No abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Abstract: Many important decisions are made under stress and they often involve risky alternatives. There has been ample evidence that stress influences decision making in cognitive as well as in affective domains, but still very little is known about whether individual attitudes to risk change with exposure to acute stress. To directly evaluate the causal effect of stress on risk attitudes, we adopt an experimental approach in which we randomly expose participants to a psychosocial stressor in the form of a standard laboratory stress-induction procedure: the Trier Social Stress Test for Groups. Risk preferences are elicited using an incentive compatible task, which has been previously shown to predict risk-oriented behavior out of the laboratory. Using three different measures (salivary cortisol levels, heart rate and multidimensional mood questionnaire scores), we show that stress was successfully induced on the treatment group. Our main result is that acute psychosocial stress significantly increases risk aversion. The effect is mainly driven by males; men in our control group are less risk-averse than women, which is a standard result in the literature, but this difference almost disappears when under psychosocial stress. Terms of use: Documents in

A previous study explored the cognitive biases that underlie individual differences in supernatural beliefs using path models in samples of Canadian and Americans (Willard and Norenzayan, 2013). We replicated and extended these path models in new nationally representative samples from the Czech Republic and Slovakia (total N = 2022). As in the original model, we found that anthropomorphism was unrelated to belief in God, but was consistently related to paranormal beliefs. Living in a highly religious area was related to a lower tendency to anthropomorphize. We further examined this relationship and found that anthropomorphism is related to belief in God for non-religious participants only, and is inversely related to belief in God among religious Slovaks, but not religious Czechs. These findings suggest that religious beliefs and societal context can change the relationship between cognitive biases and supernatural beliefs.

HIV care integrated into primary health care (PHC) encourages reorganized service delivery but could increase workload. In 2012-2013, we surveyed 910 patients and caregivers at two time points after integration in four clinics in Free State, South Africa. Likert surveys measured quality of care (QoC) and satisfaction with staff (SwS). QoC scores were lower for females, those older than 56 years, those visiting clinics every 3 months, and child health participants. Regression estimates showed QoC scores higher for ages 36-45 versus 18-25 years, and lower for those attending clinics for more than 10 years versus 6-12 months. Overall, SwS scores were lower for child health attendees and higher for tuberculosis attendees compared to chronic disease care attendees. Research is needed to understand determinants of disparities in QoC and SwS, especially for child health, diabetes, and hypertension attendees, to ensure high-quality care experiences for all patients attending PHC clinics with integrated HIV care.

Because many key career events, such as examinations and interviews, involve competition and stress, gender differences in response to these factors could help to explain the labor market gender gap. In a laboratory experiment, we manipulate psychosocial stress using the Trier Social Stress Test and confirm that this is effective by measuring salivary cortisol level and heart rate. Subjects perform in a real-effort task under both tournament and piece-rate incentives, and we elicit willingness to compete. We find that women under heightened stress perform worse than women in the control group when compensated with tournament incentives, whereas there is no treatment difference under piece-rate incentives. For men, stress does not affect output under competition or under piece rate. The gender gap in willingness to compete is not affected by stress, but stress decreases competitiveness overall, which is related to performance for women. Our results could explain gender differences in performance under competition, with implications for hiring practices and incentive structures in firms. This paper was accepted by Axel Ockenfels, behavioral economics.