Three Recommendations for Improving the Use of p-Values

ABSTRACT

Researchers commonly use p-values to answer the question: How strongly does the evidence favor the alternative hypothesis relative to the null hypothesis? p-Values themselves do not directly answer this question and are often misinterpreted in ways that lead to overstating the evidence against the null hypothesis. Even in the “post p?<?0.05 era,” however, it is quite possible that p-values will continue to be widely reported and used to assess the strength of evidence (if for no other reason than the widespread availability and use of statistical software that routinely produces p-values and thereby implicitly advocates for their use). If so, the potential for misinterpretation will persist. In this article, we recommend three practices that would help researchers more accurately interpret p-values. Each of the three recommended practices involves interpreting p-values in light of their corresponding “Bayes factor bound,” which is the largest odds in favor of the alternative hypothesis relative to the null hypothesis that is consistent with the observed data. The Bayes factor bound generally indicates that a given p-value provides weaker evidence against the null hypothesis than typically assumed. We therefore believe that our recommendations can guard against some of the most harmful p-value misinterpretations. In research communities that are deeply attached to reliance on “p?<?0.05,” our recommendations will serve as initial steps away from this attachment. We emphasize that our recommendations are intended merely as initial, temporary steps and that many further steps will need to be taken to reach the ultimate destination: a holistic interpretation of statistical evidence that fully conforms to the principles laid out in the ASA statement on statistical significance and p-values.     

An extended semiparametric model for short-term and long-term hazard ratios

In survival analysis, one way to deal with non-proportional hazards is to model short-term and long-term hazard ratios. The existing model of this nature has no control over how fast the hazard ratio is changing over time. We add a parameter to the existing model to allow the hazard ratio to change over time at different speed. A nonparametric maximum likelihood approach is used to estimate the model parameters. The existing model is a special case of the extended model when the speed parameter is 0, which leads naturally to a way of testing the adequacy of the existing model. Simulation results show that there can be substantial bias in the estimation of the short-term and long-term hazard ratio if the speed parameter is fixed incorrectly at 0 rather than estimated. The extended model is fitted to three real data sets to shed new insights, including the observation that converging hazards does not necessarily imply the odds are proportional.     

On Hotelling’s T2 test in a special paired sample case

In a special paired sample case, Hotelling’s T² test based on the differences of the paired random vectors is the likelihood ratio test for testing the hypothesis that the paired random vectors have the same mean; with respect to a special group of affine linear transformations it is the uniformly most powerful invariant test for the general alternative of a difference in mean. We present an elementary straightforward proof of this result. The likelihood ratio test for testing the hypothesis that the covariance structure is of the assumed special form is derived and discussed. Applications to real data are given.     

Responsibility,planning and risk management: moralizing everyday finance through financial education

The individualization, privatization and marketization of risk management represent a fundamental dimension of the financialization of everyday life. As individuals are required to engage with financial products and services as the main way of protecting themselves from risks and uncertainties, their economic welfare and security are construed as depending largely on their own financial decisions. Within this setting, the concept of financial literacy and accompanying practices of financial education have emerged as a prominent institutional field handling the formulation and communication of the attributes and dispositions that arguably constitute the proper financial actor. This article analyzes financial education programmes currently conducted by state agencies in Israel, examining the notions and principles they articulate when defining and explaining proper financial conduct. The study indicates that moral themes and categories occupy a salient place in the formulation of the character traits that constitute the desired literate financial actor. Notions of individual responsibility, planning ahead and rational risk management are presented not merely as instrumental resources, but as moral imperatives. Through these notions, the programmes moralize a broad array of everyday practices of personal finance such as saving, investing, borrowing and budget management, thereby connecting the sphere of financial matters to the domain of moral virtues. Offering a representation of particular modes of financial conduct as constitutive components of morally virtuous personhood, these practices imbue the financial field as a whole, especially its current generalized logic of individualized and marketized risk management, with moral meanings, hence contributing to the normalization and depoliticization of the financialization of everyday life.     

