Affective polarisation, that is, “view[ing] opposing partisans negatively and copartisans positively” (Iyengar & Westwood, 2015 p. 691), seems to have become a buzzword in field of political behaviour. Since the seminal article of Iyengar et al. (2012), where the concept was delineated for the first time, a plethora of studies engaged with it, making it one of the most popular constructs of the last decade. However, until about four years ago, the study of affective polarisation was primarily a US-centric endeavour. In Europe, affective polarisation has attracted scholarly attention only in about the last four years. This is likely due to the fact that in countries that do not have a two-party system, the feelings of in-group and out-group membership, on which affective polarisation rests, are less immediately visible.
Tushar Pranav, Eshan Pandey, Austria Lyka Diane Bala
et al.
Vision-Language Models (VLMs) excel in multimodal tasks but often exhibit Western-centric biases, limiting their effectiveness in culturally diverse regions like Southeast Asia (SEA). To address this, we introduce RICE-VL, a novel benchmark evaluating VLM cultural understanding across 11 ASEAN countries. RICE-VL includes over 28,000 human-curated Visual Question Answering (VQA) samples -- covering True or False, Fill-in-the-Blank, and open-ended formats -- and 1,000 image-bounding box pairs for Visual Grounding, annotated by culturally informed experts across 14 sub-ground categories. We propose SEA-LAVE, an extension of the LAVE metric, assessing textual accuracy, cultural alignment, and country identification. Evaluations of six open- and closed-source VLMs reveal significant performance gaps in low-resource countries and abstract cultural domains. The Visual Grounding task tests models' ability to localize culturally significant elements in complex scenes, probing spatial and contextual accuracy. RICE-VL exposes limitations in VLMs' cultural comprehension and highlights the need for inclusive model development to better serve diverse global populations.
The ability of Natural Language Processing (NLP) methods to categorize text into multiple classes has motivated their use in online content moderation tasks, such as hate speech and fake news detection. However, there is limited understanding of how or why these methods make such decisions, or why certain content is moderated in the first place. To investigate the hidden mechanisms behind content moderation, we explore multiple directions: 1) training classifiers to reverse-engineer content moderation decisions across countries; 2) explaining content moderation decisions by analyzing Shapley values and LLM-guided explanations. Our primary focus is on content moderation decisions made across countries, using pre-existing corpora sampled from the Twitter Stream Grab. Our experiments reveal interesting patterns in censored posts, both across countries and over time. Through human evaluations of LLM-generated explanations across three LLMs, we assess the effectiveness of using LLMs in content moderation. Finally, we discuss potential future directions, as well as the limitations and ethical considerations of this work. Our code and data are available at https://github.com/causalNLP/censorship
R. Rahman, Martin Heinberg, Sourindra Banerjee
et al.
Abundant consumer data has made decision-making more complicated, rather than simple, for marketers. This raises an important question about which variables in the data contain reliable information for retailers to predict future consumer purchase value (CPV) to guide strategic decisions. The authors address this question by exploring the variables “distinctive choice of brand country of origin” (DBCOO) and “country of origin diversity” (COO diversity) as analytical tools to extract insights from consumer purchase data. Building on signaling theory, the authors theorize and empirically test that DBCOO and COO diversity in a consumer's purchase history can signal, and therefore help predict, CPV. Moreover, the authors explore high-involvement product categories and purchase frequency as boundary conditions to develop a comprehensive framework of COO signals as strategic analytical tools. They find that DBCOO in a consumer's purchase history indeed increases CPV and that this relationship is enhanced for high-involvement product categories but moderated curvilinearly by purchase frequency. Moreover, they find that the COO diversity–CPV link is positive but interacts negatively with both moderators. This allows retailers to successfully distinguish high- from low-CPV consumers and thus enables them to manage the marketing mix and resources more effectively.
This paper proposes a Few-shot Learning (FSL) approach for detecting Presentation Attacks on ID Cards deployed in a remote verification system and its extension to new countries. Our research analyses the performance of Prototypical Networks across documents from Spain and Chile as a baseline and measures the extension of generalisation capabilities of new ID Card countries such as Argentina and Costa Rica. Specifically targeting the challenge of screen display presentation attacks. By leveraging convolutional architectures and meta-learning principles embodied in Prototypical Networks, we have crafted a model that demonstrates high efficacy with Few-shot examples. This research reveals that competitive performance can be achieved with as Few-shots as five unique identities and with under 100 images per new country added. This opens a new insight for novel generalised Presentation Attack Detection on ID cards to unknown attacks.
Ted Hsuan Yun Chen, Arttu Malkamäki, Ali Faqeeh
et al.
