetic, blood flow and neuronal activities of the human brain. In addition, it is necessary to devise localized FLA scale with ideal validity and reliability in accordance with the cultural background and educational environment of the country or region where the research objects are located. Moreover, the dynamic nature of FLA requires more longitudinal studies to explain how FLA changes dynamically and what impacts FLA exerts on FLL. Finally, future studies can pay more attention to the comparative study of FLA differences among different groups, which is more conducive to understanding the characteristics and distribution of FLA among different groups of FL learners, so as to put forward targeted strategies to mitigate FLA in FLL.

Research themes

Research themes of the studies on FLA in System were of rich variety. However, no research has been found on translation anxiety and interpretation anxiety. Besides, there was a scarcity of research on the effectiveness of alleviating FLA. Studies on strategies to reduce FLA were mostly conducted from the perspective of teachers, and few studies revealed how to alleviate FLA from the perspective of learners. And most of the specific strategies to mitigate FLA were only at the theoretical level, lacking sufficient theoretical and empirical evidence, which were not applicable in practical FL teaching.

The following research themes deserve more attention in future research: translation anxiety and interpretation anxiety, types and effectiveness of strategies for alleviating FLA among different groups of FL learners, FLA among learners of heritage languages as well as non-heritage languages, and comparative studies on the effects of regional locations and mother languages on FLA. Moreover, future studies should not only focus on the theoretical research of FLA, but also carry out more empirical studies on strategies on how to alleviate FLA among different FL learners, such as learners from different regional locations, learners in monolingualism, bilingualism and multilingualism, and the effectiveness of FLA-alleviating strategies.

By reviewing the 49 studies on FLA published in System between 2004 and 2023, this paper demonstrates that the journal’s commitment to FLA research embraces a wide range of research themes being explored with different research methods. Based on the findings of the review, some research lacunae regarding samples, methodologies and themes of FLA research are discussed, and some possible directions for future FLA research are also suggested.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Author contributions

QY: Writing – review & editing, Writing – original draft.

The author(s) declare that financial support was received for the research, authorship, and/or publication of this article. This work was supported by grants from the Research Project of Humanities, Foreign Languages and Arts, Xi’an University of Technology (110-451623011) and the Research Project on Graduate Education and Teaching Reform, Xi’an University of Technology (310-252042342).

Conflict of interest

The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Aida, Y. (1994). Examination of Horwitz, Horwitz, and Cope's construct of foreign language anxiety: the case of students of Japanese. Mod. Lang. J. 78, 155–168. doi: 10.2307/329005

Crossref Full Text | Google Scholar

Alrabai, F. (2022). The role of mixed emotions in language learning and teaching: a positive psychology teacher intervention. System 107:102821. doi: 10.1016/j.system.2022.102821

Arnold, J., and Brown, H. D. (1999). “A map of the terrain” in Affect in language learning . ed. J. Arnold (Cambridge: Cambridge University Press)

Google Scholar

Bashori, M., van Hout, R., Strik, H., and Cucchiarini, C. (2021). Effects of ASR-based websites on EFL learners’ vocabulary, speaking anxiety, and language enjoyment. System 99:102496. doi: 10.1016/j.system.2021.102496

Bekleyen, N. (2009). Helping teachers become better English students: causes, effects, and coping strategies for foreign language listening anxiety. System 37, 664–675. doi: 10.1016/j.system.2009.09.010

Briesmaster, M., and Briesmaster-Paredes, J. (2015). The relationship between teaching styles and NNPSETs’ anxiety levels. System 49, 145–156. doi: 10.1016/j.system.2015.01.012

Chastain, K. (1975). Affective and ability factors in second language acquisition. Lang. Learn. 25, 153–161. doi: 10.1111/j.1467-1770.1975.tb00115.x

Chen, S., Du, H., Wang, S., and Yang, L. (2022). Understanding EFL reading anxiety in relation to learning motivation, attitudes and strategies for Chinese and Spanish undergraduates. System 108:102842. doi: 10.1016/j.system.2022.102842

Cheng, Y. (2017). Development and preliminary validation of four brief measures of L2 language-skill-specific anxiety. System 68, 15–25. doi: 10.1016/j.system.2017.06.009

Cochrane Collaboration (2003). Glossary . London: Cochrane Collaboration.

Dewaele, J. M., and Li, C. (2021). Teacher enthusiasm and students’ social-behavioral learning engagement: the mediating role of student enjoyment and boredom in Chinese EFL classes. Lang. Teach. Res. 25, 922–945. doi: 10.1177/13621688211014538

Dewaele, J. M., and Macintyre, P. D. (2014). The two faces of Janus? Anxiety and enjoyment in the foreign language classroom. Stud. Second Lang. Learn. Teach. 4, 237–274. doi: 10.14746/ssllt.2014.4.2.5

Dryden, S., Tankosić, A., and Dovchin, S. (2021). Foreign language anxiety and translanguaging as an emotional safe space: migrant English as a foreign language learners in Australia. System 101:102593. doi: 10.1016/j.system.2021.102593

Elkhafaifi, H. (2005). Listening comprehension and anxiety in the Arabic language classroom. Mod. Lang. J. 89, 206–220. doi: 10.1111/j.1540-4781.2005.00275.x

Ellis, R. (1994). The study of second language acquisition . Oxford: Oxford University Press.

Gardner, R. C. (1985). Social psychology and second language learning: the role of attitudes and motivation . London: Edward Arnold.

Gungle, B. W., and Taylor, V. (1989). “Writing apprehension and second language writers” in Richness in writing: Empowering ESL students . eds. D. M. Johnson and D. H. Roen (White Plains, NY: Longman), 235–248.

Guo, Y., and Xu, J. (2014). A multi-dimensional study on foreign language anxiety of non-English major college students in China. Foreign Lang. World 4, 2–11.

Guo, Y., Xu, J., and Liu, X. (2018). English language learners’ use of self-regulatory strategies for foreign language anxiety in China. System 76, 49–61. doi: 10.1016/j.system.2018.05.001

Guo, Q., Zhou, X., and Gao, X. (2021). Research on the teaching and learning of languages other than English in system. System 100:102541. doi: 10.1016/j.system.2021.102541

Hamada, A., and Takaki, S. (2021). Effects of multidimensional foreign language reading anxiety on achievement in Japanese EFL classrooms. System 101:102613. doi: 10.1016/j.system.2021.102613

Horwitz, E. K. (2010). Foreign and second language anxiety. Lang. Teach. 43, 154–167. doi: 10.1017/S026144480999036X

Horwitz, E. K., Horwitz, M. B., and Cope, J. (1986). Foreign language classroom anxiety. Mod. Lang. J. 70, 125–132. doi: 10.1111/j.1540-4781.1986.tb05256.x

Hurd, S. (2007). Anxiety and non-anxiety in a distance language learning environment: the distance factor as a modifying influence. System 35, 487–508. doi: 10.1016/j.system.2007.05.001

In’nami, Y. (2006). The effects of test anxiety on listening test performance. System 34, 317–340. doi: 10.1016/j.system.2006.04.005

Jiang, Y., and Dewaele, J. M. (2019). How unique is the foreign language classroom enjoyment and anxiety of Chinese EFL learners? System 82, 13–25. doi: 10.1016/j.system.2019.02.017

Jiang, Y., and Dewaele, J. M. (2020). The predictive power of sociobiographical and language variables on foreign language anxiety of Chinese university students. System 89:102207. doi: 10.1016/j.system.2020.102207

Jin, Y. X., and Dewaele, J. M. (2018). The effect of positive orientation and perceived social support on foreign language classroom anxiety. System 74, 149–157. doi: 10.1016/j.system.2018.01.002

Jin, Y., Dewaele, J. M., and MacIntyre, P. D. (2021). Reducing anxiety in the foreign language classroom: a positive psychology approach. System 101:102604. doi: 10.1016/j.system.2021.102604

Kim, J. H. (2000). Foreign language listening anxiety: a study of Korean students learning English . Austin: The University of Texas.

Koga, T. (2010). Dynamicity of motivation, anxiety and cooperativeness in a semester course. System 38, 172–184. doi: 10.1016/j.system.2010.03.001

Kralova, Z., Skorvagova, E., Tirpakova, A., and Markechova, D. (2017). Reducing student teachers' foreign language pronunciation anxiety through psycho-social training. System 65, 49–60. doi: 10.1016/j.system.2017.01.001

Lee, E. J. (2016). Reducing international graduate students’ language anxiety through oral pronunciation corrections. System 56, 78–95. doi: 10.1016/j.system.2015.11.006

Lee, J. S., and Hsieh, J. C. (2019). Affective variables and willingness to communicate of EFL learners in in-class, out-of-class, and digital contexts. System 82, 63–73. doi: 10.1016/j.system.2019.03.002

Lee, H., Jin, S., and Lee, J. H. (2023). Ending the cycle of anxiety in language learning: a non-recursive path analysis approach. System 118:103154. doi: 10.1016/j.system.2023.103154

Li, H. (2018). Recasts and output-only prompts, individual learner factors and short-term EFL learning. System 76, 103–115. doi: 10.1016/j.system.2018.05.004

Li, C. (2020). A positive psychology perspective on Chinese EFL students’ trait emotional intelligence, foreign language enjoyment and EFL learning achievement. J. Multiling. Multicult. Dev. 41, 246–263. doi: 10.1080/01434632.2019.1614187

Li, C., Huang, J., and Li, B. (2021). The predictive effects of classroom environment and trait emotional intelligence on foreign language enjoyment and anxiety. System 96:102393. doi: 10.1016/j.system.2020.102393

