Pub Date : 2024-06-12DOI: 10.1177/08944393241258763
Shelley Boulianne, A. O. Larsson
Comparative communication research needs to catch up to other disciplines. In this special issue and the associated International Communication Association preconference, we focus on comparative work related to digital political communication. This introduction argues that comparative digital political communication needs to consider comparisons across various dimensions, including countries, platforms, and time, whereas existing comparative communication research focuses on country or territorial comparison. We highlight the six submissions’ approaches to comparative work. Each submission provides at least one of these three dimensions of contrast. We conclude with a discussion of enduring gaps in this field of research, such as the lack of studies using time as a dimension of comparison. Time is crucial for understanding ever-changing digital media platforms. We also conclude by discussing some ongoing challenges in political communication research.
{"title":"Comparative Digital Political Communication: Comparisons Across Countries, Platforms, and Time","authors":"Shelley Boulianne, A. O. Larsson","doi":"10.1177/08944393241258763","DOIUrl":"https://doi.org/10.1177/08944393241258763","url":null,"abstract":"Comparative communication research needs to catch up to other disciplines. In this special issue and the associated International Communication Association preconference, we focus on comparative work related to digital political communication. This introduction argues that comparative digital political communication needs to consider comparisons across various dimensions, including countries, platforms, and time, whereas existing comparative communication research focuses on country or territorial comparison. We highlight the six submissions’ approaches to comparative work. Each submission provides at least one of these three dimensions of contrast. We conclude with a discussion of enduring gaps in this field of research, such as the lack of studies using time as a dimension of comparison. Time is crucial for understanding ever-changing digital media platforms. We also conclude by discussing some ongoing challenges in political communication research.","PeriodicalId":506768,"journal":{"name":"Social Science Computer Review","volume":"141 35","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141350775","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-11DOI: 10.1177/08944393241259434
Miklós Sebők, Ákos Máté, Orsolya Ring, Viktor Kovács, Richárd Lehoczki
The article presents an open-source and freely available natural language processing system for comparative policy studies. The CAP Babel Machine allows for the automated classification of input files based on the 21 major policy topics of the codebook of the Comparative Agendas Project (CAP). By using multilingual XLM-RoBERTa large language models, the pipeline can produce state-of-the-art level outputs for selected pairs of languages and domains (such as media or parliamentary speech). For 24 cases out of 41, the weighted macro F1 of our language-domain models surpassed 0.75 (and, for 6 language-domain pairs, 0.90). Besides macro F1, for most major topic categories, the distribution of micro F1 scores is also centered around 0.75. These results show that the CAP Babel machine is a viable alternative for human coding in terms of validity at less cost and higher reliability. The proposed research design also has significant possibilities for scaling in terms of leveraging new models, covering new languages, and adding new datasets for fine-tuning. Based on our tests on manifesto data, a different policy classification scheme, we argue that model-pipeline frameworks such as the Babel Machine can, over time, potentially replace double-blind human coding for a multitude of comparative classification problems.