We identified the Twitter accounts of 941 climate change policy actors across nine countries, and collected their activities from 2017--2022, totalling 48 million activities from 17,700 accounts at different organizational levels. There is considerable temporal and cross-national variation in how prominent climate-related activities were, but all national policy systems generally responded to climate-related events, such as climate protests, in a similar manner. Examining patterns of interaction within and across countries, we find that these national policy systems rarely directly interact with one another, but are connected through consistently engaging with the same content produced by accounts of international organizations, climate activists, and researchers.
Thales Bertaglia, Catalina Goanta, Gerasimos Spanakis
et al.
This paper presents a longitudinal study of more than ten years of activity on Instagram consisting of over a million posts by 400 content creators from four countries: the US, Brazil, Netherlands and Germany. Our study shows differences in the professionalisation of content monetisation between countries, yet consistent patterns; significant differences in the frequency of posts yet similar user engagement trends; and significant differences in the disclosure of sponsored content in some countries, with a direct connection with national legislation. We analyse shifts in marketing strategies due to legislative and platform feature changes, focusing on how content creators adapt disclosure methods to different legal environments. We also analyse the impact of disclosures and sponsored posts on engagement and conclude that, although sponsored posts have lower engagement on average, properly disclosing ads does not reduce engagement further. Our observations stress the importance of disclosure compliance and can guide authorities in developing and monitoring them more effectively.
This paper presents a machine learning approach to classify countries as peaceful or non-peaceful using linguistic patterns extracted from global media articles. We employ vector embeddings and cosine similarity to develop a supervised classification model that effectively identifies peaceful countries. Additionally, we explore the impact of dataset size on model performance, investigating how shrinking the dataset influences classification accuracy. Our results highlight the challenges and opportunities associated with using large-scale text data for peace studies.
Lorenzo Lucchini, Ollin Langle-Chimal, Lorenzo Candeago
et al.
Mobile phone data have played a key role in quantifying human mobility during the COVID-19 pandemic. Existing studies on mobility patterns have primarily focused on regional aggregates in high-income countries, obfuscating the accentuated impact of the pandemic on the most vulnerable populations. Leveraging geolocation data from mobile-phone users and population census for 6 middle-income countries across 3 continents between March and December 2020, we uncovered common disparities in the behavioral response to the pandemic across socioeconomic groups. Users living in low-wealth neighborhoods were less likely to respond by self-isolating, relocating to rural areas, or refraining from commuting to work. The gap in the behavioral responses between socioeconomic groups persisted during the entire observation period. Among users living in low-wealth neighborhoods, those who commute to work in high-wealth neighborhoods pre-pandemic were particularly at risk of experiencing economic stress, facing both the reduction in economic activity in the high-wealth neighborhood and being more likely to be affected by public transport closures due to their longer commute distances. While confinement policies were predominantly country-wide, these results suggest that, when data to identify vulnerable individuals are not readily available, GPS-based analytics could help design targeted place-based policies to aid the most vulnerable.
Emanuel Kohlscheen, Richhild Moessner, Daniel Rees
We test the international applicability of Friedman s famous plucking theory of the business cycle in 12 advanced economies between 1970 and 2021. We find that in countries where labour markets are flexible (Australia, Canada, United Kingdom and United States), unemployment rates typically return to pre-recession levels, in line with Friedman s theory. Elsewhere, unemployment rates are less cyclical. Output recoveries differ less across countries, but more across episodes: on average, half of the decline in GDP during a recession persists. In terms of sectors, declines in manufacturing are typically fully reversed. In contrast, construction-driven recessions, which are often associated with bursting property price bubbles, tend to be persistent.
The study analyzed the impact of financial inclusion on the effectiveness of monetary policy in developing countries. By using a panel data set of 10 developing countries during 2004-2020, the study revealed that the financial inclusion measured by the number of ATM per 100,000 adults had a significant negative effect on monetary policy, whereas the other measure of financial inclusion i.e. the number of bank accounts per 100,000 adults had a positive impact on monetary policy, which is not statistically significant. The study also revealed that foreign direct investment (FDI), lending rate and exchange rate had a positive impact on inflation, but only the effect of lending rate is statistically significant. Therefore, the governments of these countries should make necessary drives to increase the level of financial inclusion as it stabilizes the price level by reducing the inflation in the economy.