Li, C., Wei, L., and Lu, X. (2023). Contributions of foreign language writing emotions to writing achievement. System 116:103074. doi: 10.1016/j.system.2023.103074

MacIntyre, P. D. (1995). How does anxiety affect second language learning? A reply to Sparks and Ganschow. Mod. Lang. J. 1, 90–99. doi: 10.1111/j.1540-4781.1995.tb05418.x

Macintyre, P. D. (2017). “An overview of language anxiety research and trends in its development” in New insights into language anxiety: theory, research and educational implications . ed. J. M. Dewaele (Bristol: Multilingual Matters)

MacIntyre, P. D., and Charos, C. (1996). Personality, attitudes and affect as predictors of second language communication. J. Lang. Soc. Psychol. 15, 3–26. doi: 10.1177/0261927X960151001

MacIntyre, P. D., and Gardner, R. C. (1991). Methods and results in the study of anxiety and language learning: a review of the literature. Lang. Learn. 41, 85–117. doi: 10.1111/j.1467-1770.1991.tb00677.x

Macintyre, P. D., and Gardner, R. C. (1994). The subtle effects of language anxiety on cognitive processing in the second language. Lang. Learn. 44, 283–305. doi: 10.1111/j.1467-1770.1994.tb01103.x

Matsuda, S., and Gobel, P. (2004). Anxiety and predictors of performance in the foreign language classroom. System 32, 21–36. doi: 10.1016/j.system.2003.08.002

Moher, D., Liberati, A., Tetzlaff, J., and Altman, D. G. (2009). Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 6:e1000097. doi: 10.1371/journal.pmed.1000097

PubMed Abstract | Crossref Full Text | Google Scholar

Oxford, R. L. (1999). “Anxiety and the language learner: new insights” in Affect in language learning . ed. J. Arnold (Cambridge: Cambridge University Press)

Pappamihiel, N. E. (2002). English as a second language students and English language anxiety: issues in the mainstream classroom. Res. Teach. Engl. 36, 327–355. doi: 10.1023/A:1013074218134

Park, G. P., and French, B. F. (2013). Gender differences in the foreign language classroom anxiety scale. System 41, 462–471. doi: 10.1016/j.system.2013.04.001

Popay, J., Roberts, H., Sowden, A., Petticrew, M., Arai, L., Rodgers, M., et al. (2006). Guidance on the conduct of narrative synthesis in systematic reviews. A product from the ESRC methods Programme (version 1) . Lancaster: Lancaster University.

Pyun, D. O., Kim, J. S., Cho, H. Y., and Lee, J. H. (2014). Impact of affective variables on Korean as a foreign language learners’ oral achievement. System 47, 53–63. doi: 10.1016/j.system.2014.09.017

Resnik, P., and Dewaele, J. M. (2020). Trait emotional intelligence, positive and negative emotions in first and foreign language classes: a mixed-methods approach. System 94:102324. doi: 10.1016/j.system.2020.102324

Rose, H., Syrbe, M., Montakantiwong, A., and Funada, N. (2020). Global TESOL for the 21st century: teaching English in a changing world . Bristol, Blue Ridge Summit: Multilingual Matters

Satio, Y., Garza, T. J., and Horwitz, E. K. (1999). Foreign language reading anxiety. Mod. Lang. J. 83, 202–218. doi: 10.1111/0026-7902.00016

Sellers, V. D. (2000). Anxiety and Reading comprehension in Spanish as a foreign language. Foreign Lang. Ann. 33, 512–520. doi: 10.1111/j.1944-9720.2000.tb01995.x

Spielberger, C. D. (1972). “Conceptual and methodological issues in anxiety research” in Anxiety: current trends in theory and research . ed. C. D. Spielberger (New York, NY: Academic Press)

Steinberg, F. S., and Horwitz, E. K. (1986). The effect of induced anxiety on the denotative and interpretative content of second language speech. TESOL Q. 20, 131–136. doi: 10.2307/3586395

Sun, P. P., and Teng, L. S. (2021). Why so nervous? Revisiting the sources of speech anxiety in Chinese as a second language. System 103:102647. doi: 10.1016/j.system.2021.102647

Tsai, P. S., and Liao, H. C. (2021). Students’ progressive behavioral learning patterns in using machine translation systems– a structural equation modeling analysis. System 101:102594. doi: 10.1016/j.system.2021.102594

Tsang, A., and Lee, J. S. (2023). The making of proficient young FL speakers: the role of emotions, speaking motivation, and spoken input beyond the classroom. System 115:103047. doi: 10.1016/j.system.2023.103047

Veenstra, R., and Weaver, C. (2022). Examining a continuum of FL speaking anxiety over time in an EFL classroom in Japan. System 110:102889. doi: 10.1016/j.system.2022.102889

Vogely, A. J. (1998). Listening comprehension anxiety: students’ reported sources and solutions. Foreign Lang. Ann. 31, 67–80. doi: 10.1111/j.1944-9720.1998.tb01333.x

Wang, H., Peng, A., and Patterson, M. M. (2021). The roles of class social climate, language mindset, and emotions in predicting willingness to communicate in a foreign language. System 99:102529. doi: 10.1016/j.system.2021.102529

Williams, J. E. (1996). Gender-related worry and emotionality test anxiety for high-achieving students. Psychol. Schools 33, 159–162. doi: 10.1002/(SICI)1520-6807(199604)33:23.0.CO;2-M

Wong, G., Greenhalgh, T., Westhorp, G., Buckingham, J., and Pawson, R. (2013). RAMESES publication standards: meta-narrative reviews. BMC Med. 11, 11–20. doi: 10.1186/1741-7015-11-20

Woodrow, L. (2006). Anxiety and speaking English as a second language. RELC J. 37, 308–328. doi: 10.1177/0033688206071315

Woodrow, L. (2011). College English writing affect: self-efficacy and anxiety. System 39, 510–522. doi: 10.1016/j.system.2011.10.017

Yim, S. Y. (2014). An anxiety model for EFL young learners: a path analysis. System 42, 344–354. doi: 10.1016/j.system.2013.12.022

Young, D. J. (1986). The relationship between anxiety and foreign language Oral proficiency ratings. Foreign Lang. Ann. 19, 439–445. doi: 10.1111/j.1944-9720.1986.tb01032.x

Young, D. J. (1991). Creating a low-anxiety classroom environment: what does language anxiety research suggest? Mod. Lang. J. 75, 426–437. doi: 10.2307/329492

Young, D. J. (1992). Language anxiety from the foreign language Specialist's perspective: interviews with Krashen, Omaggio Hadley, Terrell, and Rardin. Foreign Lang. Ann. 25, 157–172. doi: 10.1111/j.1944-9720.1992.tb00524.x

Zare, J., Karimpour, S., and Delavar, K. A. (2022). The impact of concordancing on English learners’ foreign language anxiety and enjoyment: an application of data-driven learning. System 109:102891. doi: 10.1016/j.system.2022.102891

Zhang, L. J., Thomas, N., and Qin, L. T. (2019). Language learning strategy research in system: looking back and looking forward. System 84, 87–92. doi: 10.1016/j.system.2019.06.002

Zuniga, M., and Simard, D. (2022). Exploring the intricate relationship between foreign language anxiety, attention and self-repairs during L2 speech production. System 105:102732. doi: 10.1016/j.system.2022.102732

Keywords: foreign language anxiety, foreign language learning, English as a foreign language, foreign language learners, literature review

Citation: Yu Q (2024) Foreign language anxiety research in System between 2004 and 2023: looking back and looking forward. Front. Psychol . 15:1373290. doi: 10.3389/fpsyg.2024.1373290

Received: 19 January 2024; Accepted: 09 April 2024; Published: 22 April 2024.

Reviewed by:

Copyright © 2024 Yu. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Qiangfu Yu, [email protected]

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

The geography of corporate fake news

Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliation Nanyang Business School, Division of Accounting, Nanyang Technological University, Singapore, Singapore

Contributed equally to this work with: Aixin Sun, Wee Peng Tay

Roles Conceptualization, Funding acquisition, Software, Writing – review & editing

Affiliation School of Computer Science and Engineering, Nanyang Technological University, Singapore, Singapore

Roles Conceptualization, Funding acquisition, Writing – review & editing

Affiliation School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore, Singapore

Alper Darendeli,
Aixin Sun,
Wee Peng Tay

Published: April 17, 2024
https://doi.org/10.1371/journal.pone.0301364
Reader Comments

Although a rich academic literature examines the use of fake news by foreign actors for political manipulation, there is limited research on potential foreign intervention in capital markets. To address this gap, we construct a comprehensive database of (negative) fake news regarding U.S. firms by scraping prominent fact-checking sites. We identify the accounts that spread the news on Twitter (now X) and use machine-learning techniques to infer the geographic locations of these fake news spreaders. Our analysis reveals that corporate fake news is more likely than corporate non-fake news to be spread by foreign accounts. At the country level, corporate fake news is more likely to originate from African and Middle Eastern countries and tends to increase during periods of high geopolitical tension. At the firm level, firms operating in uncertain information environments and strategic industries are more likely to be targeted by foreign accounts. Overall, our findings provide initial evidence of foreign-originating misinformation in capital markets and thus have important policy implications.

Citation: Darendeli A, Sun A, Tay WP (2024) The geography of corporate fake news. PLoS ONE 19(4): e0301364. https://doi.org/10.1371/journal.pone.0301364

Editor: Yasuko Kawahata, Rikkyo University, JAPAN

Received: October 30, 2023; Accepted: March 8, 2024; Published: April 17, 2024

Copyright: © 2024 Darendeli et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: We have publicly shared our manually collected and classified corporate fake news dataset. In order to comply with Twitter’s Terms of Service, we plan to share only the IDs of the accounts involved in our study, as done in the majority of research papers that are based on Twitter data. The firm-related data underlying the results presented in the study requires subscription and are mostly available from WRDS ( https://wrds-www.wharton.upenn.edu/ ). The data underlying the findings are available from https://github.com/alperdarendeli/corporatefakenews .

