Search

To extend the limited evidence on how hot spot policing (HSP) strategies affect community experiences, perceptions, and attitudes, police agencies in two cities participated in a randomized experiment involving 102 hot spots that were assigned to a control condition (n = 51) or to receive a HSP program emphasizing patrol, community engagement, and problem-solving for 14–17 months during 2019 and 2020 (n = 51). Cross-sectional surveys with hot spot community members were conducted in person before the program (n = 1082) and, due to COVID-19, by mail and internet afterwards (n = 768) to assess program effects on crime victimization, views of crime and disorder, and attitudes towards police. In both cities, the evaluation period overlapped with the COVID-19 pandemic and the national protests for police reform following the killing of George Floyd by police in Minneapolis during 2020; in one city, the program was ongoing during these events. Results showed the program had few effects on community experiences and views, though there were some indications it may have improved perceptions of police legitimacy and police misconduct in one city. The findings suggest that HSP strategies do not have harmful effects on community perceptions and might improve some aspects of police-community relations. However, weak program implementation, challenges to survey administration, and the occurrence of both COVID-19 and the George Floyd incident during the study period complicate interpretation of the results. [ABSTRACT FROM AUTHOR]

We propose a model of treatment interference where the response of a unit depends only on its treatment status and the statuses of units within its K-neighborhood. Current methods for detecting interference include carefully designed randomized experiments and conditional randomization tests on a set of focal units. We give guidance on how to choose focal units under this model of interference. We then conduct a simulation study to evaluate the efficacy of existing methods for detecting network interference. We show that this choice of focal units leads to powerful tests of treatment interference that outperform current experimental methods.

Firms adopt randomized experiments to evaluate various interventions (e.g., website design, creative content, and pricing). However, most randomized experiments are designed to identify the impact of one specific intervention. The literature on randomized experiments lacks a holistic approach to optimize a sequence of interventions along the customer journey. Specifically, locally optimal interventions unveiled by randomized experiments might be globally suboptimal when considering their interdependence as well as the long-term rewards. Fortunately, the accumulation of a large number of historical experiments creates exogenous interventions at different stages along the customer journey and provides a new opportunity. This study integrates multiple experiments within the reinforcement learning (RL) framework to tackle the questions that cannot be answered by stand-alone randomized experiments. How can we learn optimal policy with a sequence of interventions along the customer journey based on an ensemble of historical experiments? Additionally, how can we learn from multiple historical experiments to guide future intervention trials? We propose a Bayesian recurrent Q-network model that leverages the exogenous interventions from multiple experiments to learn their effectiveness at different stages of the customer journey and optimize them for long-term rewards. Beyond optimization within the existing interventions, the Bayesian model also estimates the distribution of rewards, which can guide subject allocation in the design of future experiments to optimally balance exploration and exploitation. In summary, the proposed model creates a two-way complementarity between RL and randomized experiments, and thus, it provides a holistic approach to learning and optimizing interventions along the customer journey. This paper was accepted by Anindya Ghose, information systems. Funding: This work was supported by Adobe Faculty Research Award and the Marketing Science Institute Research Grant. Supplemental Material: The data files and online appendix are available at https://doi.org/10.1287/mnsc.2023.4914. [ABSTRACT FROM AUTHOR]

This paper examines the impacts of information about COVID-19 on pig farmers’ production willingness by using endorsement experiments and follow-up surveys conducted in 2020 and 2021 in China. Our results show that, first, farmers were less willing to scale up production when they received information about COVID-19. The information in 2020 that the second wave of COVID-19 might occur without a vaccine reduced farmers’ willingness to scale up by 13.4%, while the information in 2021 that COVID-19 might continue to spread despite the introduction of vaccine reduced farmers’ willingness by 4.4%. Second, farmers whose production was affected by COVID-19 were considerably less willing to scale up, given the access to COVID-19 information. Third, farmers’ production willingness can predict their actual production behavior.

Substantial funding is provided to the healthcare systems of low-income countries. However, an important challenge is to ensure that this funding is used efficiently. This challenge is complicated by the fact that a large share of healthcare services in low-income countries is provided by nonprofit health centers that often lack (i) effective governance structures and (ii) organizational know-how and adequate training. In this paper, we argue that the bundling of performance-based incentives with auditing and feedback (A&F) is a potential way to overcome these obstacles. First, the combination of feedback and performance-based incentives—that is, feedback joined with incentives to act on this feedback and achieve specific health outcomes—helps address the knowledge gap that may otherwise undermine performance-based incentives. Second, coupling feedback with auditing helps ensure that the information underlying the feedback is reliable—a prerequisite for effective feedback. To examine the effectiveness of this bundle, we use data from a randomized governance program conducted in the Democratic Republic of Congo. Within the program, a set of health centers was randomly assigned to a "governance treatment" that consisted of performance-based incentives combined with A&F, whereas others were not. Consistent with our prediction, we find that the governance treatment led to (i) higher operating efficiency and (ii) improvements in health outcomes. Furthermore, we find that funding is not a substitute for the governance treatment; health centers that only receive funding increase their scale but do not show improvements in operating efficiency or health outcomes. This paper was accepted by Lamar Pierce, organizations. Funding: This research was supported by the Agence Nationale de la Recherche [Grant Investissements d'Avenir (LabEx Ecodec)]. Supplemental Material: The data files and online appendix are available at https://doi.org/10.1287/mnsc.2023.4846. [ABSTRACT FROM AUTHOR]

