Social Science Computer Review最新文献_第3页

Large Language Models Outperform Expert Coders and Supervised Classifiers at Annotating Political Social Media Messages 大语言模型在注释政治社交媒体信息方面优于专家编码员和监督分类器

IF 4.1 2区社会学

Social Science Computer Review Pub Date : 2024-09-23 DOI: 10.1177/08944393241286471

Petter Törnberg

{"title":"Large Language Models Outperform Expert Coders and Supervised Classifiers at Annotating Political Social Media Messages","authors":"Petter Törnberg","doi":"10.1177/08944393241286471","DOIUrl":"https://doi.org/10.1177/08944393241286471","url":null,"abstract":"Instruction-tuned Large Language Models (LLMs) have recently emerged as a powerful new tool for text analysis. As these models are capable of zero-shot annotation based on instructions written in natural language, they obviate the need of large sets of training data—and thus bring potential paradigm-shifting implications for using text as data. While the models show substantial promise, their relative performance compared to human coders and supervised models remains poorly understood and subject to significant academic debate. This paper assesses the strengths and weaknesses of popular fine-tuned AI models compared to both conventional supervised classifiers and manual annotation by experts and crowd workers. The task used is to identify the political affiliation of politicians based on a single X/Twitter message, focusing on data from 11 different countries. The paper finds that GPT-4 achieves higher accuracy than both supervised models and human coders across all languages and country contexts. In the US context, it achieves an accuracy of 0.934 and an inter-coder reliability of 0.982. Examining the cases where the models fail, the paper finds that the LLM—unlike the supervised models—correctly annotates messages that require interpretation of implicit or unspoken references, or reasoning on the basis of contextual knowledge—capacities that have traditionally been understood to be distinctly human. The paper thus contributes to our understanding of the revolutionary implications of LLMs for text analysis within the social sciences.","PeriodicalId":49509,"journal":{"name":"Social Science Computer Review","volume":"27 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142313736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Status Spill-Over in Cryptomarket for Illegal Goods 非法商品加密市场的地位溢出效应

IF 4.1 2区社会学

Social Science Computer Review Pub Date : 2024-09-21 DOI: 10.1177/08944393241286339

Filippo Andrei, Giuseppe Alessandro Veltri

{"title":"Status Spill-Over in Cryptomarket for Illegal Goods","authors":"Filippo Andrei, Giuseppe Alessandro Veltri","doi":"10.1177/08944393241286339","DOIUrl":"https://doi.org/10.1177/08944393241286339","url":null,"abstract":"Information technologies have transformed many aspects of social life, including how illegal goods are exchanged. Illegal online markets are now flourishing on various channels: the surface web (all websites accessible through a standard browser), the dark web (an encrypted internet network only accessible via anonymous browsers), and encrypted messaging applications installed on smartphones. These marketplaces take many forms, including simple web shops, chat rooms, forums, social media marketplaces, and platforms. This study focuses on the largest known darknet platform to date: AlphaBay. This cryptomarket operated from December 2014 until July 2017, when an international police operation shut it down. The dataset contains 6033 vendor profiles collected in January 2017. Using three generalized additive models (GAMs), we show that seller status positively affects sales, revenue, and sales through finalized early payment. Once sellers gain status on the platforms, they make more sales without a semi-institutionalized form of payment (e.g. escrow). On the other hand, buyers relying on status metrics as cognitive shortcuts tend to choose vendors even if they do not offer payment protection tools.","PeriodicalId":49509,"journal":{"name":"Social Science Computer Review","volume":"18 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142306251","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Network Issue Agenda Setting on Facebook: Exploring the Interplay Between Polarized Campaigns and Party Supporters Facebook 上的网络议题议程设置：探索两极分化的竞选活动与政党支持者之间的相互作用

IF 4.1 2区社会学

Social Science Computer Review Pub Date : 2024-09-20 DOI: 10.1177/08944393241286149

Zahedur Rahman Arman

引用次数: 0

Unveiling the Veiled Threat: The Impact of Bots on COVID-19 Health Communication 揭开隐性威胁的面纱：机器人对 COVID-19 健康传播的影响