文章介绍了一种用于比较政策研究的开源、免费的自然语言处理系统。CAP 巴别机可以根据比较议程项目（CAP）代码库中的 21 个主要政策主题对输入文件进行自动分类。通过使用多语言 XLM-RoBERTa 大语言模型，该管道可以为选定的语言对和领域（如媒体或议会发言）生成最先进水平的输出。在 41 个案例中有 24 个案例中，我们的语域模型的加权宏 F1 超过了 0.75（6 个语域对的加权宏 F1 超过了 0.90）。除了宏观 F1，对于大多数主要的主题类别，微观 F1 分数的分布也以 0.75 为中心。这些结果表明，就有效性而言，CAP 巴别机可替代人工编码，成本更低，可靠性更高。所提出的研究设计在利用新模型、覆盖新语言和添加新数据集进行微调方面也有很大的扩展空间。基于我们对宣言数据（一种不同的政策分类方案）的测试，我们认为，随着时间的推移，巴别机等模型管道框架有可能在众多比较分类问题上取代双盲人工编码。
{"title":"Leveraging Open Large Language Models for Multilingual Policy Topic Classification: The Babel Machine Approach","authors":"Miklós Sebők, Ákos Máté, Orsolya Ring, Viktor Kovács, Richárd Lehoczki","doi":"10.1177/08944393241259434","DOIUrl":"https://doi.org/10.1177/08944393241259434","url":null,"abstract":"The article presents an open-source and freely available natural language processing system for comparative policy studies. The CAP Babel Machine allows for the automated classification of input files based on the 21 major policy topics of the codebook of the Comparative Agendas Project (CAP). By using multilingual XLM-RoBERTa large language models, the pipeline can produce state-of-the-art level outputs for selected pairs of languages and domains (such as media or parliamentary speech). For 24 cases out of 41, the weighted macro F1 of our language-domain models surpassed 0.75 (and, for 6 language-domain pairs, 0.90). Besides macro F1, for most major topic categories, the distribution of micro F1 scores is also centered around 0.75. These results show that the CAP Babel machine is a viable alternative for human coding in terms of validity at less cost and higher reliability. The proposed research design also has significant possibilities for scaling in terms of leveraging new models, covering new languages, and adding new datasets for fine-tuning. Based on our tests on manifesto data, a different policy classification scheme, we argue that model-pipeline frameworks such as the Babel Machine can, over time, potentially replace double-blind human coding for a multitude of comparative classification problems.","PeriodicalId":506768,"journal":{"name":"Social Science Computer Review","volume":"56 23","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141358350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-11DOI: 10.1177/08944393241258767
Travis N. Ridout, Markus Neumann, Jielu Yao, Laura M. Baum, Michael M. Franz, P. Oleinikov, Erika Franklin Fowler
When it comes to the study of the messaging of online political campaigns, theory suggests that platform divergence should be common, but much research finds considerable convergence across platforms. In this research, we examine variation across digital and social media platforms in the types of paid campaign messages that are distributed, focusing on their goals, tone, and the partisanship of political rhetoric. We use data on the content of paid election advertisements placed on YouTube, Google search, Instagram, and Facebook during the 2020 elections in the United States, examining all federal candidates who advertised on these platforms during the final 2 months of the campaign. We find that YouTube is most distinct from the other platforms, perhaps because it most resembles television, but convergence better describes the two Meta platforms, Facebook and Instagram.
{"title":"Platform Convergence or Divergence? Comparing Political Ad Content Across Digital and Social Media Platforms","authors":"Travis N. Ridout, Markus Neumann, Jielu Yao, Laura M. Baum, Michael M. Franz, P. Oleinikov, Erika Franklin Fowler","doi":"10.1177/08944393241258767","DOIUrl":"https://doi.org/10.1177/08944393241258767","url":null,"abstract":"When it comes to the study of the messaging of online political campaigns, theory suggests that platform divergence should be common, but much research finds considerable convergence across platforms. In this research, we examine variation across digital and social media platforms in the types of paid campaign messages that are distributed, focusing on their goals, tone, and the partisanship of political rhetoric. We use data on the content of paid election advertisements placed on YouTube, Google search, Instagram, and Facebook during the 2020 elections in the United States, examining all federal candidates who advertised on these platforms during the final 2 months of the campaign. We find that YouTube is most distinct from the other platforms, perhaps because it most resembles television, but convergence better describes the two Meta platforms, Facebook and Instagram.","PeriodicalId":506768,"journal":{"name":"Social Science Computer Review","volume":"2 8","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141357835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Artificial intelligence (AI) is a collection of rapidly evolving disruptive technologies that radically alter various aspects of people, business, society, and the environment. AI increasingly provides significant advertising opportunities for society and business organizations. However, AI could be used to spread disinformation if it were deliberately programmed to produce misleading advertising content. Using cognitive appraisal theory and information quality theory to study how consumers assess threats and develop AI marketing coping strategies from the information generated by AI, this study examines the outcome of the dark side of AI advertising. We collected data from 451 AI-advertising users in Vietnam. The results based on PLS-SEM showed interesting and novelty results. The statistical analysis showed a negative correlation between contextual, representational, accessibility, and threat appraisals. There was also a statistically significant positive correlation between contextual, representational, accessibility, and coping appraisals. Threat appraisals were positively correlated with anger and anxiety but not loneliness. Coping appraisal was significant and negatively correlated with anxiety but not anger or loneliness. This study advances theory and management.