This paper presents results from an experimental evaluation of an intervention designed to enhance virtual student support. During the 2019–2020 school year, randomly selected mentors in a school-based mentoring program received monthly reminders with tips for communicating with youth via text, email, and phone. Unexpectedly, the results showed that although the nudges did not impact the frequency of mentor outreach (student-reported), they reduced the rate at which students reached out (mentor-reported) and saw themselves as responsive to their mentors. Moreover, and possibly as a consequence, mentors who received the intervention felt less connected to students and less satisfied with their mentoring relationships, and treated students gained less than comparison students from the mentoring program as a whole in terms of their personal and attitudinal growth. The findings add important nuance to the evidence on how behavioral interventions in educational contexts operate. Although past studies find that nudges can support engagement in discrete tasks, these findings suggest that prescribing relational practices may be less effective. Thus, mentor supports must be carefully designed in order to yield the intended benefits for students. [ABSTRACT FROM AUTHOR]

Experimental designs have been recognized as the gold standard for establishing causal mechanisms. However, the application of such designs is complicated by factors such as excessive costs, time consumption, ethical concerns, and political impossibility. Nevertheless, the Chinese government's replacement housing efforts provide a unique randomized experiment for exploring the causal effects of the built environment on travel behavior. Accordingly, based on a large-scale survey on travel patterns under an experimental design in Shanghai, this study employs a two-step modeling approach, involving logit and Tobit models, to identify the built environment's effects on auto ownership and vehicle kilometers traveled (VKT). We found that transit service improvements play a stronger role in reducing auto-drive than compact and diverse land-use characteristics. Increasing residential and employment density, as well as land-use mix, discourages car ownership, which in turn reduces VKT, but with lower elasticities than transportation system variables. The findings provide additional evidence and referential estimate for how land-use and transport strategies and policies designed to create a compact, mixed-use, and highly accessible built environment can be used in reducing auto driving. This study expands the VKT reduction elasticities' database regarding the built environment across global spatial contexts, serving as a model for similar studies elsewhere in the world. [ABSTRACT FROM AUTHOR]

This study uses a randomized controlled trial to evaluate a new program for increased labor market integration of refugees. The new program starts shortly after the residence permit is granted and uses three main components: early and intensive language training, work practice with supervisors, and job search assistance performed by professional caseworkers. The immediate and intensive assistance contrasts previous integration policies which typically constitute low-intensive help over long periods of time. We find large positive effects on employment of the program, with magnitudes corresponding to around 15 percentage points. A mediation analysis shows that 7–8 percent of the impact of the program seems to be explained by increases in documented language skills. Cost estimates suggest that the new policy is less expensive than earlier labor market programs. • The study uses an RCT to evaluate a labor market program for low-educated refugees. • The new, intensive, program starts shortly after a residence permit is granted. • Two program components are intensive language training and supervised work practice. • We find large positive effects on employment of the program. • Around 7 percent of the effect seems to be explained by increases in documented language skills. [ABSTRACT FROM AUTHOR]

For papers to have scientific impact, they need to impress our peers in their role as referees, journal editors, and members of conference committees. Does better writing help our papers make it past these gatekeepers? In this study, we estimate the effect of writing quality by comparing how 30 economists judge the quality of papers written by PhD students in economics. Each economist judged five papers in their original version and five different papers that had been language edited. No economist saw both versions of the same paper. Our results show that writing matters. Compared to the original versions, economists judge edited versions as higher quality; they are more likely to accept edited versions for a conference; and they believe that edited versions have a better chance of being accepted at a good journal. [ABSTRACT FROM AUTHOR]

This experiment tested the use of a nudge-based texting intervention on the moral reasoning development of preservice teachers. The intervention sent cellphone text messages of moral scenarios in one of two forms over the course of 42 days (6 weeks). Results showed no significant interaction between group assignment and time across all four indices of moral reasoning development. Findings present initial evidence that nudge-based text messages do not affect moral reasoning development of preservice teachers. Discussion reviews implications with design and use of a nudge-based approach to support moral reasoning development among teachers. [ABSTRACT FROM AUTHOR]