IF 4.1 2区社会学

Social Science Computer Review Pub Date : 2024-09-10 DOI: 10.1177/08944393241275641

Ali Unlu, Sophie Truong, Nitin Sawhney, Tuukka Tammi

{"title":"Unveiling the Veiled Threat: The Impact of Bots on COVID-19 Health Communication","authors":"Ali Unlu, Sophie Truong, Nitin Sawhney, Tuukka Tammi","doi":"10.1177/08944393241275641","DOIUrl":"https://doi.org/10.1177/08944393241275641","url":null,"abstract":"This article presents the results of a comprehensive study examining the influence of bots on the dissemination of COVID-19 misinformation and negative vaccine stance on Twitter over a period of three years. The research employed a tripartite methodology: text classification, topic modeling, and network analysis to explore this phenomenon. Text classification, leveraging the Turku University FinBERT pre-trained embeddings model, differentiated between misinformation and vaccine stance detection. Bot-like Twitter accounts were identified using the Botometer software, and further analysis was implemented to distinguish COVID-19 specific bot accounts from regular bots. Network analysis illuminated the communication patterns of COVID-19 bots within retweet and mention networks. The findings reveal that these bots exhibit distinct characteristics and tactics that enable them to influence public discourse, particularly showing an increased activity in COVID-19-related conversations. Topic modeling analysis uncovers that COVID-19 bots predominantly focused on themes such as safety, political/conspiracy theories, and personal choice. The study highlights the critical need to develop effective strategies for detecting and countering bot influence. Essential actions include using clear and concise language in health communications, establishing strategic partnerships during crises, and ensuring the authenticity of user accounts on digital platforms. The findings underscore the pivotal role of bots in propagating misinformation related to COVID-19 and vaccines, highlighting the necessity of identifying and mitigating bot activities for effective intervention.","PeriodicalId":49509,"journal":{"name":"Social Science Computer Review","volume":"82 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142166075","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

To Follow or Not to Follow: Estimating Political Opinion From Twitter Data Using a Network-Based Machine Learning Approach 关注或不关注：使用基于网络的机器学习方法从推特数据中估计政治观点

IF 4.1 2区社会学

Social Science Computer Review Pub Date : 2024-09-04 DOI: 10.1177/08944393241279418

Nils Brandenstein, Christian Montag, Cornelia Sindermann

{"title":"To Follow or Not to Follow: Estimating Political Opinion From Twitter Data Using a Network-Based Machine Learning Approach","authors":"Nils Brandenstein, Christian Montag, Cornelia Sindermann","doi":"10.1177/08944393241279418","DOIUrl":"https://doi.org/10.1177/08944393241279418","url":null,"abstract":"Studying political opinions of citizens stands as a fundamental pursuit for both policymakers and researchers. While traditional surveys remain the primary method to investigate individual political opinions, the advent of social media data (SMD) offers novel prospects. However, the number of studies using SMD to extract individuals’ political opinions are limited and differ greatly in their methodological approaches and levels of success. Recent studies highlight the benefits of analyzing individuals’ social media network structure to estimate political opinions. Nevertheless, current methodologies exhibit limitations, including the use of simplistic linear models and a predominant focus on samples from the United States. Addressing these issues, we employ an unsupervised Variational Autoencoder (VAE) machine learning model to extract individual opinion estimates from SMD of N = 276 008 German Twitter (now called ’X’) users, compare its performance to a linear model and validate model estimates on self-reported opinion measures. Our findings suggest that the VAE captures Twitter users’ network structure more precisely, leading to higher accuracy in following decision predictions and associations with self-reported political ideology and voting intentions. Our study emphasizes the need for advanced analytical approaches capable to capture complex relationships in social media networks when studying political opinion, at least in non-US contexts.","PeriodicalId":49509,"journal":{"name":"Social Science Computer Review","volume":"17 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142142553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Does the Media’s Partisanship Influence News Coverage on Artificial Intelligence Issues? Media Coverage Analysis on Artificial Intelligence Issues 媒体的党派倾向会影响对人工智能问题的新闻报道吗？人工智能问题的媒体报道分析

IF 4.1 2区社会学

Social Science Computer Review Pub Date : 2024-09-02 DOI: 10.1177/08944393241268526