{"title":"The Dark Sides of AI Advertising: The Integration of Cognitive Appraisal Theory and Information Quality Theory","authors":"Luan-Thanh Nguyen, Tri-Quan Dang, Dang Thi Viet Duc","doi":"10.1177/08944393241258760","DOIUrl":"https://doi.org/10.1177/08944393241258760","url":null,"abstract":"Artificial intelligence (AI) is a collection of rapidly evolving disruptive technologies that radically alter various aspects of people, business, society, and the environment. AI increasingly provides significant advertising opportunities for society and business organizations. However, AI could be used to spread disinformation if it were deliberately programmed to produce misleading advertising content. Using cognitive appraisal theory and information quality theory to study how consumers assess threats and develop AI marketing coping strategies from the information generated by AI, this study examines the outcome of the dark side of AI advertising. We collected data from 451 AI-advertising users in Vietnam. The results based on PLS-SEM showed interesting and novelty results. The statistical analysis showed a negative correlation between contextual, representational, accessibility, and threat appraisals. There was also a statistically significant positive correlation between contextual, representational, accessibility, and coping appraisals. Threat appraisals were positively correlated with anger and anxiety but not loneliness. Coping appraisal was significant and negatively correlated with anxiety but not anger or loneliness. This study advances theory and management.","PeriodicalId":506768,"journal":{"name":"Social Science Computer Review","volume":" 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141371600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-17DOI: 10.1177/08944393241246101
Mike Farjam, Anamaria Dutceac Segesten
Scholarly literature has demonstrated that hybridity transforms both legacy and new media, but that this change is not even. We treat social media platforms as arenas of remediation, where users share and add their own context to information produced by both media subtypes and compare social media conversations about migration in six European languages that include links to either traditional or new media during 2015–2019. We use a mix of computational and statistical methods to analyze 3.5 million (re)tweets and 500,000 links shared within them. We identify the main differences in agenda setting power, function, and tone present within tweets that include links to legacy or new media. Our results show that discourses are similar across languages but clearly different when remediating legacy and new media. Trust in legacy media is correlated with higher proportion of shared links from legacy media and reversely related to the proportion of shared links from new media sources. Considering the volume and timing of the remediated content, we conclude that legacy media retains its agenda setting power. New media linked content tends to cover migration in association to subjects such as Islam or terrorism and to express strong critical opinions against migrants/refugees. The language used is more toxic than in legacy media linked content. The tweets remediating legacy media articles covered topics like domestic or European politics, causes of refugee arrivals and procedures to give them protection. Thus, legacy and new media remediated content differs in both tone and function: toxicity is low and factuality high for content linking to legacy media, with the reverse being true for new media remediations.
