Social Science Computer Review

Papakyriakopoulos, Orestis; Mboya, Arwa M.

doi: 10.1177/08944393211073169pmid: N/A

We perform a socio-computational interrogation of the google search by image algorithm, a main component of the google search engine. We audit the algorithm by presenting it with more than 40 thousands faces of all ages and more than four races and collecting and analyzing the assigned labels with the appropriate statistical tools. We find that the algorithm reproduces white male patriarchal structures, often simplifying, stereotyping and discriminating females and non-white individuals, while providing more positive descriptions of white men. By drawing from Bourdieu’s theory of cultural reproduction, we link these results to the attitudes of the algorithm’s designers, owners, and the dataset the algorithm was trained on. We further underpin the problematic nature of the algorithm by using the ethnographic practice of studying-up: We show how the algorithm places individuals at the top of the tech industry within the socio-cultural reality that they shaped, many times creating biased representations of them. We claim that the use of social-theoretic frameworks such as the above are able to contribute to improved algorithmic accountability, algorithmic impact assessment and provide additional and more critical depth in algorithmic bias and auditing studies. Based on the analysis, we discuss the scientific and design implications and provide suggestions for alternative ways to design just socio-algorithmic systems.

Wang, Xingchao; Zhao, Kun

doi: 10.1177/08944393211072231pmid: N/A

A growing number of studies have suggested that partner phubbing is negatively associated with marital satisfaction. However, little is known about the mediating mechanisms between this relationship. The current study investigated whether marital interaction and marital conflict mediate the relationship between partner phubbing and marital satisfaction. A sample of 470 Chinese married adults completed questionnaires regarding demographics, partner phubbing, marital interaction, marital conflict, and marital satisfaction. The results suggested that (a) partner phubbing was negatively associated with marital satisfaction; (b) both marital interaction and marital conflict partially mediated the association between partner phubbing and marital satisfaction; and (c) marital interaction and marital conflict sequentially mediated the association between partner phubbing and marital satisfaction. These findings promote our understanding of how partner phubbing is associated with marital satisfaction.

Pirannejad, Ali; Ingrams, Alex

doi: 10.1177/08944393211063107pmid: N/A

During the last decade, a new approach to bureaucratic reform in the field of public administration, open government, has aimed to increase government transparency and accountability and improve participation of citizens and other stakeholders of government. In the current era of digital governance transformations, evaluating governmental efforts to become open is a central concern of politicians, policymakers, and researchers. Various global maturity models have been developed, but the majority of them focus on the technological capacities of government rather than the historic affinity of openness and democratic governance. In this study, we attempt to address this problem by conceptualizing how governments harness technology innovations and by prescribing developmental phases for open government. Using the qualitative meta-synthesis method, we compare 10 open government maturity models to find the similarities and differences between them. Finally, we present a comprehensive model which evaluates the open government initiatives holistically and includes the following six major stages: (1) an initial stage; (2) a transparency and accountability stage; (3) an open collaboration stage; (4) a platform stage; (5) a democratic open government stage; and, finally, (6) an open governance stage.

Nicolaidou, Iolie; Egenfeldt-Nielsen, Simon; Zupančič, Rok; Hajslund, Sara; Milioni, Dimitra L.

doi: 10.1177/08944393211073586pmid: N/A

This study describes the design of a serious game for social change (“Fact Finders”) that presents intergroup conflicts through historical inquiry and multiperspectivity. A pre-test post-test experimental design examined the game’s effect on undergraduates’ perceptions of conflicts in history. Participants included 97 Greek Cypriots (direct parties of the conflict) and 79 Slovenians (third parties of the conflict) who interacted with and evaluated the game online. Data sources included a 17-item questionnaire on perceptions of conflicts in history and gameplay learning analytics data. Findings indicated that both groups’ perceptions for historical source evaluation and understanding multiperspectivity changed significantly after the game. The game significantly changed perceptions about the constructedness of history and the ability to overcome their country’s troubled past only for direct parties of the conflict. The study provides empirical evidence demonstrating the potential value of serious games for affecting young people’s perceptions of intractable intergroup conflicts and their desire to overcome troubled pasts.

Khafaei, Taleb; Tavakoli Taraghi, Alireza; Hosseinzadeh, Mehdi; Rezaee, Ali

doi: 10.1177/08944393211055813pmid: N/A

A dynamic Online Social Network is a special type of evolving complex network in which changes occur over time. The structure of a community may change over time due to the relationship changes between its members or with other communities. This is known as a community event. In this paper, we discussed the effect of important individual community features and the lengths of adequate time intervals considered in the analysis of the behavior of social networks on the prediction accuracy of each event. Furthermore, we introduced the extra-structural features as global social network features to justify the relationship between the lengths of time intervals used in the model training by using the best prediction accuracy of events. We found a relationship between the scale of network dynamics and the length of time intervals for observing the spread and decomposed events. Finally, by comparing the accuracy of the model based on time interval length which investigated based on cps-value in this study and using the Event Prediction in Dynamic Social Network (EPDSN) model, the hypothesis of a reverse relationship between cps growth rate and time interval length to obtain better prediction accuracy for both the spread and decomposed events.

Reiter, Franz; Heiss, Raffael; Matthes, Jörg

doi: 10.1177/08944393211056224pmid: N/A

There are rising concerns that social network sites (SNS) facilitate the creation of echo chambers, in which attitude-consistent information becomes the norm while attitude-challenging information is avoided. This study aims to investigate theoretically derived predictors of attitude-consistent and attitude-challenging exposure on SNS. We theorize that three key sets of predictors may influence attitude-consistent and attitude-challenging exposure: ideology, cognitive, and behavioral indicators of political involvement, and network characteristics. In a two-wave panel study, we predict the frequency of attitude-consistent and attitude-challenging exposure as well as relative attitude-consistent exposure, measured as attitude-consistent exposure as a share of overall opinion exposure. The results demonstrate that extreme ideological positions, higher political knowledge, and low-effort political participation predicted an increase in (relative) attitude-consistent exposure. Cross-social class exposure predicted a decrease in (relative) attitude-consistent exposure. The findings challenge existing arguments that SNS may per se facilitate attitude-consistent exposure.