Mikyung Chang

{"title":"Does the Media’s Partisanship Influence News Coverage on Artificial Intelligence Issues? Media Coverage Analysis on Artificial Intelligence Issues","authors":"Mikyung Chang","doi":"10.1177/08944393241268526","DOIUrl":"https://doi.org/10.1177/08944393241268526","url":null,"abstract":"This study aims to analyze news coverage on artificial intelligence (AI) issues and highlight the characteristics and differences in reporting based on media partisanship. By examining AI-related news in the South Korean media, this study reveals how conservative and progressive outlets frame the issue differently. The analysis found that conservative media coverage predominantly focuses on positive aspects, emphasizing development value frames such as the benefits and societal progress brought by AI. In contrast, progressive media often highlight crisis value frames, focusing on issues like side effects, ethical concerns, and legislation surrounding AI. These partisan differences reflect fundamental societal priorities and influence public discourse and policy agendas. Understanding media framing is crucial for fostering informed public dialogue on the societal significance of AI and promoting evidence-based decision-making. By recognizing partisan biases and critically evaluating media coverage, citizens can engage in constructive discourse beyond ideological divides. This study underscores the role of the media in promoting interdisciplinary discussions about the future trajectory of AI and in preparing society for its impacts. Ultimately, evidence-based public discourse is essential for shaping responsible AI policies and mitigating potential risks in the digital age.","PeriodicalId":49509,"journal":{"name":"Social Science Computer Review","volume":"18 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142123492","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

TikTok Brain: An Investigation of Short-Form Video Use, Self-Control, and Phubbing 嘀嗒大脑对短视频使用、自控力和幻觉的研究

IF 4.1 2区社会学

Social Science Computer Review Pub Date : 2024-08-29 DOI: 10.1177/08944393241279422

Meredith E. David, James A. Roberts

{"title":"TikTok Brain: An Investigation of Short-Form Video Use, Self-Control, and Phubbing","authors":"Meredith E. David, James A. Roberts","doi":"10.1177/08944393241279422","DOIUrl":"https://doi.org/10.1177/08944393241279422","url":null,"abstract":"Phubbing (phone snubbing) has become the norm in (im)polite society. A vast majority of US adults report using their phones during a recent social interaction. Using one’s phone in the presence of others has been shown to have a negative impact on relationships among co-workers, friends, family, and romantic partners. Recent research suggests viewing short-form videos (SFVs) (e.g., TikTok) is more addictive/immersive than traditional social media (e.g., Facebook) leading to a greater likelihood of phubbing others. Across two studies, the present research investigates the relationship between SFV viewing and phubbing and the possible mediating effect of self-control. We also test whether TikTok has a stronger relationship with phubbing than Instagram Reels and YouTube Shorts, two popular SFV purveyors. Study 1 (282 college students) finds that viewing TikTok videos is positively associated with phubbing others and this relationship is mediated by self-control. Interestingly, Study 1 also finds that this relationship does not hold for Instagram Reels and YouTube shorts. Using two different measures of self-control, Study 2 (198 adults) provides additional support for the mediating effect of self-control on the SFV viewing—phubbing relationship. Again, the model is only supported for TikTok SFV viewing, not Instagram or YouTube. In sum, the viewing of carefully curated short TikTok videos, often 30–60 seconds in length, undermines self-control which is associated with increased phubbing behavior. Implications of the present study’s findings expand far beyond phubbing. Self-control plays a central role in nearly all human decision making and behavior. Suggestions for future research are offered.","PeriodicalId":49509,"journal":{"name":"Social Science Computer Review","volume":"10 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142100644","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

CORA: An Open-Source Software Tool for Combinational Regularity Analysis CORA：用于组合规律性分析的开源软件工具

IF 4.1 2区社会学

Social Science Computer Review Pub Date : 2024-08-29 DOI: 10.1177/08944393241275640

Lusine Mkrtchyan, Alrik Thiem, Zuzana Sebechlebská

引用次数: 0

Remember, You Can Complete This Survey Online! Web Survey Links and QR Codes in a Mixed-Mode Web and Mail General Population Survey 请记住，您可以在线完成本调查！混合模式网络和邮件普通人群调查中的网络调查链接和 QR 代码

