Frames (Automated Content Analysis)

Validation Of Results ◽

Frames describe the way issues are presented, i.e., what aspects are made salient when communicating about these issues. Field of application/theoretical foundation: The concept of frames is directly based on the theory of “Framing”. However, many studies using automated content analysis are lacking a clear theoretical definition of what constitutes a frame. As an exception, Walter and Ophir (2019) use automated content analysis to explore issue and strategy frames as defined by Cappella and Jamieson (1997). Vu and Lynn (2020) refer to Entman’s (1991) understanding of frames. The datasets referred to in the table are described in the following paragraph: Van der Meer et al. (2010) use a dataset consisting of Dutch newspaper articles (1991-2015, N = 9,443) and LDA topic modeling in combination with k-means clustering to identify frames. Walter and Ophir (2019) use three different datasets and a combination of topic modeling, network analysis and community detection algorithms to analyze frames. Their datasets consist of political newspaper articles and wire service coverage (N = 8,337), newspaper articles on foreign nations (2010-2015, N = 18,216) and health-related newspaper coverage (2009-2016, N = 5,005). Lastly, Vu and Lynn (2020) analyze newspaper coverage of the Rohingya crisis (2017-2018, N = 747) concerning frames. References/combination with other methods of data collection: While most approaches only rely on automated data collection and analyses, some also combine automated and manual coding. For example, a recent study by Vu and Lynn (2020) proposes to combine semantic networks and manual coding to identify frames. Table 1. Measurement of “Frames” using automated content analysis. Author(s) Sample Procedure Formal validity check with manual coding as benchmark* Code Vu & Lynn (2020) Newspaper articles Semantic networks; manual coding Reported Not available van der Meer et al. (2019) Newspaper articles LDA topic modeling; k-means clustering Not reported Not available Walter & Ophir (2019) (a) U.S. newspapers and wire service articles (b) Newspaper articles (c) Newspaper articles LDA topic modeling, network analysis; community detection algorithms Not reported https://github.com/DrorWalt/ANTMN *Please note that many of the sources listed here are tutorials on how to conducted automated analyses – and therefore not focused on the validation of results. Readers should simply read this column as an indication in terms of which sources they can refer to if they are interested in the validation of results. References Cappella, J. N., & Jamieson, K. H. (1997). Spiral of cynicism: The press and the public good. New York: Oxford University Press. Entman, R. M. 1991. Framing U.S. coverage of international news: contrasts in narratives of the KAL and Iran Air incidents. Journal of Communication, 41(4), 6-7. van der Meer, T. G. L. A., Kroon, A. C., Verhoeven, P., & Jonkman, J. (2019). Mediatization and the disproportionate attention to negative news: The case of airplane crashes. Journalism Studies, 20(6), 783–803. Walter, D., & Ophir, Y. (2019). News frame analysis: an inductive mixed-method computational approach. Communication Methods and Measures, 13(4), 248–266. Vu, H. T., & Lynn, N. (2020). When the news takes sides: Automated framing analysis of news coverage of the rohingya crisis by the elite press from three countries. Journalism Studies. Online first publication. doi:10.1080/1461670X.2020.1745665

Actors (Automated Content Analysis)

DOCA - Database of Variables for Content Analysis ◽

10.34778/1b ◽

2021 ◽

Author(s):

Valerie Hase

Keyword(s):