INTRODUCTION Head lice infestation is considered as a common dermatological health problem worldwide. This study was aimed to determine the pediculosis prevalence and associated risk factors among school-aged girls enrolled in public elementary schools of the city of Pave, Kermanshah province, West Iran. The study findings will provide an evidence base, upon which a multifaceted intervention against pediculosis can be formulated and implemented. METHODOLOGY In this cross-sectional study, 361 elementary school-aged girls were recruited from October to December 2018 at the beginning of the school year. Diagnosis was made by visual inspection. A structured questionnaire was utilized to collect data about past history of infestation and associated factors. RESULTS A total of 26/361 (7.2%; 95% CI: 4.50-9.90) suffered from pediculosis. Pediculosis was associated with the history of previous infestation (OR: 6.12; 95% CI: 2.68-13.99; p < 0.001), low frequency of bathing (OR: 7.90; 95% CI: 3.36-18.60; p < 0.001), low frequency of hair combing (OR: 3.64; 95% CI: 1.56-8.50; p = 0.004), screening of the student's hair by parents at home (OR: 0.39; 95% CI: 0.19-0.78; p < 0.001) and with the absence of screening by the school health officer in the schools (OR: 7.16; 95% CI: 2.91-17.61; p < 0.001). CONCLUSIONS Synchronized efforts to enhance public knowledge, periodic examination of school-aged children for pediculosis, and proper treatment of infested patients are needed to control the disease in the Iranian elementary schools. The applied strategies in low and middle income countries is suggested to be focused mainly on low cost family-based and school-based initiatives for maximum effectiveness.
Franziska Herbert, Steffen Becker, Leonie Schaewitz
et al.
Misconceptions about digital security and privacy topics in the general public frequently lead to insecure behavior. However, little is known about the prevalence and extent of such misconceptions in a global context. In this work, we present the results of the first large-scale survey of a global population on misconceptions: We conducted an online survey with n = 12, 351 participants in 12 countries on four continents. By investigating influencing factors of misconceptions around eight common security and privacy topics (including E2EE, Wi-Fi, VPN, and malware), we find the country of residence to be the strongest estimate for holding misconceptions. We also identify differences between non-Western and Western countries, demonstrating the need for region-specific research on user security knowledge, perceptions, and behavior. While we did not observe many outright misconceptions, we did identify a lack of understanding and uncertainty about several fundamental privacy and security topics.
R. Charak, J.T.V.M. de Jong, Lidewyde H. Berckmoes
et al.
Studies investigating the associations between histories of childhood maltreatment (CM) in parent-child dyads have primarily involved samples from high-income countries; however, CM rates are higher in low- and middle-income countries. The present study aimed to examine the (a) association between maltreatment in parents and maltreatment of their children through risk (i.e., parent depression) and protective (i.e., parent-child connectedness) factors and (b) associations between CM in children with aggression through posttraumatic stress symptoms (PTSS) and peer/sibling victimization. Participants were 227 parent-child dyads from Burundi, Africa, a low-income country. Parents were 18 years of age or older, and children were 12-18 years (M = 14.76, SD = 1.88, 57.7% female). Among parents, 20.7%-69.5% of participants reported a history of physical and emotional abuse and neglect; among children, the rates of sexual, physical, and emotional abuse ranged from 14.5% to 89.4%. A history of CM in parents was associated with CM in children, B = 0.19, p < .01, and CM in parents was indirectly associated with CM in children through parent-child connectedness, β = .04, 95% CI [.01, .10], and parental depression, β = .08, 95% CI [.03, .15]. In children, maltreatment was positively associated with peer/sibling victimization, and CM was associated with aggression, β = .07, 95% CI [.04, 0.11], through PTSS but not via peer/sibling victimization. Continued efforts to improve CM-related preventive strategies and the accessibility of prevention services are needed to reduce CM in low-income countries such as Burundi.
Simon Byonanuwe, Emmanuel Nzabandora, Baltazar Nyongozi
et al.
Background Premature rupture of membranes (PROM) is a common condition in developed and developing countries and poses a serious threat to the maternal and fetal well-being if not properly managed. This study delineated the prevalence and predictors of PROM in the western part of Uganda so as to guide specific preventive measures. Methods A cross-sectional study design was conducted in the months of September 2019 to November 2019. A total of 334 pregnant women above 28 weeks of gestation admitted at the maternity ward of KIU-TH were consecutively enrolled. Interviewer-administered questionnaires were used to obtain the data. Descriptive statistics followed by binary logistic regression were conducted. All data analyses were conducted using STATA 14.2. Results Of the 334 pregnant women enrolled, the prevalence of PROM was found to be 13.8%. The significant independent predictors associated with lower odds of PROM were no history of urinary tract infection (UTI) in the month preceding enrollment into the study (aOR = 0.5, 95% CI: 0.22-0.69, p = 0.038) and gestational age of 37 weeks or more (aOR = 0.3, 95% CI: 0.14-0.71, p = 0.038) and gestational age of 37 weeks or more (aOR = 0.3, 95% CI: 0.14-0.71, p = 0.038) and gestational age of 37 weeks or more (aOR = 0.3, 95% CI: 0.14-0.71, Conclusions Majorly urinary tract infections, low gestational age, and abortions influence premature rupture of membranes among women. There is a great need for continuous screening and prompt treatment of pregnant women for UTI especially those with history of 3 or more abortions at less than 34 weeks of gestation.