{"title":"The Re-mediation of Legacy and New Media on Twitter: A Six-Language Comparison of the European Social Media Discourse on Migration","authors":"Mike Farjam, Anamaria Dutceac Segesten","doi":"10.1177/08944393241246101","DOIUrl":"https://doi.org/10.1177/08944393241246101","url":null,"abstract":"Scholarly literature has demonstrated that hybridity transforms both legacy and new media, but that this change is not even. We treat social media platforms as arenas of remediation, where users share and add their own context to information produced by both media subtypes and compare social media conversations about migration in six European languages that include links to either traditional or new media during 2015–2019. We use a mix of computational and statistical methods to analyze 3.5 million (re)tweets and 500,000 links shared within them. We identify the main differences in agenda setting power, function, and tone present within tweets that include links to legacy or new media. Our results show that discourses are similar across languages but clearly different when remediating legacy and new media. Trust in legacy media is correlated with higher proportion of shared links from legacy media and reversely related to the proportion of shared links from new media sources. Considering the volume and timing of the remediated content, we conclude that legacy media retains its agenda setting power. New media linked content tends to cover migration in association to subjects such as Islam or terrorism and to express strong critical opinions against migrants/refugees. The language used is more toxic than in legacy media linked content. The tweets remediating legacy media articles covered topics like domestic or European politics, causes of refugee arrivals and procedures to give them protection. Thus, legacy and new media remediated content differs in both tone and function: toxicity is low and factuality high for content linking to legacy media, with the reverse being true for new media remediations.","PeriodicalId":506768,"journal":{"name":"Social Science Computer Review","volume":"143 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140693441","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Artificial intelligence has sophisticated social and economic effects that cannot be ignored. Based on a thorough review of the development of artificial intelligence, this paper systematically explores the mechanism of the impact of artificial intelligence on economic growth through technology, value and application three paths, which is starting from the perspective of the population external system. In order to verify the rationality of the paths, the effect of artificial intelligence on economic growth from the perspective of population external system is rigorously estimated using artificial intelligence and macroeconomic data for China from 2011 to 2019. The findings are as follows. Firstly, there is a significant positive effect of artificial intelligence on the economic growth from the perspective of the population external system. This positive effect is sufficiently robust over the sample-wide period. Secondly, there is significant regional heterogeneity in the effect of artificial intelligence on economic growth from the perspective of the population external system. The low levels of artificial intelligence development impeded the economic growth, the middle levels of artificial intelligence development contributed significantly to the economic growth, and the high levels of artificial intelligence development did not show a significant contribution to the economic growth. In view of this, future policies should be designed in terms of revitalizing the value of the artificial intelligence stock, exploring the value potential of artificial intelligence and regulating it in a hierarchical manner.
{"title":"The Impact of Artificial Intelligence on Economic Growth From the Perspective of Population External System","authors":"Xueyi Wang, Taiyi He, Shengzhe Wang, Haoxiang Zhao","doi":"10.1177/08944393241246100","DOIUrl":"https://doi.org/10.1177/08944393241246100","url":null,"abstract":"Artificial intelligence has sophisticated social and economic effects that cannot be ignored. Based on a thorough review of the development of artificial intelligence, this paper systematically explores the mechanism of the impact of artificial intelligence on economic growth through technology, value and application three paths, which is starting from the perspective of the population external system. In order to verify the rationality of the paths, the effect of artificial intelligence on economic growth from the perspective of population external system is rigorously estimated using artificial intelligence and macroeconomic data for China from 2011 to 2019. The findings are as follows. Firstly, there is a significant positive effect of artificial intelligence on the economic growth from the perspective of the population external system. This positive effect is sufficiently robust over the sample-wide period. Secondly, there is significant regional heterogeneity in the effect of artificial intelligence on economic growth from the perspective of the population external system. The low levels of artificial intelligence development impeded the economic growth, the middle levels of artificial intelligence development contributed significantly to the economic growth, and the high levels of artificial intelligence development did not show a significant contribution to the economic growth. In view of this, future policies should be designed in terms of revitalizing the value of the artificial intelligence stock, exploring the value potential of artificial intelligence and regulating it in a hierarchical manner.","PeriodicalId":506768,"journal":{"name":"Social Science Computer Review","volume":"10 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140717016","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-22DOI: 10.1177/08944393241235183
Xu Ren, Yali Hao, Jing Xu
The primary aim of this paper is to study the impact of teleworkers’ psychology on their job performance and how teleworkers relieve negative emotions to improve job performance through enterprise social media (ESM) from the conservation of resources theory perspective. An online survey was sent to 835 teleworkers from industries such as finance, automobile, IT, construction, and logistics from March to May 2022 in China. Useful data from 218 respondents were collected and analyzed to test the hypothesized relationships by partial least squares-structural equation modeling (PLS-SEM) method. The findings show that family-to-work conflict and social isolation positively influence employees’ emotional exhaustion and emotional exhaustion further negatively influences their job performance. The visibility affordance and association affordance of ESM can reduce family-to-work conflict and social isolation, thus reducing teleworkers’ emotional exhaustion. The employees’ psychological resilience negatively moderates the positive effects which family-to-work conflict and social isolation have on emotional exhaustion. This paper studies the relieving effects of ESM affordance on teleworkers’ emotional exhaustion and reveals their defending mechanism for avoiding entering into the negative psychological state. Furthermore, this paper supplies beneficial practical suggestions for managers and teleworkers to improve job performance when working from home.
