Blakeley B. McShane’s research while affiliated with Northwestern University and other places

... Finally, over-interpretation of arbitrary thresholds like 0.05 have also led to phacking (selective reporting) to achieve significant results. This has led to many calls for reform (5)(6)(7)(8)(9)(10)(11)(12) and to address this issue, multiple researchers have either proposed a shift in the language used (13)(14)(15), a shift in the scale used to introduce and interpret p values (16), or even a retirement of statistical significance altogether (17,18). ...

Reference:

From Significance to Divergence: Guiding Statistical Interpretation Through Language

... For the latter, the larger the volume of data, the better the representation of reality; but then complexity and management costs increase, and scholars may need to make compromises. That is why researchers resort to sampling and measurement error, with the consequent effects on variance (McShane et al., 2024;McShane & Bockenholt, 2016). Overall, while bias results from an incomplete representation of the true data-generating mechanism by a model because of simplifying assumptions, variance results from random variation in the data due to sampling and measurement error (Wedel & Kannan, 2016, p. 104). ...

Reference:

Sampling and sample size in B2B marketing: Current practices and recommendations

... Heterogeneity and meta-analysis in replicability. Multi-site replication projects have been used to examine the heterogeneity in effect estimates across sites (Klein et al., 2018;Coppock et al., 2018;McShane et al., 2022;Delios et al., 2022;Holzmeister et al., 2024). A prominent distinction is that these works often measure certain global notions of heterogeneity via meta-analysis (McShane et al., 2022), while we focus on generalization from one site to another. ...

Reference:

Beyond Reweighting: On the Predictive Role of Covariate Shift in Effect Generalization

... This is of particular importance, as mentioned above, given the small number of studies in certain subgroups when techniques are divided by timepoint. Lastly, we preregistered that we would perform three-level multilevel meta-analyses for each outcome individually, instead of conducting multivariate analyses 91 . Multivariate analyses could be advantageous as it allows researchers to examine various types of outcome (for example, intrusion frequency, distress and symptoms) in a single analysis, accounting Nature Human Behaviour Article https://doi.org/10.1038/s41562-024-01956-y ...

Reference:

A systematic review and meta-analysis of experimental methods for modulating intrusive memories following lab-analogue trauma exposure in non-clinical populations

... Moreover, there were no instances in which non-normality and/or heteroscedasticity in measures of root length intensity or other continuous metrics were addressed except by a transformation. That the resulting estimates and standard errors were often reported on the transformed scale suggests that, although transformation bias is of minimal concern, researchers may be giving insufficient attention to estimates, and their interpretation, relative to test statistics and p-values: statistical significance taking precedence over biological importance (McShane & Gelman, 2022). In two-thirds of these cases, the measurements were log transformed, and in a few cases a small constant was added either to all values or to zero values prior to transformation. ...

Reference:

Statistical considerations in the analysis of minirhizotron data and a review of current practice in agronomic research

... Replicability need not be narrowly defined as the probability of rejecting H 0 in the same direction over replication studies. There are other productive definitions of replication that are not based on power (e.g., Braver, Thoemmes, & Rosenthal, 2014;Kenny & Judd, 2019;Hedges & Schauer, 2019b;Hung & Fithian, 2020;Maxwell et al., 2015;Open Science Collaboration, 2015;Patil et al., 2016;Mathur & VanderWeele, 2019;McShane et al., 2022). Additionally, S. F. Anderson and Maxwell (2016) describe and review different replication goals that can be achieved with different analyses, focusing on information generated from replication studies that can reduce un-certainty about studied effects (see also Schmidt, 1996). ...

Reference:

Uses of Uncertain Statistical Power: Designing Future Studies, Not Evaluating Completed Studies

... A P-curve analysis(Simonsohn, Nelson and Simmons 2014) shows that both conditions for the full p-curve and half p-curve are met, indicating evidential value across our studies. We also conducted a single-paper metaanalysis using the method prescribed byMcShane and Böckenholt (2022). The meta-analysis included all studies reported in the main text, replication studies, file drawer studies, pilot tests, and stimuli calibration tests (k = 33). ...

Reference:

Price Partitioning of Socio-Moral Surcharges

... One of the crucial aspects of human decision making is that, as fundamentally social creatures, our preferences are strongly influenced by our social context. Viral trends, conformity, word-of-mouth, and signaling all play roles in behavior, including choices (Feinberg et al., 2020;Axsen & Kurani, 2012). Additionally, people with similar preferences, beliefs, and identities are more likely to be friends in the first place, a phenomenon known as homophily (McPherson et al., 2001). ...

Reference:

Graph-based methods for discrete choice

... Its levels are positively correlated with a range of cognitive states and behaviors that aid success in competitive situations and conflict. These states and behaviors include implicit power motivation, reduced sensitivity to threat, risk-taking, intuitive (rather than deliberative) and 'hawkish' decision-making, overconfidence, persistence, motivational drive, fearlessness, stress resilience, willingness to enter competitive interactions, and conspicuous consumption to increase perceived status-as evidenced in studies measuring endogenous T levels or natural T reactivity (Apicella et al., 2008;Carré and McCormick, 2008;Coates et al., 2009;Coates et al., 2010;Johnson et al., 2006;Mehta et al., 2017;Mehta and Josephs, 2006;Schultheiss et al., 2004;Schultheiss et al., 2005;Stanton and Schultheiss, 2009;Wu et al., 2017) or using exogenous T administration (Apicella et al., 2015;Hermans et al., 2006;Hermans et al., 2007;Knight et al., 2020;Nave et al., 2017Nave et al., , 2018. ...

Reference:

Narcissism moderates the association between basal testosterone and generosity in men

... In line with this latter conclusion, we have repeatedly demonstrated that perceiving or even processing the magnitude of centrally presented Arabic numbers (Pinto et al., 2018;Fattorini, Pinto, Merola, D'Onofrio, & Doricchi, 2016;Fattorini, Pinto, Rotondaro, & Doricchi, 2015; see also Zanolie & Pecher, 2014) or numerosities without using spatial response codes does not trigger directionally corresponding shifts of spatial attention, for example, perceiving or classifying a number or numerosity as lower than 5 does not shift attention to the left. These negative results were confirmed by a subsequent massive replication study involving 17 different laboratories (Colling et al., 2020) that, in line with our findings, provided no evidence for the presence of the attentional SNARC effect originally described by Fischer and colleagues (2003). ...

Reference:

Changes in Brain Functional Connectivity Underlying the Space–Number Association