Linking Migration Intentions with Flows: Evidence and Potential Use

In the absence of reliable, internationally available migration flow data necessary for statistical forecasting, policymakers increasingly turn to survey data on emigration intentions to evaluate future migration trends. The important assumption – i.e. that there is a measurable and systematic relationship between the intention to migrate and actual migration – has not been firmly established at the international level. We examine the association between estimated population averages of emigration intentions and official migration flow data based on data for more than 160 countries. The results show a strong association between emigration intentions and recorded bilateral flows to industrialized countries, as well as between intentions and aggregated out‐migration. The results provide policymakers with a reliability assessment of survey data on emigration intentions and encourage future attempts to incorporate survey data in formal statistical migration forecasting models.     

Value Chain Development and the Agrarian Question: Actor Perspectives on Native Potato Production in the Highlands of Peru

Value chain development seeks to integrate smallholder farmers into competitive markets to promote economic and social development. This article, conceptually based on the agrarian question, considers how the perspectives of various value chain actors, with particular focus on smallholders, have important implications for the outcomes of these market‐oriented initiatives. Utilizing Long’s concept of social interface, we present a mixed‐methods case study that analyzes how smallholders, an NGO, and agrifood corporations, including PepsiCo, partnered to establish value chains for native potatoes in the Peruvian highlands. We find that a thorough understanding of the various perspectives held by value chain actors provides important insight into why value chain initiatives have divergent trajectories. Based on the findings, we conclude that accounting for how actors are responding to development initiatives and one another helps explain development outcomes and that therefore the agrarian question remains relevant in current agricultural development discourse and practice.     

WILL GIRLS BE GIRLS? RISK TAKING AND COMPETITION IN AN ALL‐GIRLS' SCHOOL

We conduct an experiment that examines the relationship between girls only schooling and risk taking and competitive behavior. In it, we compare decisions made by students in an all‐girls' school to those made by students in a closely matched coeducational school. We further investigate the developmental nature of this behavior by comparing choices made by younger students (Grades 7 and 8) with those of older students (Grades 11 and 12). We focus on the differences between those who select into the all‐girls' school, and find that although girls educated in a single sex environment are the most risk averse, they are also among the most competitive. These results lend texture to the hypothesis that “nurture matters” in the gender differences debate. (JEL C93, J16, I21)     

Downsizing for downtown: limb lengths,toe lengths,and scale counts decrease with urbanization in western fence lizards (Sceloporus occidentalis)

Urban Ecosystems - Urbanization-induced habitat loss and alteration causes significant challenges for the survival of many species. Identifying how species respond to urbanization can yield...     

Quality of Life Enhancement Research Program: Lessons Learned

The attempted Quality of Life Enhancement program appeared to be a novel method for creating a symbiotic and sustainable relationship between a large Southeastern university and local nursing home and assisted living facilities (ALF), through which ALF residents would attend university arts and sporting events. During implementation, it was discovered that this project was unsustainable, undesired, unneeded by many care centers, difficult to implement, and required a dedicated and specialized staff. After attempting to enroll 14 local care facilities, only 2 were interested in becoming involved and produced 5 eligible participants. During the program itself, coordination with facilities and residents was difficult to maintain and unique transportation and accommodation needs were challenging to overcome. Finally, data collection was time consuming and generally unfruitful. Ultimately the project was discontinued after a year of attempted recruitment and implementation of inclusive changes to protocol. While creating an ongoing relationship with local care facilities and local universities may enhance opportunities for residents and research into important areas, such relationships take time, effort, and specialized staff to maintain.     

Bioecological Implications of Narrative Exposure Therapy in Low‐Resource Settings: Individual,Family, Community,and Socio‐Political Contexts

Traumatic stress can have detrimental effects on individuals, families, and communities. Narrative exposure therapy (NET) is an evidence‐based intervention for decreasing individuals’ post‐traumatic stress disorder (PTSD) symptoms and has been tested in some of the most challenging contexts, such as in post‐conflict refugee camps. Although the focus of NET is on reducing individual PTSD symptoms, the impact of NET can be seen beyond the individual level. The purpose of this paper is to examine some of the ecological implications of using NET with trauma‐affected populations in low‐resource settings. We describe the implications of NET at the family, community, and sociopolitical levels using several case examples. Finally, we outline limitations and future directions for improving the delivery of NET in settings with limited resources.