{"title":"How do Teleworkers Relieve Negative Emotions to Improve Job Performance Through Enterprise Social Media? The Conservation of Resources Theory View","authors":"Xu Ren, Yali Hao, Jing Xu","doi":"10.1177/08944393241235183","DOIUrl":"https://doi.org/10.1177/08944393241235183","url":null,"abstract":"The primary aim of this paper is to study the impact of teleworkers’ psychology on their job performance and how teleworkers relieve negative emotions to improve job performance through enterprise social media (ESM) from the conservation of resources theory perspective. An online survey was sent to 835 teleworkers from industries such as finance, automobile, IT, construction, and logistics from March to May 2022 in China. Useful data from 218 respondents were collected and analyzed to test the hypothesized relationships by partial least squares-structural equation modeling (PLS-SEM) method. The findings show that family-to-work conflict and social isolation positively influence employees’ emotional exhaustion and emotional exhaustion further negatively influences their job performance. The visibility affordance and association affordance of ESM can reduce family-to-work conflict and social isolation, thus reducing teleworkers’ emotional exhaustion. The employees’ psychological resilience negatively moderates the positive effects which family-to-work conflict and social isolation have on emotional exhaustion. This paper studies the relieving effects of ESM affordance on teleworkers’ emotional exhaustion and reveals their defending mechanism for avoiding entering into the negative psychological state. Furthermore, this paper supplies beneficial practical suggestions for managers and teleworkers to improve job performance when working from home.","PeriodicalId":506768,"journal":{"name":"Social Science Computer Review","volume":"16 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140441333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-12DOI: 10.1177/08944393241226607
Di Mu, Simai Zhang, Ting Zhu, Yong Zhou, Wei Zhang
Following a comprehensive analysis of the initial three generations of prisoner risk assessment tools, the field has observed a notable prominence in the integration of fourth-generation tools and machine learning techniques. However, limited efforts have been made to address the explainability of data-driven prediction models and their connection with treatment recommendations. Our primary objective was to develop predictive models for assessing the likelihood of recidivism among prisoners released from their index incarceration within 1-year, 2-year, and 5-year timeframes. We aimed to enhance interpretability using SHapley Additive exPlanations (SHAP). We collected data from 20,457 in-prison records from February 10, 2005, to August 25, 2021, sourced from a Southwestern China prison’s data management system. Recidivism records were officially determined through data mining from an official website and combined identification data from neighboring prisons. We employed five machine learning algorithms, considering sociodemographic, physical health, psychological assessments, criminological characteristics, crime history, social support, and in-prison behaviors as factors. For interpretability, SHAP was applied to reveal feature contributions. Findings indicated that young prisoners accused of larceny, previous convictions, lower fines, and limited family support faced higher reoffending risk. Conversely, middle-aged and senior prisoners with no prior convictions, lower monthly supermarket expenses, and positive psychological test results had lower reoffending risk. We also explored interactions between significant predictive features, such as prisoner age at incarceration initiation and primary accusation, and the duration of current incarceration and cumulative prior incarcerations. Notably, our models consistently exhibited high performance, as shown by AUC on the test dataset across time windows. Interpretability results provided insights into evolving risk factors over time, valuable for intervention with high-risk individuals. These insights, with additional validation, could offer dynamic prisoner information for stakeholders. Moreover, interpretability results can be seamlessly integrated into prison and court management systems as a valuable risk assessment tool.