Funding: Initials of the authors who received each award: AD, SA, WPT Grant number: NTU-ACE2019-02 The full name of the funder: Nanyang Technological University Accelerating Creativity and Excellence Funding URL of funder website: https://www.ntu.edu.sg/research/research-careers/accelerating-creativity-and-excellence-(ace) The funder did not play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

1. Introduction

The last decade has witnessed an unprecedented proliferation of fake news on social media [ 1 ]. The 2016 and 2018 U.S. elections demonstrated the vulnerability of domestic politics to false stories originating from foreign countries, and the World Economic Forum (WEF) identifies massive and systematic digital disinformation as one of the top global risks in its 2019 Global Risks Report [ 2 ]. A rich academic literature also examines the effects of fake news and foreign interference on politics [ 3 – 6 ]. Yet relatively few papers have considered the role of foreign actors in spreading corporate fake news about a country’s firms.

We define “corporate fake news” as negative false information spread about a company and later denied by a credible source (see Table 1 for several examples of corporate fake news) [ 7 ]. We focus on the dissemination of fake news on Twitter (now X), since this prominent social media platform lends itself to analyzing economic and financial issues. For example, 71% of the Twitter users in the U.S. have reported getting news on the site, and 48% of institutional investors use social media to “read timely news” [ 8 ]. The most popular fake news stories have been widely shared on Twitter [ 9 ], spreading farther, faster, more deeply, and more broadly than true news [ 10 ]. Research has also shown the use of fake news to manipulate stock prices in capital markets [ 11 ]. Indeed, a J. P. Morgan Chase director cites “a combination of domestic political groups, analysts and foreign actors who are amplifying negative headlines to sow discord and erode faith in markets” [ 12 ], and a recent article points out the potential role of foreign actors in a disinformation campaign against the Pfizer/Biontech COVID-19 vaccine [ 13 ]. Foreign actors can use the growing importance of social media (and the ease with which it can be manipulated) to amplify the effect of fake news on firms, erode confidence in capital markets, and distort efficient resource allocation.

PPT PowerPoint slide
PNG larger image
TIFF original image

https://doi.org/10.1371/journal.pone.0301364.t001

Despite its importance, however, there are few large-scale empirical investigations of corporate fake news and its geographical origins. What are the characteristics of corporate fake news on social media? Do fake news stories pertain only to accounting/financial issues, or are they also related to a wider range of issues such as the politics, products, or operations of a firm? To what extent do these fake rumors circulate on Twitter? And most importantly, what are the geographic locations and characteristics of Twitter users who start the rumors?

In answering these questions, we face two primary challenges: (i) systematically identifying corporate fake news, and (ii) predicting the locations of rumormongers. To tackle the first challenge, we collect a comprehensive sample of (verified) corporate fake and non-fake news from prominent fact-checking organizations (i.e., Snopes.com , Factcheck.org , Politifact.com , and Truthorfiction.com ). We automatically scrape the websites and use a mix of automated and manual methods to link the news to related companies. We hire and train human coders to manually match fact-checked news to firms and classify their contents to identify the topics in the text. Our approach allows us to identify not only financial news, but also news that captures other aspects of a firm’s attributes (e.g., religion, founders, products, politics, etc.). We identify 541 (144) corporate fake (non-fake) news stories about 126 (67) unique firms between January 2012 and June 2021. We also identify the source of the fake news stories by following the citation trail and find that 42.51% of the corporate fake news is initially seeded in social media (e.g., Twitter, Facebook), followed by news sites (13.68%). The news also spans a variety of topics, including firms’ politics (37.7%), products (22.6%), operations (16.8%), and founders/executives (6.8%).

To tackle the second challenge, we search mentions of the fact-checked news on Twitter to identify the country locations of users who spread the news. However, the location of Twitter users is hard to pin down, as most users do not voluntarily (or accurately) disclose their location: indeed, they may enter irrelevant text in the data field or intentionally manipulate their locations to obfuscate the origin of their tweets [ 14 ]. To overcome this challenge, we employ a machine-learning model to predict the locations of rumormongers on Twitter. We use a comprehensive global sample of around 4 million geo-tagged tweets to train a location-prediction model developed in the computer science literature [ 15 , 16 ]. We use geo-tags in tweets as reliable ground-truth data because geographic coordinates are appended to tweets based on the location of the mobile device. Specifically, we use tweet text in combination with metadata (i.e., tweet language, user-declared location, user name, and user description) to train a recurrent neural network (RNN) to predict the location of geo-tagged tweets. We concatenate the features in a text, represent them as word embeddings, and use long short-term memory (LSTM), a model that retains longer-range dependencies in text sequences [ 16 , 17 ]. Although more advanced models exist for text classification tasks in general, given our problem setting, LSTM is one of the most suitable models given its effectiveness and low cost compared to more advanced models. We split our geo-tagged data into training, validation, and test sets, and we evaluate the accuracy of the model in a test set using a variety of model architectures, features, and hyperparameters (see S1 Appendix in S1 File for details). Our fine-tuned model’s predictive accuracy is 88.76%. In other words, we can correctly predict the location of 88.76% of Twitter users’ countries based on their tweets and metadata. This accuracy is comparable to that of other location-prediction models in the literature.

Geotagging, however, requires the consent of users, and only a small proportion of tweets (around 1%) are geo-tagged. Therefore, we use the trained model to infer the locations of Twitter users without geo-tagging data. This approach allows us to infer the home location of all the users in our sample. Of the 685 fact-checked corporate news items, we identify 294 (87) fake (non-fake) stories that disseminate on Twitter. Using the trained model, we find that corporate fake news is more likely than corporate non-fake news to be initiated by non-US (foreign) accounts. The difference between the percentage of fake and non-fake corporate news originating from foreign accounts (37.56% versus 30.84%) is statistically significant ( t = 2.30, p <0.05). The accounts that spread fake news are relatively new and have more followers, and the fake news is retweeted more than the non-fake news. We also compute the percentage of users with bot-like behavior, as bots can be used to disseminate low-credibility information [ 18 ]. We find that the users who spread fake news are more likely to exhibit bot-like characteristics ( t = 36.48, p <0.00).

After introducing the data, we present several stylized facts about the geographical origins of fake news (at the country level) and the determinants of a firm being targeted by fake news (at the firm level). At the country level, we measure the percentage of Twitter users in a country who are spreading the fake news and find that fake news is more likely to originate from African and Middle Eastern countries. The top five countries spreading corporate fake news are Oman, Jordan, Morocco, Qatar, and Lebanon. In contrast, non-fake news is most likely to originate from Western countries (e.g., Austria, Finland, Poland, and Denmark). We also use two metrics for distributional analysis of geographical locations. First, we use Kullback-Leibler (KL) divergence as a non-parametric approach to compare the “distance” from distribution of the percentage of fake and non-fake news spread from individual countries. The KL divergence is 0.62 and statistically significant ( SE = 0.18, 95% CI[0.27,0.97]), suggesting that the set of countries spreading corporate fake news is different from the set of countries spreading non-fake news. Second, we compute median relative polarization (MRP) to compare the concentration of countries spreading fake and non-fake corporate news. We find that MRP is 0.33 and statistically significant ( SE = 0.11, 95% CI[0.12,0.54]), which suggests that fake news is most likely to originate from a concentrated set of countries. Finally, we show that foreign-originating fake news increases during periods of heightened geopolitical risk.

Second, we estimate firm-level regressions to explore the characteristics of firms targeted by fake news. The results of our tests suggest that two factors can explain part of the variation in exposure to foreign-originating fake news. First, ideological motivations can drive foreign actors to spread misinformation about a country’s firms. The last decade has witnessed a proliferation of cyberattacks by nation-states on the strategic industries of foreign countries. Consistent with this, we find that firms in strategic industries (i.e., the telecommunication, pharmaceutical, semiconductor, computer, and defense industries) are more likely to be targeted by foreign-originating fake news. Second, we find that firms operating in uncertain information environments are more prone to foreign fake news, consistent with information frictions slowing the price-discovery process, which may cause prices to deviate from intrinsic values for prolonged periods of time and create profit opportunities for rumormongers [ 11 ].

Our study makes several contributions. First, we contribute to the debate about fake news and misinformation on social media. While research shows that social media is the main conduit through which rumors propagate in the political sphere [ 7 ], little work has been done on the geographical origins of rumors in capital markets. By allowing us to infer the geographical locations of rumor starters, our methodology has the potential to inform policymakers on whether foreign influence operations in the political sphere can carry over to the economic domain and capital markets.

Second, we add to the growing literature on information acquisition in the era of social media. Social media platforms can facilitate price discovery by allowing for direct information transfer between firms and consumers/investors [ 19 – 23 ], but the anonymity of users also provides a breeding ground for misinformation and distorts price discovery [ 24 ]. Indeed, recent work shows that firms may disseminate truthful negative information about their competitors on social media [ 25 ]. In contrast, our focus is on the negative fake news spread about firms on social media.

Finally, location-prediction models have been gaining popularity [ 15 ], and researchers have recently used geographic online social networks to estimate population movement patterns [ 26 , 27 ], forecast economic activity [ 28 ], and monitor political events [ 29 ], public health [ 30 , 31 ], the spread of diseases [ 32 ], and conspiracy theories [ 33 ]. In this work, we use the location-prediction model in the context of fake news dissemination in capital markets.