Content Analysis ◽

Named Entities ◽

Part Of Speech Tagging ◽

Part Of Speech ◽

State Of The Union ◽

German Newspaper ◽

Speech Tagging ◽

Validation Of Results ◽

Actors in coverage might be individuals, groups, or organizations, which are discussed, described, or quoted in the news. The datasets referred to in the table are described in the following paragraph: Benoit and Matuso (2020) uses fictional sentences (N = 5) to demonstrate how named entities and noun phrases can be identified automatically. Lind and Meltzer (2020) demonstrate the use of organic dictionaries to identify actors in German newspaper articles (2013-2017, N = 348,785). Puschmann (2019) uses four data sets to demonstrate how sentiment/tone may be analyzed by the computer. Using tweets (2016, N = 18,826), German newspaper articles (2011-2016, N = 377), Swiss newspaper articles (2007-2012, N = 21,280), and debate transcripts (1970-2017, N = 7,897), he extracts nouns and named entities from text. Lastly, Wiedemann and Niekler (2017) extract proper nouns from State of the Union speeches (1790-2017, N = 233). Field of application/theoretical foundation: Related to theories of “Agenda Setting” and “Framing”, analyses might want to know how much weight is given to a specific actor, how these actors are evaluated and what perspectives and frames they might bring into the discussion how prominently. References/combination with other methods of data collection: Oftentimes, studies use both manual and automated content analysis to identify actors in text. This might be a useful tool to extend the lists of actors that can be found as well as to validate automated analyses. For example, Lind and Meltzer (2020) combine manual coding and dictionaries to identify the salience of women in the news. Table 1. Measurement of “Actors” using automated content analysis. Author(s) Sample Procedure Formal validity check with manual coding as benchmark* Code Benoit & Matuso (2020) Fictional sentences Part-of-Speech tagging; syntactic parsing Not reported https://cran.r-project.org/web/packages/spacyr/vignettes/using_spacyr.html Lind & Meltzer (2020) Newspapers Dictionary approach Reported https://osf.io/yqbcj/?view_only=369e2004172b43bb91a39b536970e50b Puschmann (2019) (a) Tweets (b) German newspaper articles (c) Swiss newspaper articles (d) United Nations General Debate Transcripts Part-of-Speech tagging; syntactic parsing Not reported http://inhaltsanalyse-mit-r.de/ner.html Wiedemann & Niekler (2017) State of the Union speeches Part-of-Speech tagging Not reported https://tm4ss.github.io/docs/Tutorial_8_NER_POS.html *Please note that many of the sources listed here are tutorials on how to conducted automated analyses – and therefore not focused on the validation of results. Readers should simply read this column as an indication in terms of which sources they can refer to if they are interested in the validation of results. References Benoit, K., & Matuso. (2020). A Guide to Using spacyr. Retrieved from https://cran.r-project.org/web/packages/spacyr/vignettes/using_spacyr.html Lind, F., & Meltzer, C. E. (2020). Now you see me, now you don’t: Applying automated content analysis to track migrant women’s salience in German news. Feminist Media Studies, 1–18. Puschmann, C. (2019). Automatisierte Inhaltsanalyse mit R. Retrieved from http://inhaltsanalyse-mit-r.de/index.html Wiedemann, G., Niekler, A. (2017). Hands-on: a five day text mining course for humanists and social scientists in R. Proceedings of the 1st Workshop Teaching NLP for Digital Humanities (Teach4DH@GSCL 2017), Berlin. Retrieved from https://tm4ss.github.io/docs/index.html

Content analysis of newspaper coverage of wolf recolonization in France using structural topic modeling

Biological Conservation ◽

10.1016/j.biocon.2018.01.029 ◽

2018 ◽

Vol 220 ◽

pp. 254-261 ◽

Cited By ~ 24

Author(s):

Marie Chandelier ◽

Agnès Steuckardt ◽

Raphaël Mathevet ◽

Sascha Diwersy ◽

Olivier Gimenez

Keyword(s):

Content Analysis ◽

Topic Modeling ◽

Newspaper Coverage ◽

Structural Topic Modeling

Clouded reality: News representations of culturally close and distant ethnic outgroups

Communications ◽

10.1515/commun-2019-2069 ◽

2020 ◽

Vol 45 (s1) ◽

pp. 744-764 ◽

Cited By ~ 1

Author(s):

Anne C. Kroon ◽

Damian Trilling ◽

Toni G. L. A. van der Meer ◽

Jeroen G. F. Jonkman

Keyword(s):

Content Analysis ◽

Cultural Distance ◽

Word Embeddings ◽

Ethnic Bias ◽

AbstractThe current study explores how the cultural distance of ethnic outgroups relative to the ethnic ingroup is related to stereotypical news representations. It does so by drawing on a sample of more than three million Dutch newspaper articles and uses advanced methods of automated content analysis, namely word embeddings. The results show that distant ethnic outgroup members (i. e., Moroccans) are associated with negative characteristics and issues, while this is not the case for close ethnic outgroup members (i. e., Belgians). The current study demonstrates the usefulness of word embeddings as a tool to study subtle aspects of ethnic bias in mass-mediated content.

Parsing, Semantic Networks, and Political Authority Using Syntactic Analysis to Extract Semantic Relations from Dutch Newspaper Articles

Political Analysis ◽

10.1093/pan/mpn006 ◽

2008 ◽

Vol 16 (4) ◽

pp. 428-446 ◽

Cited By ~ 46

Author(s):

Wouter van Atteveldt ◽

Jan Kleinnijenhuis ◽

Nel Ruigrok

Keyword(s):