A. Samaddar, R. Cuevas, Marie Claire Custodio
et al.
The EAT-Lancet Commission urgently called for “planetary health diets”. The success of encouraging dietary shifts, however, crucially hinges on people, and more specifically on consumers' culture, context, socioeconomic status, food environment, attitudes, perceptions, beliefs, and behavior towards food choice. In India, enhanced food availability and accessibility do not readily lead to improved nutritional status. Thus, developing planetary health diets in India requires an understanding of systemic drivers of food choice. Food is an essential part of Indian culture and deeply rooted to the country's history, traditions, lifestyles, and customs. Yet, the diversity and cultural drivers of food choice are still insufficiently understood. To address this knowledge gap, we use expert elicitation to contextualize the “gastronomic systems research” framework to a target population of low-to middle-income households to capture the diversity and cultural drivers of food choice and its nutritional implications in rice-based diets in two states in eastern India. The experts catalogued 131 unique dishes associated with five differentiated daily dining occasions. The majority of dishes belong to the starch food group. Morning snacks exhibit the lowest nutritional diversity while dinners feature the highest diversity in both states. In West Bengal, dish options tend to be carbohydrate-rich and energy-dense, and a significant number of dishes are fried and oily. The gastronomic system mapped by the experts provides a useful baseline for nutritionists, policymakers, and food system actors as a first step in the design of nutrition intervention strategies to develop planetary health diets in eastern India.
We studied the research performance of 69 countries by considering two different types of new knowledge: incremental (normal) and fundamental (radical). In principle, these two types of new knowledge should be assessed at two very different levels of citations, but we demonstrate that a simpler assessment can be performed based on the total number of papers (P) and the ratio of the number of papers in the global top 10% of most cited papers divided to the total number of papers (Ptop 10%/P). P represents the quantity, whereas the Ptop 10%/P ratio represents the efficiency. In ideal countries, P and the Ptop 10%/P ratio are linked to the gross domestic product (GDP) and GDP the per capita, respectively. Only countries with high Ptop 10%/P ratios participate actively in the creation of fundamental new knowledge and have Noble laureates. In real countries, the link between economic and scientific wealth can be modified by the technological activity and the research policy. We discuss how technological activity may decrease the Ptop 10%/P ratio while only slightly affecting the capacity to create fundamental new knowledge; in such countries, many papers may report incremental innovations that do not drive the advancement of knowledge. Japan is the clearest example of this, although there are many less extreme examples. Independently of technological activity, research policy has a strong influence on the Ptop 10%/P ratio, which may be higher or lower than expected from the GDP per capita depending on the success of the research policy.
Gianluca Teza, Michele Caraglio, Attilio L. Stella
We show how the Shannon entropy function can be used as a basis to set up complexity measures weighting the economic efficiency of countries and the specialization of products beyond bare diversification. This entropy function guarantees the existence of a fixed point which is rapidly reached by an iterative scheme converging to our self-consistent measures. Our approach naturally allows to decompose into inter-sectorial and intra-sectorial contributions the country competitivity measure if products are partitioned into larger categories. Besides outlining the technical features and advantages of the method, we describe a wide range of results arising from the analysis of the obtained rankings and we benchmark these observations against those established with other economical parameters. These comparisons allow to partition countries and products into various main typologies, with well-revealed characterizing features. Our methods have wide applicability to general problems of ranking in bipartite networks.
In light of the ongoing integration efforts, the question of whether CAPADR economies may benefit from a single currency arises naturally. This paper examines the feasibility of an Optimum Currency Area (OCA) within seven CAPADR countries. We estimate SVAR models to retrieve demand and supply shocks between 2009:01 - 2020:01 and determine their extent of symmetry. We then go on to compute two regional indicators of dispersion and the cost of inclusion into a hypothetical OCA for each country. Our results indicate that asymmetric shocks tend to prevail. In addition, the dispersion indexes show that business cycles have become more synchronous over time. However, CAPADR countries are still sources of cyclical divergence, so that they would incur significant costs in terms of cycle correlation whenever they pursue currency unification. We conclude that the region does not meet the required symmetry and synchronicity for an OCA to be appropiate.