An important caveat to our study is the descriptive nature of our research design, which may preclude causal interpretation. Nevertheless, it is worthwhile, even at the descriptive level, to conduct a large-scale content analysis of corporate fake news, along with an exploratory analysis of its dissemination and the role of foreign actors on social media. In addition, even if we can identify the foreign origin of a fake news story, attributing it to a foreign state actor or an intentional disinformation operation remains difficult. First, Virtual Private Networks and third-party proxies may mask foreign actors’ identities and locations, making tracking their activity incredibly difficult. Second, the use of vast networks of new and hijacked accounts across multiple platforms complicates attribution as these campaigns adapt and spread rapidly [ 34 , 35 ] Additionally, platform limitations in data access and analysis capabilities restrict researchers’ ability to pinpoint the source of rumor. Hence, we cannot make any claims about the intent of the users disseminating such news on social media (e.g., whether they are knowingly or unwittingly spreading disinformation, or just joking about the story). While the increased foreign activity during periods of high geopolitical risk and the targeting of strategic industries may provide some clues, it is often difficult to draw conclusions about the originators’ underlying motives. The dynamic nature of social media platforms could also generate some interesting patterns for further exploration. For example, rumormongers may not only initiate rumors but also amplify the fake news (e.g., via retweets or replies to tweets) initiated by another source. It is important to acknowledge that our study focuses solely on the original tweets spreading claims verified by a fact-checking organization. Finally, our dataset includes only rumors that were investigated by fact-checking organizations (and probably excludes less viral rumors), which may lead to a selection bias in the collection of fake news.

2. Data and methods

2.1. corporate fake news.

We build a comprehensive database of news about U.S. firms that has been debunked by prominent fact-checking organizations (Snopes.com, Factcheck.org, Politifact.com, and Truthorfiction.com). Each fact-checking site has its own classification scheme (e.g., Snopes.com classifies articles into six categories, whereas Politifact.com has nine categories). We normalize the verdicts across different sites by mapping these classes to fake and non-fake news categories (see S2 Appendix in S1 File for details), and we do not include mixed news in the analysis (i.e., news classified as neither true nor false). In doing so, we aim to identify a broad spectrum of corporate fake news over a long time period. Whereas previous studies focus only on the financial information of public companies [ 11 , 36 ], our approach allows us to identify a comprehensive sample of fake news that targets businesses but is not necessarily financial in nature.

First, we automatically scrape the fact-checking websites to collect all the fact-checked articles and parse the publication date, title, claim, body of the text, and fact-checking ratings for each piece of news. Because a naïve company name keyword search might not fully capture a firm’s products, executives, or subsidiaries (e.g., news about Oreo Cookies may be linked to Mondelez International), we use Named Entity Recognition (NER) (i.e., NLTK, TextBlob, and SpaCy) methods to create a filtered sample of news potentially related to firms. NER methods help us identify an initial subset of news referring to a company (or product) such as KFC, Facebook, or Pfizer or a person such Bill Gates or Steve Jobs. However, because the automated algorithms are trained on non-business textual data, they may generate false positives [ 37 ]. Therefore, we manually read the filtered subsample to exclude any non-firm-related news. After reading 100 randomly selected articles, we develop the labeling rules presented in S3 Appendix in S1 File . For example, if the news is about a CEO’s arrest, we include the news in our sample because of its importance to the firm’s operations (see, e.g., https://www.snopes.com/fact-check/italy-bill-gates-arrest/ ). However, if the news is about a CEO’s private life (or her charitable activities), we do not include it in our sample (see, e.g., https://www.snopes.com/fact-check/bill-gates-planned-parenthood/ or https://www.politifact.com/factchecks/2020/may/14/facebook-posts/no-evidence-gates-foundation-will-profit-coronavir/ ). We also exclude conspiracy theories, satire, and false statements by politicians [ 7 ] and keep only the negative-sentiment news, using the sentiment score determined by the word list in [ 37 ] (e.g., we classify news as negative if it contains more negative than positive words and manually read the news to determine the sentiment if the difference between positive and negative words does not exceed 1% of total words).

Second, we manually classify news content to identify the specific topics in the text. We prefer manual annotation (instead of automated annotation algorithms) to develop a deeper understanding of the features of the text given the high level of domain expertise required in our setting. To do this, we start by determining categories based on 50 randomly selected subsets of news. Then, we train three research assistants to annotate the same text independently and compare results. After ensuring the degree of consensus regarding the topics, research assistants continue to independently annotate a larger set of news and identify major categories of news topics. When the research assistants disagree (on a new topic category), the authors discuss the news before reaching an agreement. This way, we identify a broad range of topics including firms’ products and services, operations, data privacy, politics, and founders and executive management. Table 1 shows several examples of corporate fake news classified into the various topical categories.

Finally, we identify the origin of a story by relying primarily on URL links mentioned in fact-checking articles. Fact-checking articles often discuss and cite the source of a claim while debunking the claim. We manually read each URL link in fact-checked articles to determine the origin of the claim and identify its publication date. If there is more than one relevant source, we keep the earliest published source.

2.2. Twitter data

We collect historical tweet- and user-level data about corporate news using Twitter Academic API. The API grants academics full access to historical tweets dating back to 2006 (except for deleted tweets and accounts). We identify the original (or source) tweets spreading fake and non-fake corporate news in two steps. First, we collect all the tweets that contain a link to a fact-checking website that evaluates the veracity of a corporate news story. We exclude the original tweets containing a link to a fact-checking website, because our goal is to identify the spread of unverified and contested information, not information verified by fact-checking organizations. The remaining tweets are replies to an original tweet or replies to replies. We also remove tweets that do not directly reply to an original tweet to ensure that a reply containing a link to a fact-checking website is in fact addressing the original tweet.

Second, we extract the URL links to external articles mentioned in the above original tweets (spreading fake and non-fake news) or the URL links mentioned in the fact-checking articles. After manually reading the extracted articles, we identify the URL links about the (fake and non-fake) news. We then extract the original tweets containing a link to any of these articles (which are not necessarily fact-checked through replies). In our search, we transform links to canonical URLs by removing http://, https:// and analytic tracking parameters (i.e., Urchin Tracking Module parameters), and we transform short URLs to the expanded form (to merge different links referring to the same article). Through this process, we identify a sample of original tweets mentioning a fact-checked corporate news story on Twitter. We define the sender of the original tweet (i.e., the source of the Twitter cascade) as the rumor source.

These sample filters leave us with 342,818 original tweets (i.e., no retweets or replies) mentioning corporate news verified by a fact-checking organization. Tweet-level variables include the tweet text, timestamp, tweet language, and number of retweets. We also separately query for author-level data—user name, user description, the number of followers, the number of followees, and account creation date—for each unique Twitter user authoring one of the collected tweets. We obtain data for 189,158 unique Twitter users. Finally, we manually match Twitter data to other academic datasets (e.g., COMPUSTAT) using company names. Our code and dataset are available at https://github.com/alperdarendeli/corporatefakenews . The collection and analysis methods comply with the terms and conditions for the source of the data. Our study was also reviewed and approved by the Institutional Review Board of Nanyang Technological University (IRB-2022-349).

2.3. Twitter location-prediction model

The geolocation of a rumor source on social media is hard to pin down, either because users do not disclose their home location or do not enter data that correspond to their actual locations [ 14 ]. For example, users frequently enter fake locations or sarcastic comments in their profiles (e.g., Jupiter, Outta Space, Out of this world, etc.), making it difficult to infer user location solely from self-declared profile data. The profile locations could also be inaccurate because individuals choose not to publicly share their country of location or intentionally obfuscate the origin of their tweets, as is frequently the case with user-generated input. Hence, location information on Twitter is often far from complete and reliable. To tackle this issue, we employ an LTSM model for location prediction, which is particularly suitable for processing reasonably long and sequential text data [ 15 ].

The LSTM model is a type of RNN that can utilize information further back in a sequential chain. To do this, LSTM models use mechanisms called gates that regulate the flow of information being passed from one step to the next. Unlike a vanilla RNN, LSTM includes a forget gate that determines the sort of information that is passed across the sequence. The operations within an LSTM allow the model to keep relevant information (and forget irrelevant information) from the previous steps no matter what the length of the sequence is. Thus, LSTM models do not suffer from the vanishing gradient problem (i.e., the short-term memory problem). This makes LSTM suitable for processing sequential data with long-term dependencies (such as text). We use the model to predict the geographic location of Twitter users [ 16 ].

For prediction, we use tweet content in combination with user metadata because the recent literature shows that the metadata can contribute substantially to predictive accuracy and provide valuable location signals [ 15 ]. We use a large sample of English and non-English geolocated tweets between 2014 and 2019 as our labeled dataset. We use the locations in geo-tagged tweets as our ground-truth because they are based on the GPS coordinates of mobile devices, which are reliable and difficult to manipulate. The training data consist of 3,927,563 geotagged tweets covering 149 countries and 2,187 cities (see S1 Appendix in S1 File for details).

First, we extract the following text features from the geotagged tweets: tweet text, tweet language, user-declared location, user description, and user name. Other features such as time zone, UTC offset, URL links, and messenger source (e.g., iPhone or Android) can also help predict locations. Time zone and UTC offset, however, cannot be extracted from Twitter API at the time of our study ( https://twittercommunity.com/t/upcoming-changes-to-the-developer-platform/104603 ). The previous literature also shows the limited benefit of URL links and messenger source for the prediction task [ 16 ]. We then clean the text by (i) removing links, user name, punctuation, and extra spaces, (ii) separating emoticons, (iii) making all text lower case, and (iv) concatenating the tweet features. We concatenate as follows, inserting special tokens at the front of each text field. A [BLANK] token is also inserted if the specific text field is blank.