Bakkeli, Nan Zou

doi: 10.1177/08944393211069622pmid: N/A

Background and purpose:The COVID-19 pandemic has posed considerable challenges to people’s mental health, and the prevalence of anxiety and depression increased substantially during the pandemic. Early detection of potential depression is crucial for timely preventive interventions; therefore, there is a need for depression prediction.Data and methods:This study was based on survey data collected from 5001 Norwegians (3001 in 2020 and 2000 in 2021). Machine learning models were used to predict depression risk and to select models with the best performance for each pandemic phase. Probability thresholds were chosen based on cost-sensitive analysis, and measures such as accuracy (ACC) and the area under the receiver operating curve (AUC) were used to evaluate the models’ performance.Results:The study found that decision tree models and regularised regressions had the best performance in both 2020 and 2021. For the 2020 predictions, the highest accuracies were obtained using gradient boosting machines (ACC = 0.72, AUC = 0.74) and random forest algorithm (ACC = 0.71, AUC = 0.75). For the 2021 predictions, the random forest (ACC = 0.76, AUC = 0.78) and elastic net regularisation (ACC = 0.76, AUC = 0.78) exhibited the best performances. Highly ranked predictors of depression that remained stable over time were self-perceived exposure risks, income, compliance with nonpharmaceutical interventions, frequency of being outdoors, contact with family and friends and work–life conflict. While epidemiological factors (having COVID symptoms or having close contact with the infected) influenced the level of psychological distress to a larger extent in the relatively early stage of pandemic, the importance of socioeconomic factors (gender, age, household type and employment status) increased substantially in the later stage.Conclusion:Machine learning models consisting of demographic, socioeconomic, behavioural and epidemiological features can be used for fast ‘first-hand’ screening to diagnose mental health problems. The models may be helpful for stakeholders and healthcare providers to provide early diagnosis and intervention, as well as to provide insight into forecasting which social groups are more vulnerable to mental illness in which social settings.

Li, Xintong; Martens, Matthew P.; Wiedermann, Wolfgang

doi: 10.1177/08944393211073168pmid: N/A

Conditional Direction Dependence Analysis (CDDA) has recently been proposed as a statistical framework to test reverse causation (x → y vs. y → x) and potential of confounding (x ← c → y) of variable relations in linear models when moderation is present. Similar to standard DDA, CDDA assumes that the “true” predictor is a continuous, non-normal, exogenous variable. Under non-normality, a conditional causal effect of one variable does not only change means, variances, and covariances, but also the distributional shape (i.e., skewness, kurtosis, co-skewness, and co-kurtosis) of another variable given the moderator. Such distributional changes can be used to study underlying mechanisms of heterogenous causal effects. The present study introduces conditional direction of dependence modeling and presents SPSS macros to make CDDA easily accessible to applied researchers. A real-world data example from the field of gambling addiction research is used to introduce the functionality of CDDA SPSS macros. Limitations of CDDA due to violated assumptions and poor data quality are discussed. The CDDA installation package is available at no charge from www.ddaproject.com.

Geng, Jingyu; Wang, Jue; Wang, Yuhui; Wang, Xingchao; Lei, Li; Wang, Pengcheng

doi: 10.1177/08944393221074602pmid: N/A

Non-suicidal self-injury has been increasingly acknowledged as a health problem. However, the association of cyberbullying victimization with non-suicidal self-injury and the mechanisms connecting this link have not been sufficiently studied. Therefore, the current study aimed to test the link between cyberbullying victimization and non-suicidal self-injury and to explore the mediating roles of basic psychological needs satisfaction and the moderating role of self-compassion in this relation. A sample of 1007 senior high students participated in the current study and completed multiple questionnaires on cyberbullying victimization, basic psychological needs satisfaction, non-suicidal self-injury, and self-compassion. The bivariate correlations between variables, mediation model, and moderated mediation model were tested by correlation analysis and Model 4 and Model 8 of the PROCESS macro, respectively. The results indicated that cyberbullying victimization was significantly and positively associated with non-suicidal self-injury. Autonomy needs satisfaction mediated the relationship between cyberbullying victimization and non-suicidal self-injury. The direct effect of cyberbullying victimization on non-suicidal self-injury was alleviated by strong self-compassion. Besides, self-compassion strengthened the direct relation between cyberbullying victimization and basic psychological needs and further strengthened the indirect association of cyberbullying victimization with non-suicidal self-injury. Our study highlights the potential mechanisms underlining the relationship between cyberbullying victimization and non-suicidal self-injury, and it has important theoretical and practical implications for adolescent non-suicidal self-injury.

Zhao, Xinyan

doi: 10.1177/08944393211073746pmid: N/A

Social media influentials play key roles in information creation and impact people’s crisis interpretations and behavioral intentions. Applying a multilevel perspective, this study examined how various social media influentials’ frame building was affected by crisis clusters and message characteristics. Social media influentials’ tweets were extracted from over a million tweets from eight crises. A random sample of 2,000 tweets was analyzed by content analysis. The results showed that social media influentials’ use of responsibility and topic frames was affected by both crisis-level factors (i.e., crisis origin and organization type) and message-level factors (i.e., communicative functions and types of influentials). These findings support the importance to understand the contextual factors that condition influentials' frame building on social media.