Network Analysis ◽

Political Communication ◽

Semantic Network ◽

Semantic Networks ◽

Syntactic Analysis ◽

Semantic Network Analysis ◽

Political Actors ◽

Name Matching ◽

Automatic Coding ◽

Analysis of political communication is an important aspect of political research. Thematic content analysis has yielded considerable success both with manual and automatic coding, but Semantic Network Analysis has proven more difficult, both for humans and for the computer. This article presents a system for an automated Semantic Network Analysis of Dutch texts. The system automatically extracts relations between political actors based on the output of syntactic analysis of Dutch newspaper articles. Specifically, the system uses pattern matching to find source constructions and determine the semantic agent and patient of relations, and name matching and anaphora resolution to identify political actors. The performance of the system is judged by comparing the extracted relations to manual codings of the same material. Results on the level of measurement indicate acceptable performance. We also estimate performance at the levels of analysis by using a case study of media authority, resulting in good correlations between the theoretical variables derived from the automatic and manual analysis. Finally, we test a number of substantive hypotheses with regression models using the automatic and manual output, resulting in highly similar models in each case. This suggests that our method has sufficient performance to be used to answer relevant political questions in a valid way.

Spotlight on Athletes with a Disability: Malaysian Newspaper Coverage of the 2012 London Paralympic Games

Adapted Physical Activity Quarterly ◽

10.1123/apaq.2015-0021 ◽

2016 ◽

Vol 33 (1) ◽

pp. 15-32 ◽

Cited By ~ 6

Author(s):

Jadeera Phaik Geok Cheong ◽

Selina Khoo ◽

Rizal Razman

Keyword(s):

Content Analysis ◽

English Language ◽

Female Athletes ◽

Newspaper Coverage ◽

Paralympic Games ◽

This study analyzed newspaper coverage of the 2012 London Paralympic Games by 8 Malaysian newspapers. Articles and photographs from 4 English-language and 4 Malay-language newspapers were examined from August 28 (1 day before the Games) to September 10, 2012 (1 day after the Games closing). Tables, graphs, letters, fact boxes, and lists of events were excluded from analysis. A total of 132 articles and 131 photographs were analyzed. Content analysis of the newspaper articles revealed that most (62.8%) of the articles contained positive reference to the athletes with a disability. There were equal numbers (39.1%) of action and static shots of athletes. More articles and photographs of Malaysian (58%) than non-Malaysian (42%) athletes with a disability were identified. Only 14.9% of the articles and photographs were related to female athletes with a disability.

Krantenberichtgeving over het rookverbod in de horeca

Tijdschrift voor Communicatiewetenschappen ◽

10.5117/2011.039.003.004 ◽

2011 ◽

Vol 39 (3) ◽

Author(s):

Gera E. Nagelhout ◽

Bas van den Putte ◽

Hein de Vries ◽

Marc C. Willemsen

Keyword(s):

Content Analysis ◽

Passive Smoking ◽

Hospitality Industry ◽

Newspaper Coverage ◽

Smoking Ban ◽

Economic Aspects ◽

Hospitality Workers ◽

Health Damage ◽

The Media ◽

Newspaper coverage about the smoking ban in the hospitality industry: A content analysis. Newspaper coverage about the smoking ban in the hospitality industry: A content analysis. In the Netherlands relatively few people support the hospitality industry smoking ban. Possibly this is due to the way the media covered the smoking ban. A content analysis of 1,041 articles in six Dutch newspapers showed that when there were economic aspects of the ban in the newspaper articles, the articles were mostly negative towards the smoking ban (62% negative, 29% positive, 9% mixed or neutral). The same was true when the newspaper articles dealt with resistance against the ban (69% negative, 26% positive, 5% mixed). When there were health aspects in the articles, the articles were equally often positive as negative (42% positive, 43% negative, 15% mixed or neutral). Although the smoking ban was implemented to protect hospitality workers from the health damage of passive smoking, economic aspects (59%) and the resistance against the ban (46%) appeared more often in the newspapers than health aspects (22%).

The manifest and latent functions of Twitter use by journalists: An observational study among economic journalists

Journal of Applied Journalism & Media Studies ◽

10.1386/ajms_00004_1 ◽

2019 ◽

Vol 8 (3) ◽

pp. 311-329

Author(s):

Michiel Johnson ◽

Steve Paulussen ◽

Peter Van Aelst

Keyword(s):

Content Analysis ◽

Social Network ◽

Social Network Analysis ◽

Network Analysis ◽

Observational Study ◽

Dominant Position ◽

The Social ◽

Potential Sources ◽

The Media

This study focuses on Twitter use among economic journalists working for print media in Belgium. By looking into their tweeting and following behaviour, the article examines how economic journalists use Twitter for promotional, conversational and sourcing purposes. Based on an automated content analysis of what they tweet and a social network analysis of whom they follow, the results show that economic journalists mainly use Twitter to promote themselves and their news organization rather than to engage in public conversation on the platform. In addition, the study looks into their following behaviour to investigate which actors they consider as 'potential sources'. Here, the findings are consistent with previous studies among political and health journalists, indicating that journalists are more likely to follow institutionally affiliated rather than non-affiliated sources on Twitter. Furthermore, the social network analysis gives additional evidence of the media-centered of journalists' Twitter use, as media-affiliated actors maintain a dominant position in the economic journalists' Twitter networks.