‘[TEXT] <cleaned_text> [LANG] <tweet_lang> [LOC] <cleaned_user_declared_location> [DESC] <cleaned_user_description> [NAME] <cleaned_user_name>‘

We then feed the concatenated text into an LSTM model to predict geographic locations. By and large, we follow the model architecture in [ 16 ] and pose the geolocation-prediction task as a multi-label classification task of 2,187 cities. In the model, the concatenated text is represented as word embeddings because it better captures words with similar locational semantics in low-dimensional vectors (and produces a more efficient representation of words than one-hot encoding). After transforming the features into machine-readable vectors, we train an LSTM model with a prediction layer. We split 80% of the data into training, 10% into validation, and 10% into test sets. We use stratified sampling to ensure that the geographical distribution of each set is approximately equal. The model learns a weight (or coefficient) for each tweet feature and uses Adam optimization to minimize cross-entropy loss over all possible weights. We fine-tune the model using different hyperparameters on the validation set. Table 2 , Panel A reports the selected model hyperparameters. S1 Appendix in S1 File provides technical details about data pre-processing, parameter tuning, and model training steps. The model predicts the geographic location of a tweet at the city level (after which we map the prediction to the related countries). We do not use city-level predictions because the predictive accuracy at the city level is much lower than the accuracy at the country level (51.4% versus 88.8%). And, more importantly, our research question, which examines the prevalence of fake news from foreign accounts, is cast at the country level.

https://doi.org/10.1371/journal.pone.0301364.t002

We then evaluate the accuracy of the model at the country level using the test set.

The individual performance of different features is presented in Table 2 , Panel B. If we had always predicted that the United States would be the most frequently observed country in the training data, our prediction would be 21% correct. As a simple baseline, we assess our model’s performance against this majority predictor. We start our bottom-up analysis with individual tweet features. We find that tweet text is the most relevant feature for location prediction. Using the text alone, we can correctly predict the location of 72.48% of all tweets. Consistent with the prior literature, we find that augmenting text with user metadata improves accuracy. For example, combining text with user-declared location improves accuracy to 86.34%. When we concatenate all features (tweet text, user-declared location, user description, and user name), the model achieves the best predictive accuracy of 88.76%.

Next, we perform a leave-one-out feature importance analysis to rank the factors in terms of their contribution to model performance. The analysis relies on the idea that if a feature is not important, excluding it from the predictor set should not noticeably decrease the model’s out-of-sample performance. We iteratively remove one feature at a time and evaluate accuracy. Specifically, we calculate the decrease in the predictive accuracy when a feature is excluded from the predictor set. We then scale the decrease by the predictive accuracy when all predictors are used. Surprisingly, while tweet text separately has the largest influence on the prediction of user locations, we find, in Fig 1 , that user-declared location is the top explanatory predictor in the feature importance analysis. This implies that user-declared location may not carry a large weight on its own, but combining it with other features might help improve the predictive accuracy of the model.

https://doi.org/10.1371/journal.pone.0301364.g001

Overall, our model significantly outperforms a naïve majority predictor in predicting the geographic locations of tweets. Our model accuracy is also comparable to that of other location-prediction models in the literature. For example, a maximum entropy classifier [ 38 ] predicts the country origins of tweets with 88.90% accuracy using tweet language, user-declared location, user language, time zone, offset, user name, user description, and tweet text. A linear classifier model [ 39 ] predicts the country origins of tweets with 87.34% accuracy using tweet text, profile location, time zone, and time (in UTC time). Our model’s predictive accuracy is comparable despite our inability to use additional features such as time zone, offset, and user language, which were not accessible via Twitter API at the time of this study.

As the authors of [ 38 ] show that geo-located and non-geolocated tweets have similar characteristics, we can use a model trained on geolocated tweets to predict the locations of users without geo-tagged data. Therefore, we use our fine-tuned model to predict the locations of tweets disseminating fact-checked corporate news. Fig 2 illustrates the location-prediction procedure at the user level. We use metadata and multiple tweets of each user to predict the user’s country location.

This figure illustrates the location prediction of a Twitter user. We implement a two-step procedure. First, we use our model to predict locations at the tweet level. Specifically, we apply our fine-tuned model’s weights to produce a score (or multiclass logit) for each tweet location. Second, we take the average probabilities of tweet location at the user level to predict the country with the highest score as the location of the Twitter user.

https://doi.org/10.1371/journal.pone.0301364.g002

We apply the fine-tuned model’s weights to our sample to produce a score for each tweet’s location. Specifically, we use the softmax function to convert these weights to multiclass logits (or probabilities) for each city and country. For each tweet, the model generates a vector of probabilities Pr( Cu k ) = {pr( C 1uk ), pr( C 2uk ),…, pr( C Luk )}, where pr( C 1uk ) is the probability that the user u k will be assigned to country 1, and so on. Then, we take the average probabilities of tweet locations at the user level and predict that the Twitter user is located in the country with the highest score.

3. Results and discussion

3.1 empirical construct.

We use the model’s predictions to construct Foreign Corporate Fake News (%) as the key empirical construct of our study. Our model predicts the geographical locations of Twitter users across 139 countries. We exclude, however, the countries where Twitter is blocked (i.e., China, North Korea, Russia, Iran, Uzbekistan, Turkmenistan, and Belarus), which account for 0.62% of the users in our sample. Our findings are robust to including these countries in the analysis.

We categorize a corporate fake news story as foreign originated if its rumor source (i.e., the Twitter user who initiates the cascade) is outside the U.S. To put it differently, we define a continuous variable— Foreign Corporate Fake News (%) —as the percentage of original tweets spreading fake news initiated by an account in a foreign country. For example, if we have 10 original tweets spreading a fake news story, and five of the tweets are initiated by foreign accounts, the value of Foreign Corporate Fake News (%) is 50%. The measure accounts for the fact that fake news can be spread via multiple cascades on Twitter. In robustness tests, we replace Foreign Corporate Fake News (%) with Foreign Corporate Fake News (dummy) , as binning the data can reduce measurement errors caused by a noisy continuous variable [ 40 ]. We define Foreign Corporate Fake News (dummy) as an indicator variable equal to one if at least one Twitter account spreading fake news is located in a foreign country, and zero otherwise. We also use corporate non-fake news to construct a corresponding benchmark. We define Foreign Corporate Non-fake News (dummy) as an indicator variable equal to one if at least one Twitter account spreading non-fake news is located in a foreign country, and zero otherwise. Similarly, Foreign Corporate Non-fake News (%) is the percentage of original tweets spreading non-fake news initiated by an account outside the U.S. We construct the measures at the news level. For much of the analysis, however, we aggregate the measures to the country-year (Section 3.3) and firm-year (Section 3.4) level.

3.2. Sampling design and summary statistics

We begin our analysis by reporting descriptive statistics. First, we report data across fake and non-fake corporate news. We identify 541 fake news stories (about 126 unique firms) and 144 non-fake news stories (about 67 unique firms) over the period from January 2012 to June 2021. Within each news category, we tabulate news content and provide statistics about the characteristics of rumor starters. Table 3 , Panel A presents the results. Most of the fake news revolves around the firm’s politics, product, or operations. 37.7% of corporate fake news involves political discussions (e.g., support for social movements such as Black Lives Matter, funding of political parties, government contracts, etc.), whereas product-related news (e.g., contaminated products, product discontinuation, product safety, etc.) accounts for 22.6%, and operations-related news (e.g., downsizing, employee policy, etc.) accounts for 16.8% of the fake news. The distribution of topics for non-fake news follows a similar pattern, and the difference across news categories is not statistically significant.

https://doi.org/10.1371/journal.pone.0301364.t003

In Panel B, we identify the individual characteristics of the rumor starters. We use Twitter metadata to extract user profiles. Specifically, we check the number of people who follow the user on Twitter (i.e., followers), the number of people whom the user follows in Twitter (i.e., followees), the age of the user’s account (measured in years), and the percentage of bot accounts. We find that starters of fake news (i.e., Twitter users who initiate the news) have more followers than starters of non-fake news. They are also relatively new accounts. We also use Botometer, a popular public tool, for bot detection on Twitter (see S4 Appendix in S1 File for details). Using this tool, we calculate a Bot Activity score between zero and one for each Twitter user. A higher score indicates a higher likelihood that a Twitter account is a bot. We find that bot score is higher for fake news accounts (0.51 versus 0.46), suggesting that fake news spreaders are more likely to exhibit bot-like behavior. We also find that bot-like activity is significantly higher for fake news than for non-fake news, which indicates that a large group of automatically operated accounts may be promoting the fake news ( t = 36.48, p <0.00). Finally, we find that corporate fake news, on average, is retweeted around 480 times, whereas non-fake news is retweeted around 20 times. This finding is consistent with the faster diffusion of fake news on social media, as documented by [ 10 ].

In Panels C and D, we report the distribution of news categories across industries and over time. At the industry level, the wholesale, retail, business equipment, and consumer nondurable industries are more likely to be targeted by fake news (columns 1 and 2). Fake news, however, is more likely to be foreign originated in the finance, chemicals, and consumer durables industries (column 3). In temporal trends, the fake news peaks in 2015, after which it gradually starts to decline (Panel D). Panel E reports the distribution of news by fact-checking organizations. Most of the fact-checked news is collected by Snopes (67%). We also calculate the number of days it takes for a fact-checking organization to check a claim. To do this, we manually read and identify the original date of the claim (i.e., the source news). We then subtract the publication date of the fact-checking article from the date when the source news began to disseminate news in the public domain. We find Politifact to be faster than other organizations in fact-checking the claims. We also manually identify the platforms where the source news is disseminated. Panel F reports the results. We find that the fake news is disseminated mostly on social media (e.g., Twitter, Facebook, Youtube). In contrast, non-fake news originates mostly from news sites, consistent with the filtering role that the editorial process plays in traditional media.