{"title":"Prediction of Recidivism and Detection of Risk Factors Under Different Time Windows Using Machine Learning Techniques","authors":"Di Mu, Simai Zhang, Ting Zhu, Yong Zhou, Wei Zhang","doi":"10.1177/08944393241226607","DOIUrl":"https://doi.org/10.1177/08944393241226607","url":null,"abstract":"Following a comprehensive analysis of the initial three generations of prisoner risk assessment tools, the field has observed a notable prominence in the integration of fourth-generation tools and machine learning techniques. However, limited efforts have been made to address the explainability of data-driven prediction models and their connection with treatment recommendations. Our primary objective was to develop predictive models for assessing the likelihood of recidivism among prisoners released from their index incarceration within 1-year, 2-year, and 5-year timeframes. We aimed to enhance interpretability using SHapley Additive exPlanations (SHAP). We collected data from 20,457 in-prison records from February 10, 2005, to August 25, 2021, sourced from a Southwestern China prison’s data management system. Recidivism records were officially determined through data mining from an official website and combined identification data from neighboring prisons. We employed five machine learning algorithms, considering sociodemographic, physical health, psychological assessments, criminological characteristics, crime history, social support, and in-prison behaviors as factors. For interpretability, SHAP was applied to reveal feature contributions. Findings indicated that young prisoners accused of larceny, previous convictions, lower fines, and limited family support faced higher reoffending risk. Conversely, middle-aged and senior prisoners with no prior convictions, lower monthly supermarket expenses, and positive psychological test results had lower reoffending risk. We also explored interactions between significant predictive features, such as prisoner age at incarceration initiation and primary accusation, and the duration of current incarceration and cumulative prior incarcerations. Notably, our models consistently exhibited high performance, as shown by AUC on the test dataset across time windows. Interpretability results provided insights into evolving risk factors over time, valuable for intervention with high-risk individuals. These insights, with additional validation, could offer dynamic prisoner information for stakeholders. Moreover, interpretability results can be seamlessly integrated into prison and court management systems as a valuable risk assessment tool.","PeriodicalId":506768,"journal":{"name":"Social Science Computer Review","volume":"13 8","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139532757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-04DOI: 10.1177/08944393231224539
Tin Trung Nguyen, Van Thi Thanh Tran, Minh Tu Tran Hoang
Social networking sites (SNSs) have emerged as parallel societies, providing individuals with a platform to interact with peers and construct their desired self-identities. However, maintaining a positive image and safeguarding oneself from social judgment often necessitate self-censorship in self-identity expression. Drawing upon the privacy calculus theory, this study investigates how SNS users engage in a rational cost–benefit analysis between peer privacy concerns and self-presentation when deciding whether to actively or passively use SNSs. Findings from a variance-based analysis—partial least squares structural equation modeling (PLS-SEM)—to a sample of 394 Facebook users revealed that active use was primarily driven by perceived benefits, while passive use was triggered by perceived privacy costs. However, employing a case-based analysis—fuzzy-set qualitative comparative analysis (fsQCA), the present study uncovered that while some SNS users do not conform to the privacy calculus, many others do, thereby confirming the proposed dual privacy calculus model for SNS use. These findings resolve the contradictory findings from previous research on the privacy calculus model. This study extends the literature on the privacy calculus theory by developing a dual peer privacy calculus model to understand SNS users’ passive and active uses and validate the significance of peer privacy concerns on these behavioral patterns. This study underscores critical factors influencing SNS usage patterns, empowering platform developers to provide users with effective tools to combat privacy violations by peers, thereby promoting increased active engagement.