Recommender systems have been introduced to help consumers navigate large sets of alternatives. They usually lead to more sales, which may increase consumer surplus and firm profit. In this paper, we ask whether the firms' choice of recommender system might hurt consumers. We use data from a large-scale field experiment in video-on-demand to measure the price elasticity of demand for movies placed in salient and nonsalient slots on the TV screen. During this experiment, the firm randomized the prices and slots in which movies were recommended to consumers. This setting readily allows for identifying the effects of price and slot on demand, and thus computes consumer surplus. We find that consumers are less price-elastic toward movies placed in salient slots. Using the outcomes of this experiment, we simulate how consumer surplus and welfare change when the firm implements several types of recommender systems, namely one that maximizes profit. We show that this system hurts consumer surplus and welfare relative to a system that maximizes welfare. We also show that, in our setting, the system that maximizes profit does not generate less consumer surplus than several recommender systems frequently used in practice. Yet, the amount of extra rent the firm can extract from strategically placing movies in salient slots is still a function of the popularity and quality of movies used to do so. Ultimately, our results question whether recommender systems embed mechanisms that extract excessive surplus from consumers, which may call for better scrutiny. [ABSTRACT FROM AUTHOR]

The scholarly information-seeking process for behavioral research consists of three phases: searching, accessing, and processing of past research. Existing IT artifacts, such as Google Scholar, have in part addressed the searching and accessing phases, but fall short of facilitating the processing phase, creating a knowledge inaccessibility problem. We propose a behavioral ontology learning from text (BOLT) design framework that presents concrete prescriptions for developing systems capable of supporting researchers during their processing of behavioral knowledge. Based upon BOLT, we developed a search engine— TheoryOn—to allow researchers to directly search for constructs, construct relationships, antecedents, and consequents, and to easily integrate related theories. Our design framework and search engine were rigorously evaluated through a series of data mining experiments, a randomized user experiment, and an applicability check. The data mining experiment results lent credence to the design principles prescribed by BOLT. The randomized experiment compared TheoryOn with EBSCOhost and Google Scholar across four information retrieval tasks, illustrating TheoryOn’s ability to reduce false positives and false negatives during the information-seeking process. Furthermore, an in-depth applicability check with IS scholars offered qualitative support for the efficacy of an ontology-based search and the usefulness of TheoryOn during the processing phase of existing research. The evaluation results collectively underscore the significance of our proposed design artifacts for addressing the knowledge inaccessibility problem for behavioral research literature. [ABSTRACT FROM AUTHOR]

We approach attrition in feld experiments with baseline data as an identifcation problem in a panel model. A systematic review of the literature indicates that there is no consensus on how to test for attrition bias. We establish identifying assumptions for treatment effects for both the respondent subpopulation and the study population. We propose randomization and regression-based procedures to test their sharp implications. We then relate our proposed tests to current empirical practice, and demonstrate that the most commonly used test can over-reject internal validity. Simulations and applications further support the relevance of our analysis.

Many papers in the literature have analyzed the role search rank plays in influencing user choices for the same product offered through different links. However, the literature has not analyzed whether search position can cause users to change their consumption between two ex ante distinct product categories. We analyze this question in the context of the ongoing public policy debate surrounding the role search engines can play in anti-piracy efforts. Specifically, we analyze whether reducing the prominence of infringing links can impact users choices between legal and infringing content. To do this we design a customized search engine that allows us to manipulate the positions of infringing and legal links in users’ search results. We then use this search engine to conduct experiments on a general population of users and on a subset of college-aged users. Our data show that reducing the prominence of infringing links in search results causes users who otherwise would have consumed infringing content to switch their consumption to paid legal content, and that these results hold even among users whose initial search queries express an explicit preference for infringing content. These results suggest that even small changes in the cost of discovering pirated content can have a relatively large impact on user behavior. As such, our results inform an important public policy debate by showing that search engines have a vital role to play in the fight against online copyright theft. [ABSTRACT FROM AUTHOR]

Patients with mental illness are less likely to receive the same physical healthcare as those without mental illness and are less likely to be treated in accordance with established guidelines. This study employed a randomized experiment to investigate the influence of comorbid depression on diagnostic accuracy. Physicians were presented with an interactive vignette describing a patient with a complex presentation of pernicious anemia. They were randomized to diagnose either a patient with or without (control) comorbid depression and related behaviors. All other clinical information was identical. Physicians recorded a differential diagnosis, ordered tests, and rated patient likeability. Fifty-nine physicians completed the study. The patient with comorbid depression was less likeable than the control patient (p=0.03, 95 % CI [0.09, 1.53]). Diagnostic accuracy was lower in the depression compared to control condition (59.4 % vs. 40.7 %), however this difference was not statistically significant χ2(1)=2.035, p=0.15. Exploratory analyses revealed that patient condition (depression vs. control) interacted with the number of diagnostic tests ordered to predict diagnostic accuracy (OR=2.401, p=0.038). Accuracy was lower in the depression condition (vs. control) when physicians ordered fewer tests (1 SD below mean; OR=0.103, p=0.028), but there was no difference for physicians who ordered more tests (1 SD above mean; OR=2.042, p=0.396). Comorbid depression and related behaviors lowered diagnostic accuracy when physicians ordered fewer tests – a time when more possibilities should have been considered. These findings underscore the critical need to develop interventions to reduce diagnostic error when treating vulnerable populations such as those with depression. [ABSTRACT FROM AUTHOR]