3.3 Country-level analysis

We next explore the distribution of corporate news at the country level. Our analysis is motivated by the fact that ideologically motivated foreign actors may spread misinformation to attack the reputation of a country’s firms. For example, a recent study finds evidence of social media manipulation campaigns in 70 countries organized by government agencies to shape public attitudes [ 41 ]. Several state actors target foreign countries to influence global audiences, amplify hate speech, or harass political figures or journalists, and countries use troll farms in the Middle East and Africa to spread rumors about target countries (see, e.g., https://www.theguardian.com/technology/2020/mar/13/facebook-uncovers-russian-led-troll-network-based-in-west-africa .) The country-level analysis may help shed light on the geographical origin of rumor spreaders (and potential foreign involvement) in capital markets.

Because the number of Twitter users varies across countries, we begin our analysis by normalizing the number of fake (and non-fake) news stories originating from a country by the total number of users in that country in a given year. We use the number of users in the randomly collected geo-tagged tweet dataset as a proxy for the total number of users in that country (as we could not obtain the historical country-level user statistics from Twitter or external data providers). To reduce outlier effects, we include countries with at least 300 tweets in the training data. The top five countries with the highest number of Twitter users are the United States, Indonesia, Brazil, Turkey, and Great Britain. We take the average of the yearly normalized Foreign Corporate Fake News (%) at the country-year level to construct our measure at the country level.

Table 4 reports the results. In Panel A, we show that corporate fake news originates primarily from Middle Eastern and African countries. Oman, Jordan, Morocco, Qatar, and Ghana are the top five countries from which most tweets with corporate fake news originate. The probability of each Omani user originating corporate fake news is 2.01%, while the probabilities for Jordanian and Moroccan users are 1.76% and 1.47%, respectively. In contrast, non-fake news originates mostly from Western countries. The probability of each Austrian user originating non-fake news is 0.63%, while the probabilities for Finnish and Polish users are 0.60% and 0.59%, respectively. In Fig 3 , we plot the geographical distribution of 10 countries spreading fake (Panel A) and non-fake (Panel B) news about U.S. public firms. In Panel B, to account for the general business interests of users from a specific country, we construct an adjusted measure by subtracting the probability of disseminating non-fake news from that of disseminating fake news. Using this measure, we still find that the foreign fake news originates primarily from the Middle East and Africa (i.e., Oman, Jordan, Qatar, and Morocco), consistent with the anecdotal evidence that foreign actors use troll farms in this region. For example, a Russian-led network of professional trolls (operated by local residents) targeting the U.S. was discovered in Ghana and Nigeria [ 42 ]. There is also evidence of Lebanese, Nicaraguan, and Moroccan governments running disinformation campaigns for political motives (see, e.g., https://www.newarab.com/analysis/disinformation-and-electronic-armies-lebanons-elections for Lebanon, see https://www.bbc.com/news/world-latin-america-59129894 for Nicaragua, and see https://www.accessnow.org/how-pro-government-media-in-morocco-use-fake-news-to-target-and-silence-rif-activists/ for Morocco).

This graph plots the geographical location of users spreading corporate news about U.S. public firms. We construct Foreign Fake News (%) and Foreign Non-Fake News (%) at the country level in two steps. First, we normalize the number of users spreading (fake or non-fake) news on Twitter by the total number of users in each country. We then take the averages of this normalized country-year-level measure to convert it to a country-level measure. In Panel A, we plot the top 10 countries spreading corporate fake news about U.S. public firms. In Panel B, we plot the top 10 countries spreading corporate non-fake news about U.S. public firms. The map in the figure is made with Natural Earth ( https://www.naturalearthdata.com/about/terms-of-use/ ).

https://doi.org/10.1371/journal.pone.0301364.g003

https://doi.org/10.1371/journal.pone.0301364.t004

Next, we employ the KL divergence metric to compare the country distribution of the percentage of Twitter accounts spreading fake and non-fake news. KL divergence is a measure of how one probability distribution differs from a second distribution. Intuitively, we examine how far away the geographical distribution of the percentage of accounts spreading fake news is from the geographical distribution of the accounts spreading non-fake news. If the two distributions match perfectly, KL divergence is zero; otherwise, it can take values between zero and infinity. We find that the KL divergence is 0.62 and statistically significant ( SE = 0.13, 95% CI[0.37,0.88]), which suggests that the geographical distribution of fake news is different from that of non-fake news. We also compare the concentration of geographic locations using a median relative polarization index (MRP). A positive MRP implies a more uneven distribution of news across countries relative to a benchmark. We take the distribution of locations spreading non-fake news as our benchmark and compute the MRP. We find that the MRP is relatively higher within countries spreading fake news ( MRP = 0.33, SE = 0.11, 95% CI[0.12,0.54]), suggesting that the distribution of countries spreading fake is more concentrated than that of countries spreading non-fake news.

Finally, we examine if the probability of foreign fake news is more pronounced during periods of high geopolitical risks. To do this, we plot a time series of the news originating from foreign accounts and geopolitical risks. We use two proxies for geopolitical risks. First, we employ a news-based measure of the Geopolitical Risk Index (GPR) developed by [ 43 ]. The authors of [ 43 ] (on page 1195) define geopolitical risks “as the threat, realization, and escalation of adverse events associated with wars, terrorism, and any tensions among states and political actors that affect the peaceful course of international relations.” They construct an index at the country-year level by counting the share of articles mentioning adverse geopolitical events in leading newspapers in the U.S. A higher level of GPR corresponds to escalated geopolitical tensions facing the U.S. Second, we use the Global Database on Event, Location, and Tone (GDELT) to construct an Interstate Conflict Index . GDELT is the most widely used database for studying international relations and conflicts. It contains more than 200 million geolocated events compiled from international news sources [ 44 ]. The sources include AfricaNews, Agence France Presse, Associated Press, BBC Monitoring, United Press International, The Washington Post, The New York Times, and Google News. After machine coding the relevant information in the text of a news story (e.g., related countries, type of event, intensity of conflict or cooperation) into events, GDELT merges all duplicate events into a single event record. GDELT then provides the Goldstein scale [ 45 ], measuring the impact of each event from -10 (most conflictual) to +10 (most cooperative). We calculate the annual average Goldstein scale between the U.S. and other countries and reverse the scale to make it an increasing function of the conflicts that the U.S. faces.

We use these two indices— GPR and the Interstate Conflict Index —to explore the relation between the news originating from foreign accounts and geopolitical tensions. We calculate the annual average of Foreign Corporate Fake News (%) (at the U.S. level) to capture the prevalence of foreign accounts targeting U.S. firms. For ease of interpretation, we standardize Foreign Corporate Fake News (%) and the indices. We plot the temporal evolution of the variables in Fig 4 .

This figure plots the time series of corporate news originating from foreign countries and geopolitical tensions. Foreign Fake News (%) is the percentage of original tweets spreading corporate fake news initiated by a foreign (non-U.S.) Twitter account. We reconstruct this measure by aggregating all the corporate news at the yearly level. Panel A plots the relation between Foreign (Non-fake) Fake News (%) and Interstate Conflict Risk . Interstate Conflict Risk is an index measuring interstate conflicts (based on the Goldstein scale) using daily reported events in the global news media. Panel B plots the relation between Foreign Fake (Non-fake) News (%) and Geopolitical Risk Index . Geopolitical Risk Index is the share of articles mentioning adverse geopolitical events in leading newspapers in the U.S. For ease of interpretation, we standardize the indices and news measures.

https://doi.org/10.1371/journal.pone.0301364.g004

Panel A shows the close comovement of Foreign Corporate Fake News (%) and Interstate Conflict Index . There is a positive correlation (0.40) between the proportion of fake news from foreign accounts and interstate conflict risk. The correlation, however, is weaker (0.17) for Foreign Corporate Non-fake News (%) . We observe a similar pattern in Panel B using GPR Index . The correlation between GPR and Foreign Corporate Fake News (%) is positive (0.09), but that between GPR and Foreign Corporate Non-fake News (%) is negative (-0.48). Overall, the increased fake news originating from foreign countries during periods of high geopolitical tensions suggests a link between foreign accounts and state actors. However, given our data limitations and descriptive research design, we interpret our evidence as merely suggestive.

3.4 Firm-level analysis

It is an empirical question what kind of firms are more likely to be targeted by fake news.

At the country level, we show a comovement between heightened geopolitical risks and foreign-originating fake news. At the firm level, we predict that firms in strategic industries (i.e., the telecommunication, pharmaceutical, semiconductor, military, and computer industries) and firms that are leaders in their industries are more likely to be targeted by foreign actors, given the recent proliferation of foreign-originated cyberattacks that have damaged the reputations of targeted firms in strategic industries (see, e.g., https://nyti.ms/3jsdGSR , https://bit.ly/3twMIho ). Similarly, foreign actors can strategically disseminate fake news on social media to tarnish the reputations of strategically important firms, as reputation is often considered a firm’s most important intangible asset [ 46 ]. A significant body of research shows the effect of reputation on asset prices [ 47 , 48 ], firm sale [ 49 ], risk and financial policy [ 50 ], investor preferences [ 51 ], and consumer behavior [ 52 , 53 ]. Ideologically motivated actors may use Twitter to strategically target important industries, as reiterated news conveyed via multiple channels can reach a wide audience of individual and institutional investors [ 20 ].