Sentiment/tone (Automated Content Analysis)

DOCA - Database of Variables for Content Analysis ◽

10.34778/1d ◽

2021 ◽

Author(s):

Valerie Hase

Keyword(s):

Machine Learning ◽

Content Analysis ◽

Sentiment Analysis ◽

Supervised Machine Learning ◽

Data Sets ◽

Financial News ◽

State Of The Union ◽

Course Material ◽

Sentiment/tone describes the way issues or specific actors are described in coverage. Many analyses differentiate between negative, neutral/balanced or positive sentiment/tone as broader categories, but analyses might also measure expressions of incivility, fear, or happiness, for example, as more granular types of sentiment/tone. Analyses can detect sentiment/tone in full texts (e.g., general sentiment in financial news) or concerning specific issues (e.g., specific sentiment towards the stock market in financial news or a specific actor). The datasets referred to in the table are described in the following paragraph: Puschmann (2019) uses four data sets to demonstrate how sentiment/tone may be analyzed by the computer. Using Sherlock Holmes stories (18th century, N = 12), tweets (2016, N = 18,826), Swiss newspaper articles (2007-2012, N = 21,280), and debate transcripts (2013-2017, N = 205,584), he illustrates how dictionaries may be applied for such a task. Rauh (2019) uses three data sets to validate his organic German language dictionary for sentiment/tone. His data consists of sentences from German parliament speeches (1991-2013, N = 1,500), German-language quasi-sentences from German, Austrian and Swiss party manifestos (1998-2013, N = 14,008) and newspaper, journal and news wire articles (2011-2012, N = 4,038). Silge and Robinson (2020) use six Jane Austen novels to demonstrate how dictionaries may be used for sentiment analysis. Van Atteveldt and Welbers (2020) use state of the Union speeches (1789-2017, N = 58) for the same purpose. The same authors (van Atteveldt & Welbers, 2019) show based on a dataset of N = 2,000 movie reviews how supervised machine learning might also do the trick. In their Quanteda tutorials, Watanabe and Müller (2019) demonstrate the use of dictionaries and supervised machine learning for sentiment analysis on UK newspaper articles (2012-2016, N = 6,000) as well as the same set of movie reviews (n = 2,000). Lastly, Wiedemann and Niekler (2017) use state of the Union speeches (1790-2017, N = 233) to demonstrate how sentiment/tone can be coded automatically via a dictionary approach. Field of application/theoretical foundation: Related to theories of “Framing” and “Bias” in coverage, many analyses are concerned with the way the news evaluates and interprets specific issues and actors. References/combination with other methods of data collection: Manual coding is needed for many automated analyses, including the ones concerned with sentiment. Studies for example use manual content analysis to develop dictionaries, to create training sets on which algorithms used for automated classification are trained, or to validate the results of automated analyses (Song et al., 2020). Table 1. Measurement of “Sentiment/Tone” using automated content analysis. Author(s) Sample Procedure Formal validity check with manual coding as benchmark* Code Puschmann (2019) (a) Sherlock Holmes stories (b) Tweets (c) Swiss newspaper articles (d) German Parliament transcripts Dictionary approach Not reported http://inhaltsanalyse-mit-r.de/sentiment.html Rauh (2018) (a) Bundestag speeches (b) Quasi-sentences from German, Austrian and Swiss party manifestos (c) Newspapers, journals, agency reports Dictionary approach Reported https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/BKBXWD Silge & Robinson (2020) Books by Jane Austen Dictionary approach Not reported https://www.tidytextmining.com/sentiment.html van Atteveldt & Welbers (2020) State of the Union speeches Dictionary approach Reported https://github.com/ccs-amsterdam/r-course-material/blob/master/tutorials/sentiment_analysis.md van Atteveldt & Welbers (2019) Movie reviews Supervised Machine Learning Approach Reported https://github.com/ccs-amsterdam/r-course-material/blob/master/tutorials/r_text_ml.md Watanabe & Müller (2019) Newspaper articles Dictionary approach Not reported https://tutorials.quanteda.io/advanced-operations/targeted-dictionary-analysis/ Watanabe & Müller (2019) Movie reviews Supervised Machine Learning Approach Reported https://tutorials.quanteda.io/machine-learning/nb/ Wiedemann & Niekler (2017) State of the Union speeches Dictionary approach Not reported https://tm4ss.github.io/docs/Tutorial_3_Frequency.html *Please note that many of the sources listed here are tutorials on how to conducted automated analyses – and therefore not focused on the validation of results. Readers should simply read this column as an indication in terms of which sources they can refer to if they are interested in the validation of results. References Puschmann, C. (2019). Automatisierte Inhaltsanalyse mit R. Retrieved from http://inhaltsanalyse-mit-r.de/index.html Rauh, C. (2018). Validating a sentiment dictionary for German political language—A workbench note. Journal of Information Technology & Politics, 15(4), 319–343. doi:10.1080/19331681.2018.1485608 Silge, J., & Robinson, D. (2020). Text mining with R. A tidy approach. Retrieved from https://www.tidytextmining.com/ Song, H., Tolochko, P., Eberl, J.-M., Eisele, O., Greussing, E., Heidenreich, T., Lind, F., Galyga, S., & Boomgaarden, H.G. (2020) In validations we trust? The impact of imperfect human annotations as a gold standard on the quality of validation of automated content analysis. Political Communication, 37(4), 550-572. van Atteveldt, W., & Welbers, K. (2019). Supervised Text Classification. Retrieved from https://github.com/ccs-amsterdam/r-course-material/blob/master/tutorials/r_text_ml.md van Atteveldt, W., & Welbers, K. (2020). Supervised Sentiment Analysis in R. Retrieved from https://github.com/ccs-amsterdam/r-course-material/blob/master/tutorials/sentiment_analysis.md Watanabe, K., & Müller, S. (2019). Quanteda tutorials. Retrieved from https://tutorials.quanteda.io/ Wiedemann, G., Niekler, A. (2017). Hands-on: a five day text mining course for humanists and social scientists in R. Proceedings of the 1st Workshop Teaching NLP for Digital Humanities (Teach4DH@GSCL 2017), Berlin. Retrieved from https://tm4ss.github.io/docs/index.html