Much concern has been raised about the power of political microtargeting to sway voters' opinions, influence elections, and undermine democracy. Yet little research has directly estimated the persuasive advantage of microtargeting over alternative campaign strategies. Here, we do so using two studies focused on U.S. policy issue advertising. To implement a microtargeting strategy, we combined machine learning with message pretesting to determine which advertisements to show to which individuals to maximize persuasive impact. Using survey experiments, we then compared the performance of this microtargeting strategy against two other messaging strategies. Overall, we estimate that our microtargeting strategy outperformed these strategies by an average of 70% or more in a context where all of the messages aimed to influence the same policy attitude (Study 1). Notably, however, we found no evidence that targeting messages by more than one covariate yielded additional persuasive gains, and the performance advantage of microtargeting was primarily visible for one of the two policy issues under study. Moreover, when microtargeting was used instead to identify which policy attitudes to target with messaging (Study 2), its advantage was more limited. Taken together, these results suggest that the use of microtargeting--combining message pretesting with machine learning--can potentially increase campaigns' persuasive influence and may not require the collection of vast amounts of personal data to uncover complex interactions between audience characteristics and political messaging. However, the extent to which this approach confers a persuasive advantage over alternative strategies likely depends heavily on context. [ABSTRACT FROM AUTHOR]

Objectives: Outfitting police officers with body-worn cameras (BWCs) has been suggested to improve police-community relations. This study evaluates whether the deployment of BWCs on NYPD officers impacted resident perceptions of the police. Methods: A cluster randomized controlled trial design was used to test the influence of BWCs on resident perceptions of the NYPD in treatment precincts relative to control precincts. Dual-frame randomly selected telephone surveys were used to collect pre-intervention and post-intervention resident perception data. Results: We find no statistically significant differences between BWC treatment and control precincts in general perceptions of the NYPD or the average assessment of police officer behavior among those who have had recent encounters with the NYPD. Conclusion: Strong community support for BWC adoption and citizen expectations for videos of controversial policing events suggests the continued use of this technology. However, BWCs should be implemented with other evidence-based programs to enhance police-community relations. [ABSTRACT FROM AUTHOR]

Many instructors of political science wrestle with the question of whether to reveal their political ideology to their students and recent polarization in the United States intensifies those concerns. Prior research suggests that liberal and moderate students are wary of taking a course with a conservative professor, but do students react to economically conservative professors the same as they do socially conservative professors? We conducted an online survey experiment of current students at two public universities in the southern United States to test whether a reputation for expressing conservative opinions on either economic or social issues affected a professor's appeal to students. Participants split along ideological lines on both professor profiles, but greater skepticism was directed at the socially conservative professor. Preference for a socially conservative professor was greatest among more religious students. [ABSTRACT FROM AUTHOR]

In this research, we study an effective method to encourage users to generate stronger passwords. Specifically, we propose a novel design of password strength meters that incorporates contextual information to help users digest the message generated by the password strength meter. We evaluate our design by leveraging three independent and complementary methods: a survey-based experiment using students to evaluate the saliency of our conceptual design (proof of concept), a controlled laboratory experiment conducted on Amazon Mechanical Turk to test the effectiveness of the proposed design (proof of value), and a randomized field experiment conducted in collaboration with an online forum in Asia to establish proof of use. In each study, we observe that users exposed to the proposed password strength meter are more likely to change their passwords, leading to a new password that is significantly stronger. Our findings suggest that the proposed design of augmented password strength meters is an effective method for promoting secure password behavior among end users. Our design also requires minimal computational resources and technical capabilities. Password-based authentication is the most commonly used method for gaining access to secured systems. Unfortunately, empirical evidence highlights the fact that most passwords are significantly weak, and encouraging users to create stronger passwords is a significant challenge. In this research, we propose a theoretically augmented password strength meter design that is guided by the elaboration likelihood model of persuasion (ELM). We evaluate our design by leveraging three independent and complementary methods: a survey-based experiment using students to evaluate the saliency of our conceptual design (proof of concept), a controlled laboratory experiment conducted on Amazon Mechanical Turk to test the effectiveness of the proposed design (proof of value), and a randomized field experiment conducted in collaboration with an online forum in Asia to establish proof of use. In each study, we observe the changes in users' behavior in response to our proposed password strength meter. We find that the ELM-augmented password strength meter is significantly effective at addressing the challenges of password-based authentication. Users exposed to this strength meter are more likely to change their passwords, leading to a new password that is significantly stronger. Our findings suggest that the proposed design of augmented password strength meters is an effective method for promoting secure password behavior among end users. History: Alessandro Acquisti, Senior Editor; Pallab Sanyal, Associate Editor. Funding: The authors gratefully acknowledge financial support from the Social Science and Humanities Research Council of Canada (SSHRC) [Grant 435-2018-0605]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/isre.2022.1125. [ABSTRACT FROM AUTHOR]