To empirically test this prediction, we construct Foreign Fake News (%) by identifying the proportion of fake news originating from foreign accounts for a specific firm in a given year. For example, if 100 accounts initiate fake news about a firm in a given year, and 40 of these accounts are from a foreign country, Foreign Fake News (%) is 40%. We employ both univariate and regression analyses to examine the characteristics of firms targeted by fake news. Table 5 Panel A reports univariate results. Our firm-level analysis does not include 22 (11) private firms that are subject to fake (non-fake) news because we do not have financial data for this subset of firms. In the sample, we have 59 unique public firms (465 firm-years) targeted by fake news and 17 unique public firms (145 firm-years) subject to non-fake news. The firms targeted by fake news are significantly more profitable and have greater foreign sales. Target firms have higher pooled average Return on Assets (ROA) (0.12 versus 0.08) and lower Book-to-Market ratio (0.30 versus 0.50). Industry competition (as measured by TNIC HHI ) also increases the probability of a firm being targeted by fake news. The TNIC HHI of target firms (0.35) is higher than that of firms with non-fake news (0.24). Firms in strategic industries (i.e., pharmaceutical, semiconductor, computer, defense, and telecommunication) are more likely to be targeted by fake news (0.22 versus 0.12). More importantly, the fake news is more likely to originate from foreign accounts. 14.21% of fake news is initiated by a non-US Twitter account versus 6.37% of non-fake news (see S5 Appendix in S1 File for the variable definitions).

https://doi.org/10.1371/journal.pone.0301364.t005

In the model, we include a vector of firm-level characteristics, including size ( Total Assets ), profitability ( Return on Assets) , leverage ( Leverage ), book-to-market ratio ( Book-to-Market ), dividend ( Dividend Dummy ), and a loss dummy ( Loss ). We do not have a clear directional prediction for these variables. To capture a firm’s visibility in foreign markets, we also control for Foreign Sales , defined as a dummy variable equal to one if at least 10% of a firm’s sales are to foreign (non-U.S.) markets. In addition, we control for TNIC HHI and Product Similarity to capture market competition. TNIC HHI is calculated as the sum of the squared market shares of all firms operating in the same industry, using the time-varying Text-based Network Industry Classification (TNIC) developed by [ 55 ]. Product Similarity is a firm-level measure based on product descriptions from 10-K filings [ 55 ]. We include these measures because peer firms can spread disinformation about their rivals, especially in competitive markets [ 25 ]. For example, the authors of [ 25 ] document negative peer disclosure (NPD) as an emerging corporate strategy firms use to publicize adverse news about their industry peers on social media. Moreover, NPD propensity increases with product market rivalry, as highly competitive industries provide greater incentives to spread negative news.

We employ Institutional Ownership and Return Volatility to control for the relation between the information environment and the fake news. Institutional Ownership represents the percentage of a firm’s stock held by institutional investors. Return Volatility captures uncertainty regarding a firm’s underlying fundamentals, which we measure as the standard deviation of a firm’s daily stock returns during a fiscal year. A poor information environment can create incentives for rumormongers to spread fake news, as investors are more likely to be influenced by news when access to alternative information sources is limited. We use these variables to control for the relation between a firm’s information environment and the likelihood of being targeted by fake news.

Table 5 , Panel B summarizes the results. In column 1, we use Foreign Fake News (%) as the dependent variable in the baseline model. In column 2, we estimate a benchmark model using Foreign Non-fake News (%) for comparison. We find that larger firms with foreign sales and growth opportunities are more likely to be targeted by foreign fake news. Column 3 compares the coefficients across the baseline and benchmark models. The coefficient estimates of target firms are also significantly different from those of non-target firms. Second, we find that firms operating in less competitive markets (i.e., industries with higher TNIC HHI ) are more likely to be targeted by foreign fake news. The coefficient estimate on TNIC HHI , however, is not statistically different from that on non-fake news ( χ 2 = 0.00, p <0.99).

Next, we examine whether firms with an uncertain information environment (i.e., higher return volatility) and a less sophisticated investor base (i.e., high retail ownership) are more prone to foreign-originating fake news. Information frictions may slow the price-discovery process and can cause prices to deviate from intrinsic values for prolonged periods [ 56 , 57 ], which increases the influence of rumormongers on stock prices. Theoretical literature suggests that rumors create profit opportunities for rumormongers [ 58 ], and recent work shows that corporate fake news can affect stock prices [ 11 , 59 , 60 ]. Consistent with our conjecture, we find the incidence of fake news to be higher for firms that have a less robust information environment. The likelihood of being targeted by a foreign source is higher for firms with lower institutional ownership (-0.010 versus -0.006, χ 2 = 5.63, p <0.02) and higher return volatility (0.027 versus 0.010, χ 2 = 3.07, p <0.08).

In columns 4 and 5, we examine whether fake news from foreign accounts is concentrated in strategically important firms. To test this, we use two variables. First, we define Industry Leader as an indicator equal to one if a firm is the largest member of its industry in terms of revenue, and zero otherwise. Second, we define Strategic Industry as an indicator equal to one if a firm is in the computer, telecommunication, pharmaceutical, semiconductor, or defense industry, and zero otherwise. We predict and find that firms that are leaders in their industries, as well as firms operating in strategic industries, are more likely be targeted by fake news originating from foreign countries. Being a member of a strategic industry increases the proportion of fake news from foreign countries by 2.9% (0.004/0.137). The coefficient estimate on Strategic Industry for target firms (0.004) is statistically different from the coefficient estimate for non-target firms (0.002) with χ 2 = 2.63 and p <0.10. Industry leaders are also more likely to be targeted by foreign accounts (0.021 versus 0.016). However, the difference of coefficient estimates is not statistically significant between firms subject to fake and non-fake news ( χ 2 = 0,58 and p <0.45).

Overall, the firm-level analysis complements our country-level findings by showing that strategic industries have a higher probability of being targeted by fake news originating from foreign countries. That said, as discussed in the Introduction, we cannot cleanly attribute our findings to a foreign state actor or an intentional disinformation operation. Other potential reasons for disseminating negative fake news include the financial incentives of investors holding short positions or participating in other schemes that would benefit from a negative market reaction. In addition, inter-firm competition incentives may induce competitors to spread false news about their rivals. Finally, social media algorithms may unintentionally amplify certain content, including misinformation, based on factors like user engagement or click-through rates. While these are all plausible possibilities, they do not line up well with our country- and firm-level findings (i.e., the positive correlation between foreign-originated fake news and heightened geopolitical risks, as well as the targeting of strategic industries). We acknowledge, however, that we cannot fully rule out these alternative explanations, given the aforementioned limitations and the lack of granular data.

3.5 Additional analysis

In this section, we conduct three additional tests. First, we exclude political corporate fake news from our sample as politics is the most common topic found in corporate fake news (37.7%). Table 6 , Panel A summarizes the results. Our analysis shows that the results hold in the sample of fake and non-fake corporate news (except that some differences in the coefficient estimates are statistically insignificant).

https://doi.org/10.1371/journal.pone.0301364.t006

Second, in the analysis above, we use Foreign Fake News (%) as a continuous measure to capture the prevalence of foreign-originated fake news. Alternatively, we create an indicator variable Foreign Fake News (dummy) that is equal to one if a firm is targeted by foreign-originated fake news, and zero otherwise. Unlike the continuous measure, Foreign Fake News (dummy) takes a value of one if at least one initiator of fake news is a foreign account. Panel B summarizes the results. The findings are robust to the alternative use of this indicator variable.

Third, our inferences are robust to alternative clustering of standard errors, such as two-way clustering at the firm and year level (untabulated).

4. Discussion

This paper exploits a machine-learning approach to infer the geographical distribution of fake news spreaders on Twitter. We find that corporate fake news is more likely to originate from foreign countries, is more pronounced during periods of high geopolitical tension, and is more likely to target strategic industries and firms operating in uncertain information environments.

Our findings have both policy and practical implications. First, they will be of interest to policymakers, as they provide initial evidence of foreign-originating misinformation in capital markets. While we cannot attribute misinformation to a foreign state actor (or an intentional disinformation operation), we provide preliminary evidence that foreign influence operations in the political sphere can carry over to the economic domain. The findings may encourage policymakers to establish fact-checking organizations dedicated to financial information. Or, adopting a proactive approach, they can implement (or develop) advanced AI tools to detect and track misinformation campaigns in real time to combat online falsehood [ 61 ]. Educating investors about identifying red flags for misleading information can also help investors make informed decisions.

Second, our findings show the importance of a holistic approach to information risk. In today’s world, companies can be political targets. In this new environment, executives must not focus narrowly on financial or accounting information but must also pay attention to broader information risks (e.g., disinformation campaigns on social media). Geopolitical risks can further incentivize foreign actors to seed fake news about U.S. firms. Our work suggests that executives should plan for disinformation campaigns, be ready to respond to incidents online, and plan for these events in the new geopolitical and information environment. To do this, companies may leverage new technologies to monitor, detect, and respond to misinformation campaigns, e.g., by building internal fact-checking teams, partnering with independent fact-checking organizations, or using generative AI tools. They can actively engage in ‘social media listening’ to monitor real-time online conversations about the firm and gather insights about their brands, industry, or products. In doing so, companies can communicate more accurate information (and debunk false claims) during periods of high misinformation risk.

