Random projections for Bayesian regression

This article deals with random projections applied as a data reduction technique for Bayesian regression analysis. We show sufficient conditions under which the entire d-dimensional distribution is approximately preserved under random projections by reducing the number of data points from n to \(k\in O({\text {poly}}(d/\varepsilon ))\) in the case \(n\gg d\). Under mild assumptions, we prove that evaluating a Gaussian likelihood function based on the projected data instead of the original data yields a \((1+O(\varepsilon ))\)-approximation in terms of the \(\ell _2\) Wasserstein distance. Our main result shows that the posterior distribution of Bayesian linear regression is approximated up to a small error depending on only an \(\varepsilon \)-fraction of its defining parameters. This holds when using arbitrary Gaussian priors or the degenerate case of uniform distributions over \(\mathbb {R}^d\) for \(\beta \). Our empirical evaluations involve different simulated settings of Bayesian linear regression. Our experiments underline that the proposed method is able to recover the regression model up to small error while considerably reducing the total running time.     

Andere Positionierungen?! Analytische und politische Perspektiven dekolonialistischer Sozial- und Kulturtheorie in und aus Lateinamerika

The contribution discusses fundamental problematics of decolonialist social and cultural theory. Using the examples of the approaches of five authors – Aníbal Quijano, Enrique Dussel, Catherine Walsh, Gloria Anzaldúa, and Néstor García Canclini –, the text explores how the historicial construction of alterity was countered theoretically. Finally the article makes the case that the reactions to alterity – understood as based on social rank and cultural codes – are analytically described more precisely and politically better manageable by the notion of positioning than by any other concept of the decolonial debate.     

Konfliktmoderation mit Gruppen

One professional form of conflict resolution known as ‘mediation’ has itself established globally as an accepted alternative to litigation within the past four decades—albeit most notably for dyadic conflict between two individuals. Conflicts involving a higher number of parties both within and between social groups in organizations or public communities are far more complex and much tougher to be defined than in dyadic settings. They require an expanded repertoire of intervention techniques and more advanced competencies of neutrals as—in addition to the resolution efforts—the conflicting parties’ group dynamics demands attention and participative facilitation. Moreover, this type of conflict is difficult to study. An action strategy for professional group-based conflict resolution is exemplified and contrasted to dyadic mediation of conflict. Based on the analysis, problems for future research are outlined.     

Do Simulated Gambling Activities Predict Gambling with Real Money During Adolescence? Empirical Findings from a Longitudinal Study

As technology has developed, the international gambling market has changed markedly in recent years. The supply of internet-based gambling opportunities has become ever more significant. At the same time, the introduction of new gambling opportunities always brings a demand for evidence-based scientific evaluation, with regard to the associated risks of addiction. Simulated internet gambling, which is the focus of this study, represents a relatively new product group located at the interface between gambling and computer gaming. Concerns have been raised in scientific literature, especially with regard to the adolescent age group, as to whether participation in simulated internet gambling directly promotes recruitment to the world of monetary gambling, as defined in the gateway hypothesis. The research design was based on a standardized, representative longitudinal survey (over a 1-year period) with a total of 1178 school pupils from Northern Germany (M = 13.6 years; 47.5% male). It must be borne in mind that 12% of the adolescents belonged to the subgroup of “onset gamblers” and first reported experience with monetary gambling at the second stage of surveying. Logistic regression analysis demonstrates that this migration process is fostered by (1) participation from home in simulated gambling on social networks and (2) significant exposure to advertising (relating to both simulated and monetary gambling). Within the subgroup of simulated internet gamblers, variables such as particular patterns of use (including breadth and depth of involvement with simulated internet gambling, certain motives for participation, and microtransactions) do not serve as significant predictors. Despite this, important needs for action for the purposes of prevention and research can be identified.     

Near-Misses and Stop Buttons in Slot Machine Play: An Investigation of How They Affect Players,and May Foster Erroneous Cognitions

In modern casinos, multiline slot machines are becoming increasingly popular compared to traditional, three-reel slot machines. A paucity of research has examined how the unique presentation of near-misses and the use of a stop button in multiline slot machines impact erroneous cognitions related to the perception of skill and agency during play. Our goal therefore was to determine the prevalence of erroneous cognitions pertaining to near-miss outcomes and the usage of a stop button and then to see whether the stop button affected players’ experiences of winning, losing and near-miss outcomes. We recruited 132 gamblers from a casino in Ontario. They played two versions of a slot machine simulator: one with a stop button and one without a stop button. We measured player’s arousal [skin conductance responses (SCRs), pressure on the spin-button), and behavioural responses (post-reinforcement pauses (PRPs)] to wins, losses and near-misses during play. We predicted more robust physiological SCRs and longer PRPs to wins in the stop button game. We also predicted that near-misses encountered in the stop button game would trigger greater levels of arousal and frustration in players, as indexed by larger SCRs, and greater force applied to the spin button to initiate the next spin. Erroneous cognitions pertaining to the stop button and near-misses respectively were assessed following play. Results showed that a small but meaningful percentage of players held erroneous cognitions about the stop button (13.6%) and near-misses (16%). Players depressed the spin button harder, and had larger SCRs for all outcomes when using the stop button. Players also paused longer for near-misses in the game involving the stop button. Our findings converge to suggest that the stop button encourages an erroneous perception of skill in some players, and consequentially impacts how such players perceive their outcomes in multiline slot machines.     