{"title":"How Peer Privacy Concerns Affect Active and Passive Uses of Social Networking Sites: A Dual Peer Privacy Calculus Model","authors":"Tin Trung Nguyen, Van Thi Thanh Tran, Minh Tu Tran Hoang","doi":"10.1177/08944393231224539","DOIUrl":"https://doi.org/10.1177/08944393231224539","url":null,"abstract":"Social networking sites (SNSs) have emerged as parallel societies, providing individuals with a platform to interact with peers and construct their desired self-identities. However, maintaining a positive image and safeguarding oneself from social judgment often necessitate self-censorship in self-identity expression. Drawing upon the privacy calculus theory, this study investigates how SNS users engage in a rational cost–benefit analysis between peer privacy concerns and self-presentation when deciding whether to actively or passively use SNSs. Findings from a variance-based analysis—partial least squares structural equation modeling (PLS-SEM)—to a sample of 394 Facebook users revealed that active use was primarily driven by perceived benefits, while passive use was triggered by perceived privacy costs. However, employing a case-based analysis—fuzzy-set qualitative comparative analysis (fsQCA), the present study uncovered that while some SNS users do not conform to the privacy calculus, many others do, thereby confirming the proposed dual privacy calculus model for SNS use. These findings resolve the contradictory findings from previous research on the privacy calculus model. This study extends the literature on the privacy calculus theory by developing a dual peer privacy calculus model to understand SNS users’ passive and active uses and validate the significance of peer privacy concerns on these behavioral patterns. This study underscores critical factors influencing SNS usage patterns, empowering platform developers to provide users with effective tools to combat privacy violations by peers, thereby promoting increased active engagement.","PeriodicalId":506768,"journal":{"name":"Social Science Computer Review","volume":"82 8","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139387323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-01-02DOI: 10.1177/08944393231220165
Piotr Konieczny, Włodzimierz Lewoniewski
As one of the most popular sources of information in the world, Wikipedia is edited by a large, global community of contributors. User-generated nature of this online encyclopedia ensures that the information reflects a wide range of topics. Hovewer, Wikipedia articles are created and edited independently in each language version. Therefore, some topics may be presented with varying degrees of completeness depending on their importance in a particular language community. In this paper, we quantified the concept of Americanization on a global scale through comparative analysis of the coverage of American topics in different language versions of Wikipedia. For this purpose, we analyzed over 90 million Wikidata items and 40 million Wikipedia articles in 58 languages. We discussed whether Americanization is more or less dominant in different languages, regions, and cultures. We showed that the interest in American topics is not universal. Western, developed countries are more Americanized (more interested in topics related to America) than the rest of the world. This is the first global, quantitative confirmation of issues often hypothesized, or assumed, in the literature on Americanization and related phenomena. This study shows that Wikipedia and Wikidata can allow quantification of social science concepts that previously were considered not realistically measurable. Finally, the presented research is also relevant to the discourses on the biases of Wikipedia.
{"title":"Quantifying Americanization: Coverage of American Topics in Different Wikipedias","authors":"Piotr Konieczny, Włodzimierz Lewoniewski","doi":"10.1177/08944393231220165","DOIUrl":"https://doi.org/10.1177/08944393231220165","url":null,"abstract":"As one of the most popular sources of information in the world, Wikipedia is edited by a large, global community of contributors. User-generated nature of this online encyclopedia ensures that the information reflects a wide range of topics. Hovewer, Wikipedia articles are created and edited independently in each language version. Therefore, some topics may be presented with varying degrees of completeness depending on their importance in a particular language community. In this paper, we quantified the concept of Americanization on a global scale through comparative analysis of the coverage of American topics in different language versions of Wikipedia. For this purpose, we analyzed over 90 million Wikidata items and 40 million Wikipedia articles in 58 languages. We discussed whether Americanization is more or less dominant in different languages, regions, and cultures. We showed that the interest in American topics is not universal. Western, developed countries are more Americanized (more interested in topics related to America) than the rest of the world. This is the first global, quantitative confirmation of issues often hypothesized, or assumed, in the literature on Americanization and related phenomena. This study shows that Wikipedia and Wikidata can allow quantification of social science concepts that previously were considered not realistically measurable. Finally, the presented research is also relevant to the discourses on the biases of Wikipedia.","PeriodicalId":506768,"journal":{"name":"Social Science Computer Review","volume":"93 13","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139390481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：