Our findings provide initial suggestive evidence on the potential involvement of foreign actors in the economic domain. Future work can explore the benefit the rumormongers derive from spreading negative fake news and the ultimate impact of negative fake news on stock market behavior. It would be also interesting to examine the optimal response of firms to the fake news. Finally, investigating the cross-platform spread of misinformation and the use of new technologies could provide a broader perspective. Recent advances in technology (e.g., generative AI) may allow for the creation of highly realistic but entirely fabricated video or audio content. These deepfakes can be used to spread misinformation without relying on traditional text-based methods. We leave these and other considerations for future research.

Supporting information

S1 file. supporting information files containing s1-s5 appendices..

https://doi.org/10.1371/journal.pone.0301364.s001

Acknowledgments

We thank Philip Lee Hann, Daiyue Li, Yiming Chen, Yuqing Zhao, Li Yu Kua, Ong Li Jing, Wang Jing, and Gautam Rohan for excellent research assistance. We also thank conference participants at NBS Research Day, Chang Xin (Simba), and Byoung-Hyoun Hwang for helpful comments and suggestions.

View Article
PubMed/NCBI
Google Scholar
60. Xu R. Corporate fake news on social media. Doctoral Dissertation, University of Miami. 2021. Available from: https://scholarship.miami.edu/esploro/outputs/doctoral/Corporate-Fake-News-on-Social-Media/991031553788502976 .

IMAGES

foreign literature.docx
foreign literature
SOLUTION: Chapter 2 review of related literature and studies docx
Foreign Studies and Foreign Literature About Stress and Sources of
(DOC) Foreign Literature
🐈 Example of foreign literature. Related Local and Foreign Literature

VIDEO

The Literature Review
STAGES OF BEING
Studying English Literature in Context: Study Guide & Critical Reflections
Spotlight on the Comparative Literature BA
Writing a Research Proposal for PhD in English Literature / Arts and Humanities
Wednesday Sem |Issues and Factors Shaping Chinese Research on Indian Literature| 27 Mar'24 @ 3PM IST

COMMENTS

Foreign Language Literatures
Tips for doing research on literary topics. Start with the MLA International Bibliography. The MLA International Bibliography is both international and multilingual, making it a great general tool for research in literary scholarship. You can use the drop-down list to specify a Subject Literature by nation or region (Scottish, North African, etc.).
Full article: Perspectives on translation and world literature
In 1993, he lectured again on comparative literature at Shanghai and Beijing (Wang, Citation 2001, p. 38). This was a time of intense intellectual exchange between China and the West, a period that showed a great interest in foreign literature and culture, and also aimed to disseminate Chinese cultural artifacts abroad.
Language and Literature: Sage Journals
Language and Literature is an invaluable international peer-reviewed journal that covers the latest research in stylistics, defined as the study of style in literary and non-literary language. We publish theoretical, empirical and experimental research that aims to make a contribution to our understanding of style and its effects on readers.
Foreign Languages & Literatures Research Guide: Articles
Foreign Languages & Literature Research Guide. This guide is designed to help get you started with your research. Many of the resources are paid for by Rasmuson Library. Best Bets for Finding Articles. MLA International Bibliography. Covers literature, language and linguistics, folklore, film, literary theory and criticism, dramatic arts, as ...
Foreign Literatures in America
Foreign Literatures in America (FLA) is a project devoted to the recovery and understanding of the significance of foreign authored literary works, as well as immigrant authored literary works, in the U.S. throughout U.S. history. ... Our principal mission is to challenge conceptions of "American literature" that turn upon the American ...
The language and non-language benefits of literature in foreign
Set against the backcloth of increasing recognition and attention to literature in foreign language education (FLE) globally, this article reports part of a large-scale study of 1,190 secondary-level learners' views of the benefits of literature, as instantiated by short stories (ShS) and poems and songs (PS) in their English-as-a-foreign-language learning.
Research Guides: Foreign Language Literatures: Finding Articles
Literature, Film, Couse Assignments in Languages. The Literary Encyclopedia publishes biographies of major and minor writers; scholarly descriptions of all interesting texts written by these authors, including those often neglected; and a variety of descriptive and critical essays on literary, cultural and historical matters, which provide a finer understanding of the social contexts in which ...
Literature: A Research Guide for Graduate Students
Welcome! This guide is maintained by Odile Harter, the library liaison to the departments of Comparative Literature and English.Here you'll find: Get Started - a to-do list for new graduate students. Find a Database - how to locate the best search engines for your field or project. Research Dos & Don'ts - tips to help you be productive and efficient. All of which supplement Literary Research ...
Research Guides: Foreign Language Literatures: Home
This guide is meant to assist you in searching for foreign language literatures. It has information on how to search our library for foreign language titles, as well as selected resources for various language literatures. Use the green navigation tabs for more specific resources: Article databases. Book Search and In-library locations
Key principles for an integrated intercultural literary pedagogy: An
Despite an abundance of theoretical research on the benefits of using literature in the foreign language classroom for intercultural development, empirical studies on effective pedagogical approaches are scarce. ... (Byram et al., 2017), research in a CLIL classroom where literature is the topic aiming at intercultural development has been ...
(PDF) Textbooks on Foreign Literature as a Means of Students
This article describes an action research experience carried out in an advanced English as a foreign language classroom of the language program at a university in Bogotá, Colombia, in 2010.
Libraries: Foreign Language and Literature: Research Methods
Subjects: Foreign Language and Literature 344 J L Seehorn Jr Rd Johnson City, TN 37614-1701
Foreign Literature
Foreign Literature, founded in 1980, has the longest history and the biggest influence among professional academic publications in the foreign literature research in China.Editors of the journal are researchers from the Institute of Foreign Literature, Beijing Foreign Studies University, and the editorial board is comprised of renowned scholars from home and abroad.
Research Methods
For example, we may be studying the relationship between literature and society, between author and text, or the status of a work in the literary canon. We may want to know about a work's form, genre, or thematics. We may want to know about the audience's reading and reception, or about methods for teaching literature in schools ...
PDF Foreign Literature at the Lessons of Individual Reading ...
organized being based on foreign literature. Literary read is a reflection of the history, traditions, mentality and identity of one or another nation. Reading foreign literature enriches students' vocabulary, expands knowledge of phraseological units and set expressions. The language of fiction has a special
PDF A Literature Review of Foreign Studies on the Impact of CALL on ...
A Literature Review of Foreign Studies on the Impact of CALL on Second Language Acquisition from 2015 Qiuxin Zhang1 1Lanzhou Jiaotong University, School of Foreign Language, Anning West Road, Anning District 88, China ... and language teaching is no exception. As a research hot issue, computer-assisted language learning (CALL) has attracted ...
How to Write a Literature Review
Literature review research question example What is the impact of social media on body image among Generation Z? Make a list of keywords. Start by creating a list of keywords related to your research question. Include each of the key concepts or variables you're interested in, and list any synonyms and related terms. You can add to this list ...
Literature: A Research Guide for Graduate Students
A guide to help get you started on your graduate work in English, Comparative Literature, and related fields. NOTE: This guide is a supplement to the general topic guide Literary Research in Harvard Libraries.
CHAPTER 2 Review of Related Literature and Studies Foreign Literature
RELATED STUDIES FOREIGN Physical Breus (2006) More and more research studies demonstrate that daytime sleepiness from chronic sleep deprivation and poor quality sleep has significant impacts on daytime behavior and academic performance, as well as concentration, attention, and mood.
Foreign aid and poverty reduction: A review of international literature
Presently, the empirical literature on aid effectiveness is dominated by studies on the effectiveness of foreign aid on economic growth. There is a general dearth of the empirical literature on the effectiveness of foreign aid on poverty reduction. Thus, the poverty-reducing effects of aid are not well documented (White, Citation 2015, p. 187).
(PDF) Local and foreign literary genres on students' reading
Local and foreign literary genres on students' reading comprehension and attitude is a study investigating whether local or foreign genres have significant effect on grade 8 students' reading ...
Frontiers
Keywords: foreign language anxiety, foreign language learning, English as a foreign language, foreign language learners, literature review. Citation: Yu Q (2024) Foreign language anxiety research in System between 2004 and 2023: looking back and looking forward. Front. Psychol. 15:1373290. doi: 10.3389/fpsyg.2024.1373290
Chapter 2 Local and Foreign Literature
Foreign Literature : India's higher education system is the third largest in the world, after China andUnited State. The main governing body at tertiary level is the University GrantsCommission. Which enforces its standards, advises the government, and help coordinate between the centre and the state? Accreditation for higher learning is ...
The geography of corporate fake news
Although a rich academic literature examines the use of fake news by foreign actors for political manipulation, there is limited research on potential foreign intervention in capital markets. To address this gap, we construct a comprehensive database of (negative) fake news regarding U.S. firms by scraping prominent fact-checking sites. We identify the accounts that spread the news on Twitter ...
Small and medium-sized enterprises in emerging markets and foreign
The rest of this study is organized as follows. First, after the introduction section, the literature review of the various CSFs is depicted in Section II. Section III reveals research methods, where MCDM, the step-wise weight assessment ratio analysis, the sampling procedure, and the Delphi method are presented in detail.
An examination into pre-Service English-as-a-Foreign-Language teachers
A literature review indicates that no research has been conducted to date to examine pre-service English as a foreign language ... Her research concentrations include foreign/second language learning, teaching language skills to young and adult learners and foreign language teacher education.
Models of private-public partnership in medical activities (according
DOI: 10.32687/1561-5936-2024-28-1-47-52 Corpus ID: 268959934; Models of private-public partnership in medical activities (according to foreign and domestic literature) @article{Sarkisjan2024ModelsOP, title={Models of private-public partnership in medical activities (according to foreign and domestic literature)}, author={A. D. Sarkisjan and T. V. Shapovalenko and S. P. Darenkov and Z. H ...