We approach attrition in field experiments with baseline data as an identification problem in a panel model. A systematic review of the literature indicates that there is no consensus on how to test for attrition bias. We establish identifying assumptions for treatment effects for both the respondent subpopulation and the study population, and propose procedures to test their sharp implications. We then relate our proposed tests to current empirical practice, and demonstrate that the most commonly used test in the literature is not a test of internal validity in general. We illustrate the relevance of our analysis using several empirical applications.

Public health organizations increasingly use social media advertising campaigns in pursuit of public health goals. In this paper, we evaluate the impact of about $40 million of social media advertisements that were run and experimentally tested on Facebook and Instagram, aimed at increasing COVID-19 vaccination rates in the first year of the vaccine roll-out. The 819 randomized experiments in our sample were run by 174 different public health organizations and collectively reached 2.1 billion individuals in 15 languages. We find that these campaigns are, on average, effective at influencing self-reported beliefs—shifting opinions close to 1% at baseline with a cost per influenced person of about $3.41. Combining this result with an estimate of the relationship between survey outcomes and vaccination rates derived from observational data yields an estimated cost per additional vaccination of about $5.68. There is further evidence that campaigns are especially effective at influencing users’ knowledge of how to get vaccines. Our results represent, to the best of our knowledge, the largest set of online public health interventions analyzed to date. [ABSTRACT FROM AUTHOR]

Dynamic pricing programs have the potential to better align the price and social marginal costs of electricity production in the presence of time-varying external impacts on environmental quality. Using a particular dynamic pricing implementation, we show that the resulting impact on short-term operating costs and emissions is at best ambiguous. Moreover, we highlight the difficulty in quantifying impacts across geographic regions even in ideal scenarios where data is plentiful and the behavioral response can be estimated precisely using a randomized control trial of customers of an electric utility. Our findings point to price-induced reallocation of electricity consumption within a day as the underlying causal mechanism that may increase net emissions depending on the source-generation mix of a region. [ABSTRACT FROM AUTHOR]

Causal pluralism can be defended not only in respect to causal concepts and methodological guidelines, but also at the finer-grained level of causal inference from a particular source of evidence for causation. An argument for this last variety of pluralism is made based on an analysis of causal inference from randomized experiments (RCTs). Here, the causal interpretation of a statistically significant association can be established via multiple paths of reasoning, each relying on different assumptions and providing distinct elements of information in favour of a causal interpretation. [ABSTRACT FROM AUTHOR]

To improve covariate balance over a complete randomization, a number of methods have been proposed recently to utilize modern computational capabilities to find allocations with balance in observed covariates. Asymptotic inference on treatment effects based on these designs is more complicated than that under complete randomization, and this is why Fisher randomization tests often are suggested. This article suggests model‐based Bayesian inference as a general method of inference in these designs, which can deal with complications such as arbitrary covariate balancing criteria and complex estimands. As an illustration, we focus on the case when the outcome is linearly related to the covariates and the estimand of interest is the Sample Average Treatment Effect (SATE). We use a large Monte Carlo simulation to compare the finite sample performance of the model‐based Bayesian inference with that of two previous methods which are valid for asymptotic inference of SATE under Mahalanobis distance based rerandomization. We find that for experiments with small to moderate sample sizes, Bayesian inference is to be preferred to the previous methods. As a byproduct, we also find that regression adjustment together with small sample adjusted estimators of standard errors perform better than the previous methods. [ABSTRACT FROM AUTHOR]

We ran a field experiment to investigate whether individual performance in teams was affected by the gender of the leader. About 430 students from an Italian University took an intermediate exam that was partly evaluated on the basis of teamwork. Students were randomly matched in teams of three and, in each team, we randomly chose a leader entrusted with the task of coordinating the work of the team. We find a positive and significant effect of female leadership on team performance. This effect is driven by the higher performance of team members in female-led teams rather than by an improvement in leader performance. In spite of the higher performance of female-led teams, male members tended to evaluate female leaders as less effective, whereas female members have provided more favorable judgments. Our results are consistent with stereotypically feminine personality traits influencing leadership style and the decision on the amount of effort to exert in a task where females are contravening stereotypical behavior and their traditional societal role. [ABSTRACT FROM AUTHOR]

One of the problems that can affect the value that individuals attribute to the environment is related to the quality and level of information that agents have about the environmental resource. The purpose of this article is to test the hypothesis that persuasive communication affects individuals' perceptions of the environment. To test the hypothesis, we applied randomized questionnaires to students at the Federal University of Pelotas. The results of the random experiment show that groups formed by young, inexperienced and uninformed are more influenced by environmental rhetoric. Furthermore, the results show that individual perception can be intentionally altered by biased information.