IF 4.1 2区社会学

Social Science Computer Review Pub Date : 2024-08-24 DOI: 10.1177/08944393241277553

Kristen Olson, Amanda Ganshert

{"title":"Remember, You Can Complete This Survey Online! Web Survey Links and QR Codes in a Mixed-Mode Web and Mail General Population Survey","authors":"Kristen Olson, Amanda Ganshert","doi":"10.1177/08944393241277553","DOIUrl":"https://doi.org/10.1177/08944393241277553","url":null,"abstract":"Recruitment materials for concurrent mixed-mode self-administered web and mail surveys must communicate information about multiple modes simultaneously. Providing the link to the web survey on the cover of the paper questionnaire or including a QR code to access the web survey may increase the visibility of the web mode and thus increase the proportion of people who participate via the web, but whether including either piece of information does so has received surprisingly little empirical attention. In this paper, we examine the results of experiments embedded in two general population probability-based concurrent mixed-mode surveys of Nebraska adults. First, in the Labor Availability Survey of the Greater Omaha Area, respondents were randomly assigned to receive the web link and login information on the cover or the paper questionnaire without this information (all had web information in the cover letter). We then replicated and extended this experiment in the Labor Availability Survey of Northeast Nebraska. The questionnaire cover experiment was fully crossed with the presence or absence of a QR code to access the web survey. Neither of these design features affected response rates or speed of response, but the link on the questionnaire significantly increased the proportion of respondents who participated by web and the QR code significantly increased the proportion of respondents who participated by smartphone. Sample composition was largely unaffected on most characteristics, although the respondent pool was less similar to the population on education when the link was on the questionnaire. About 20% of respondents used a smartphone when typing in a survey link, but virtually all respondents used a smartphone when scanning the QR code. Survey researchers can include a link on the cover of the questionnaire to increase web participation rates in mixed-mode surveys. QR codes can be used when smartphone participation is desired.","PeriodicalId":49509,"journal":{"name":"Social Science Computer Review","volume":"7 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142050629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Understanding Narratives of Uncertainty in Fertility Intentions of Dutch Women: A Neural Topic Modeling Approach 理解荷兰妇女生育意愿中的不确定性叙述：神经主题建模方法

IF 4.1 2区社会学

Social Science Computer Review Pub Date : 2024-08-24 DOI: 10.1177/08944393241269406

Xiao Xu, Anne Gauthier, Gert Stulp, Antal van den Bosch

{"title":"Understanding Narratives of Uncertainty in Fertility Intentions of Dutch Women: A Neural Topic Modeling Approach","authors":"Xiao Xu, Anne Gauthier, Gert Stulp, Antal van den Bosch","doi":"10.1177/08944393241269406","DOIUrl":"https://doi.org/10.1177/08944393241269406","url":null,"abstract":"Uncertainty in fertility intentions is a major obstacle to understanding contemporary trends in fertility decision-making and its outcomes. Quantifying this uncertainty by structural factors such as income, ethnicity, and housing conditions is recognized as insufficient. A recently proposed framework on subjective narratives has opened up a new way to gauge factors behind fertility decision-making and uncertainty. Through surveys, such narratives can be elicited with open-ended questions (OEQs). However, analyzing answers to OEQs typically involves extensive human coding, imposing constraints on sample size. Natural Language Processing (NLP) techniques assist researchers in grasping aspects of the underlying reasoning behind responses with much less human effort. In this study, using automatic neural topic modeling methods, we identify and interpret topics and themes underlying the narratives on fertility intention uncertainty of women in the Netherlands. We used Contextualized Topic Models (CTMs), a neural topic model using pre-trained representations of Dutch language, to conduct our analyses. Our results show that nine topics dominate the narratives about fertility planning, with age and health-related issues as the most prominent ones. In addition, we found that uncertainty in fertility intentions is not homogeneous, as women who feel uncertain due to real-life constraints and those who have no fertility plans at all put their stress on vastly different narratives.","PeriodicalId":49509,"journal":{"name":"Social Science Computer Review","volume":"25 1","pages":""},"PeriodicalIF":4.1,"publicationDate":"2024-08-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142050627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"社会学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0