The social order of markets

In this article I develop a proposal for the theoretical vantage point of the sociology of markets, focusing on the problem of the social order of markets. The initial premise is that markets are highly demanding arenas of social interaction, which can only operate if three inevitable coordination problems are resolved. I define these coordination problems as the value problem, the problem of competition and the cooperation problem. I argue that these problems can only be resolved based on stable reciprocal expectations on the part of market actors, which have their basis in the socio-structural, institutional and cultural embedding of markets. The sociology of markets aims to investigate how market action is structured by these macrostructures and to examine their dynamic processes of change. While the focus of economic sociology has been primarily on the stability of markets and the reproduction of firms, the conceptualization developed here brings change and profit motives more forcefully into the analysis. It also differs from the focus of the new economic sociology on the supply side of markets, by emphasizing the role of demand for the order of markets, especially in the discussion of the problems of valuation and cooperation.

---

Backlash or a Positive Response?: Public Opinion of LGB Issues After Obergefell v. Hodges

ABSTRACT

Following Obergefell v. Hodges, same-sex marriage remains controversial and anti-LGBT state legislation has been passed, which raises questions about whether the Supreme Court’s ruling may have created a backlash. We use data from two waves of a general population survey of Nebraskans conducted before and after the decision to answer three questions. First, we test three theories of how the court decision influenced public opinion. We find that support for same-sex marriage was significantly higher following the ruling, suggesting that there was not a backlash to it. Second, we assess whether people perceive that the court accurately reflects the public’s opinion. We find that people who favor same-sex marriage are more likely to think that the ruling refects public opinion very well; those who oppose same-sex marriage are more likely to think that the ruling does not at all reflect public opinion. Third, we examine the association between discussing gay rights and support for same-sex marriage, finding that those who talk about LGB issues very often are more likely to favor same-sex marriage. We discuss the implications of these findings in relation to two of the themes of this special issue: the influence of marriage equality on Americans’ understandings of marriage and the impact of marriage equality on future LGBT activism.     

Cost-efficient observation of working postures from video recordings--more videos, more observers or more views per observer?

In ergonomics, assessing the working postures of an individual by observation is a very common practice. The present study investigated whether monetary resources devoted to an observational study should preferably be invested in collecting many video recordings of the work, or in having several observers estimate postures from available videos multiple times. On the basis of a data set of observed working postures among hairdressers, necessary information in terms of posture variability, observer variability, and costs for recording and observing videos was entered into equations providing the total cost of data collection and the precision (informative value) of the resulting estimates of two variables: percentages time with the arm elevated <15 degrees and >90 degrees. In all 160 data collection strategies, differing with respect to the number of video recordings and the number of repeated observations of each recording, were simulated and compared for cost and precision. For both posture variables, the most cost-efficient strategy for a given budget was to engage 4 observers to look at available video recordings, rather than to have one observer look at more recordings. Since the latter strategy is the more common in ergonomics practice, we recommend reconsidering standard practice in observational posture assessment.     

Reinforcing Small Wins and Frustrating Near-Misses: Further Investigation Into Scratch Card Gambling

Scratch card games are incredibly popular in the Canadian marketplace. However, only recently have researchers started to systematically analyze their structural characteristics and how these in turn affect the gambler. We present two studies designed to further understand the underlying physiological and psychological effects that scratch cards have on gamblers. We had gamblers (63 in Experiment 1, 68 in Experiment 2) play custom made scratch cards involving a small win, a regular loss and a near-miss—where they uncovered two out of the three symbols needed to win the top prize. Our predictions were that despite near-misses and losses being objectively equivalent (the gambler wins nothing) gamblers’ reactions to these outcomes would differ dramatically. During game play, skin conductance levels and heart rate were recorded, as well as how long gamblers paused between each game. Gamblers’ subjective reactions to the different outcomes were then assessed. In both studies, near-misses triggered higher levels of physiological arousal (skin conductance levels and heart rates) than losses. Gamblers paused significantly longer following small wins than other outcomes, and reported high arousal, positive affect and urge to gamble—a constellation of results consistent with their rewarding properties. Importantly near-miss outcomes were rated as highly arousing, negative in emotional tone, and the most frustrating of all three outcome types examined. In Experiment 2, when we measured subjective urge to gamble immediately after each outcome, urge to gamble was significantly higher following near-misses than regular losses. Thus, despite not rewarding the gambler with any monetary gain, these outcomes nevertheless triggered higher arousal and larger urges to gamble than regular losses, a finding that may explain in part, the allure of scratch cards as a gambling activity.     

Boundary dimensions of social innovation: negotiating conflicts and compatibilities when developing a national agenda

In order to add to the existing knowledge on the complex institutional dynamics of negotiating conflicts and compatibilities in innovation processes that address societal challenges and social needs, this article scrutinizes the scientific and societal debate on the theoretical and practical application of “social innovation” by means of the concept “boundary object” linked to institutional theory. A participatory case study of the development of an R&I agenda for social innovation in Sweden, formulated through dialog between stakeholders from the public, private and non-profit sectors, is presented. The results show that the unpredictable, emergent, evolving and adaptable character of needs, challenges and solutions in social innovation processes is made more theoretically and practically manageable by acknowledging social innovation as a boundary object. This is since this concept, despite its ambiguous character, helps balance complexity and usefulness in a way that serves to expand existing knowledge on institutional dynamics.