Obesity Communication with Etiology and Disease: Automated Content Analysis of Digital News, 2010-2019 (Preprint)

10.2196/preprints.26660 ◽

2020 ◽

Author(s):

Angela Chang ◽

Matthew Tingchi Liu ◽

Wen Jia

Keyword(s):

Content Analysis ◽

Mainland China ◽

Personal Attributes ◽

Analysis Tool ◽

Obesity Epidemic ◽

News Stories ◽

Journalism Studies ◽

Digital News ◽

Online Texts

BACKGROUND The fact that the number of population suffering from obesity has increased worldwide calls into question on media efforts for informing the public. This research attempts to determine the ways in which the mainstream digital news covers the etiology of obesity and diseases associated with the burden of obesity. OBJECTIVE The dual objectives of this study are to obtain an understanding of what the news says about obesity and to explore meanings in data by extending preconceived grounded theory. METHODS We propose an automatic content analysis tool, DiVoMiner. This computer-aided platform is designed to organize and filter large sets of data based on patterns of word occurrence and latent topics. Another programming language Python 3 is employed to explore connections and patterns created by the aggregated interactions. The 10 years of news text compared the development of obesity coverage and its potential impact on perception in Mainland China, Hong Kong, and Taiwan. Digital news stories covering obesity along with affliction and consequence inferences in nine newspapers were sampled. RESULTS A total of 30,968 news stories were identified with increasing attention since 2016. The highest intensity of newspaper coverage of obesity communication was found in Taiwan. Overall, a stronger focus on two shared causative attributes of obesity are on stress (n = 4,483, 33.0%) and tobacco use (n = 3143, 23.2.0%). Similar to the previous studies, the discourse between the obesity epidemic and personal afflictions is the most emphasized approach (n = 13,587, 80.0%). Additionally, the burden of obesity and cardiovascular diseases are implied the most (n = 8,477, 42.2%) despite the aggregated interaction of edge centrality shows the highest link between the use of “obesity” and “cancer”. The discussion indicated that the inclination of blaming personal attributes for health afflictions potentially limits social and governmental responsibility for addressing this issue. The strategy of various obesity communication for news gatekeepers, health communication researchers, and policy-makers are noted. CONCLUSIONS This study goes beyond traditional journalism studies by extending the framework of computational and customizable online texts. This could set a norm for researchers and practitioners who work on the data projects largely for a different and innovative attempt. However, challenges of methods should be faced, including the lack of standards of automated content measures. CLINICALTRIAL not available.