Analogous to side effects in medical research, spillover and halo effects refer to unintended effects of interventions. Tourism and research in related areas have only begun to acknowledge these exist. Future research should quantify their magnitude to enable cost–benefit analysis, for example of whether an intervention to reduce carbon footprint does so when the total effect of interventions is included. Utilizing data from an experiment aimed at changing tourists' food consumption behaviour in hotels, this paper provides a methodology to quantify spillover and halo effects, including statistical inference to test hypotheses and produce confidence intervals. [ABSTRACT FROM AUTHOR]

Attitudes toward violence are important in theoretical explanations of violent behavior and efforts to reduce violent behavior. Though an association between attitudes and violent behavior has been demonstrated, most studies have used correlational/observational research designs. We conducted a randomized experiment to test the effect of attitudes toward violence on violent behavior with 285 men from the community. Participants were randomly assigned to receive material to make attitudes toward violence more negative or to a control condition. Violent behavior was then approximated by asking participants to select from a range of violent and nonviolent options in response to a series of interpersonal conflict vignettes. Participants in the negative attitude condition responded with less violence on the vignette questionnaire than did participants in the control condition (Cohen's d = −0.23, 95% CI [−0.46, 0.01]). Participants also completed a measure of attitudes toward violence at the end of the experiment; more positive attitudes toward violence showed a strong association with more violent responding on the vignette questionnaire (r =.62, 95% bootstrapped CI [.54,.69]). Consistent with theory and practice, our findings suggest that attitudes toward violence may play a role in violent behavior. [ABSTRACT FROM AUTHOR]

Objective: Superiority experiments supply a critical lens to probe higher education interventions because they can dismantle the collective effect of a program and identify the relative contribution of the core components. However, such lines of inquiry are generally absent in higher education literature. We draw on experimental data to probe the dominant mechanisms of a simple but promising academic probation intervention. Methods: We used a 2 × 2 factorial experiment designed to unpack the effects of an academic probation intervention and identify the dominant mechanism. Within this framework, 500 academic probation students were assigned to one of four treatment conditions or a control condition. Treatment groups were assigned to receive notification of academic probation status by either certified or regular mail (factor A) that contained either a standard or embellished letter (factor B). Results: We found that students who received embellished warning letters significantly improved subsequent academic performance. The impact on performance was 0.35 grade points when notification sent by regular mail and amplified to 0.48 grade points when sent via certified mail. The routine notification communication message did not differ from the treatment as usual method regardless of mail delivery method. Conclusions: These findings suggest the availability of low-cost, scalable interventions are available and can be identified to positively impact at-risk students' academic achievement. [ABSTRACT FROM AUTHOR]

The timing of an investigation after an officer-involved shooting (OIS) is influenced by conflicting forces. The public demands expedited resolution, but police officers are provided several protections that can delay investigations of their actions. This study conducts a randomized experiment to determine the impact of question timing after an OIS on the accuracy of police officers' memory recall. Officers were randomly assigned to one of two groups. The treatment group completed a questionnaire after participating in a live-action, active shooter training scenario and again 2 days later, whereas the control group only completed the questionnaire 2 days later. Our findings suggest the timing of interviews after training did not influence officers' recall of the scenario. There is little empirical understanding of how police officers reconstruct OIS events; further interdisciplinary research can help clarify these cognitive processes. This research could strengthen a traditional pathway to provide accountability for officers through investigations. [ABSTRACT FROM AUTHOR]

Court-related fines and fees are widely levied on criminal defendants who are frequently poor and have little capacity to pay. Such financial obligations may produce a criminalization of poverty, where later court involvement results not from crime but from an inability to meet the financial burdens of the legal process. We test this hypothesis using a randomized controlled trial of court-related fee relief for misdemeanor defendants in Oklahoma County, Oklahoma. We find that relief from fees does not affect new criminal charges, convictions, or jail bookings after 12 months. However, control respondents were subject to debt collection efforts at significantly higher rates that involved new warrants, additional court debt, tax refund garnishment, and referral to a private debt collector. Despite significant efforts at debt collection among those in the control group, payments to the court totaled less than 5 percent of outstanding debt. The evidence indicates that court debt charged to indigent defendants neither caused nor deterred new crime, and the government obtained little financial benefit. Yet, fines and fees contributed to a criminalization of low-income defendants, placing them at risk of ongoing court involvement through new warrants and debt collection. [ABSTRACT FROM AUTHOR]

Crime prevention through environmental design (CPTED) is a non-punitive method for reducing crime through the design of the built environment. The relevance of CPTED strategies, however, is less clear in the context of computing environments. Building upon prior research indicating that computing environments may change computer users' behaviors, this study tests the effectiveness of CPTED-based approaches in mitigating system-trespassing events. Findings from this randomized controlled field trial demonstrate that specific CPTED strategies can mitigate hacking events by reducing the number of concurrent activities on the target computer, attenuating the number of commands typed in the attacked computer, and decreasing the likelihood of hackers returning to a previously hacked environment. Our findings suggest some novel and readily implemented strategies for reducing cybercrime. [ABSTRACT FROM AUTHOR]

Because company cars add to corporate CO 2 footprints, companies are beginning to replace cars with mobility budgets that employees can use for leisure and commuting trips. This study examines whether nudges can encourage sustainable travel in such a subsidized setting. We conduct a field experiment with 341 employees of a large German company. Observing expenditure items charged to the mobility budget, we test if social comparisons and a climate-related moral appeal induce a shift towards low-emissions transport modes. We find that simultaneous application of both nudges causes a reduction in car use, particularly taxi and ride sharing, as well as substitution towards micromobility, but not public transport. The social comparison alone is not effective, and the treatment effects of the combined nudge vanish in the second half of the treatment period. Survey evidence suggests that these results are driven by a minority that complies with the communicated social norm. [ABSTRACT FROM AUTHOR]

This study analyses the effects of group differentiation by students' learning strategies of around 1200 students in 46 classes from eight secondary schools in the Netherlands. In an experimental setup with randomization at the class level, division of students over three groups per class (an instruction-independent group, an average group, and an instruction-dependent group) is based on learning strategies, measures using the Motivated Strategies for Learning Questionnaire (MSLQ). Each group is offered instruction fitting their own learning strategy. The results show that student performance is higher in classes where the differentiation was applied, and that these students score higher at some scales of the posttest of the questionnaire on motivation, metacognition and self-regulation. However, there are differences between classrooms from different teachers. Additional teacher questionnaires confirm the discrepancy in teacher attitudes towards the intervention. [ABSTRACT FROM AUTHOR]

We provide a principled way for investigators to analyze randomized experiments when the number of covariates is large. Investigators often use linear multivariate regression to analyze randomized experiments instead of simply reporting the difference of means between treatment and control groups. Their aim is to reduce the variance of the estimated treatment effect by adjusting for covariates. If there are a large number of covariates relative to the number of observations, regression may perform poorly because of overfitting. In such cases, the least absolute shrinkage and selection operator (Lasso) may be helpful. We study the resulting Lasso-based treatment effect estimator under the Neyman-Rubin model of randomized experiments. We present theoretical conditions that guarantee that the estimator is more efficient than the simple difference-of-means estimator, and we provide a conservative estimator of the asymptotic variance, which can yield tighter confidence intervals than the difference-of-means estimator. Simulation and data examples show that Lasso-based adjustment can be advantageous even when the number of covariates is less than the number of observations. Specifically, a variant using Lasso for selection and ordinary least squares (OLS) for estimation performs particularly well, and it chooses a smoothing parameter based on combined performance of Lasso and OLS.

Despite the growing evidence of informational interventions on college and major choices, we know little about how such light-touch interventions affect the gender gap in STEM majors. Linking survey data to administrative records of Chinese college applicants, we conducted a large-scale randomized experiment to examine the STEM gender gap in the major preference beliefs, application behaviors, and admissions outcomes. We find that female students are less likely to prefer, apply to, and enroll in STEM majors, particularly Engineering majors. In a school-level cluster randomized controlled trial, we provided treated students with major-specific wage information. Students' major preferences are easily malleable that 39% of treated students updated their preferences after receiving the wage informational intervention. The wage informational intervention has no statistically significant impacts on female students' STEM-related major applications and admissions. In contrast, those male students in rural areas who likely lack such information are largely shifted into STEM majors as a result of the intervention. We provide supporting evidence of heterogeneous major preferences for extrinsic incentives: even among those students who are most likely to be affected by the wage information (prefer high paying majors and lack the wage information), female students are less responsive to the informational intervention. [ABSTRACT FROM AUTHOR]

In a randomized field experiment, we show that a low-cost intervention, weekly email reminders, increases the weekly exercising frequency of gym members by 13%, with the effect being larger for class training (19%) than for free training (11%). We find that nearly all types of individuals benefit from the reminders (e.g., students, nonstudents, women, men, new members, recurring members), and this effect persists in a three-month posttreatment period. Furthermore, the increase in class visits is partly driven by an increase in the number of bookings of gym classes and a slight decrease in the share of bookings that are canceled. Limited attention and habit formation can best explain these results. In contrast to the increase in gym attendance, we do not find any effect on the duration and renewal of membership contracts. [ABSTRACT FROM AUTHOR]

Big-Fish-Little-Pond (BFLP) theory posits that being surrounded by high-ability peers reduces students' self-perceived ability - i.e., their academic self-concept (ASC), which is known to regulate the effort students put into education. We add to this literature the finding that the school context alone, without referring to peers' ability, triggers social comparison and can decrease students' ASC. We hypothesize that attenuating the school context increases students' ASC. The paper shows two examples of how attenuating school context leads to an increase in ASC. In Study 1, we exploited COVID-19-induced home-based education where students learned online without their usual school context. We find that students' ASC in reading and writing increased outside the school context. In the randomized survey experiment of Study 2, we treated students by asking first for their ASC and subsequently their grades - thus, we attenuated the school context by ordering the questions. We asked control students first about their grades and subsequently their ASC. We found that treated students' ASC increased in reading (but not in writing) compared to the control students who first reported their grades and then their ASC. The results indicate that the school context alone renders the perception of ability relative to peers, which triggers an unintended negative effect on students' ASC. [ABSTRACT FROM AUTHOR]

We study attitudes, perceptions, and valuations of a convenience sample of Chilean employees from an electric distribution company who applied for a subsidized electric vehicle (EV) acquisition program. The subsidy was randomly assigned among the interested applicants. We use this data in an experiment to assess the impact that being a user, or a non-user of an EV has on the factors under study, using focus groups and a stated preference (SP) experiment. In the focus groups, users mentioned relatively more benefits and barriers of EVs, while non-users spontaneously stated that a limited charging network at the urban level could be an issue. The SP survey suggested that being a user did not trigger a change in environmental attitudes; on the other hand, it significantly boosted perceptions of maintenance costs and driving range. Additionally, discrete choice models estimated from the SP data suggested a possibly null willingness to pay for urban charging infrastructure, expressed as a percentage of current gas stations, among the user group. We discuss possible policy implications that can be inferred from this analysis, considering the size and source limitations of the available sample. [ABSTRACT FROM AUTHOR]

This paper analyzes whether moderate weight reduction improves subjective health perception in obese individuals. Besides simple regression models, in a simultaneous equation framework we use randomized monetary weight loss incentives as instrument for weight change, to address possible endogeneity bias. In contrast to related earlier work that also employed instrumental variables estimation, identification does not rely on long-term, between-individuals weight variation, but on short-term, within-individual weight variation. Yet, our result does not suggest that the simple regressions suffer from much endogeneity bias, since instrumental variables estimation yields similar—though far noisily estimated and statistically insignificant—estimates. In qualitative terms, our results do not contradict previous findings pointing to weight loss in obese individuals resulting in improved subjective health. Our results suggest that a reduction of body weight by one BMI unit is associated with an increase in the probability of reporting self-rated health to be 'satisfactory' or better by 3 to 4 percentage points. This finding may encourage obese individuals in their weight loss attempts, since they are likely to be immediately rewarded for their efforts by subjective health improvements. [ABSTRACT FROM AUTHOR]

Ed Latessa's career was devoted to using evidence-based interventions to improve the lives of offenders. This article honors Ed by contributing to this line of inquiry. The main aim is to review criminological experiments with offending outcomes and a follow-up period of at least 10 years. Twenty major experiments are summarized, with intervention targets ranging from mothers in pregnancy to adult offenders. Follow-up periods ranged from 10 to 72 years. Effects of interventions were generally desirable and often persisted, but sometimes were undesirable and increased or decreased over time. In addition to offending, many other outcomes were often measured at different ages, and subgroup analyses were useful in identifying categories of participants associated with more or less desirable effects. It is concluded that more long-term follow-ups of criminological experiments should be carried out, preferably including cost-benefit analyses and measuring the prevalence, frequency, and seriousness of offending in different age ranges. [ABSTRACT FROM AUTHOR]

Digital media affordances create interesting potential opportunities for educators to integrate these technologies into the classroom in an effort to achieve greater student success. While there are many reasons to suspect that using digital media in the political science classroom may yield positive effects, there are few experimental studies of these effects. In this study, I leverage a longitudinal experimental design to study the effectiveness of using Twitter in introductory American politics courses to increase political interest, news consumption, and student learning outcomes. The findings show that using Twitter in the classroom has no significant influence on political interest or news consumption when controlling for prior levels of these variables. However, there may be a small positive influence on student learning outcomes that should be further explored. [ABSTRACT FROM AUTHOR]