Why and How We Should Join the Shift From Significance Testing to Estimation

A paradigm shift away from null hypothesis significance testing seems in progress. Based on simulations, we illustrate some of the underlying motivations. First, P-values vary strongly from study to study, hence dichotomous inference using significance thresholds is usually unjustified. Second, statistically significant results have overestimated effect sizes, a bias declining with increasing statistical power. Third, statistically non-significant results have underestimated effect sizes, and this bias gets stronger with higher statistical power. Fourth, the tested statistical hypotheses generally lack biological justification and are often uninformative. Despite these problems, a screen of 48 papers from the 2020 volume of the Journal of Evolutionary Biology exemplifies that significance testing is still used almost universally in evolutionary biology. All screened studies tested the default null hypothesis of zero effect with the default significance threshold of p = 0.05, none presented a pre-planned alternative hypothesis, and none calculated statistical power and the probability of ‘false negatives’ (beta error). The papers reported 49 significance tests on average. Of 41 papers that contained verbal descriptions of a ‘statistically non-significant’ result, 26 (63%) falsely claimed the absence of an effect. We conclude that our studies in ecology and evolutionary biology are mostly exploratory and descriptive. We should thus shift from claiming to “test” specific hypotheses statistically to describing and discussing many hypotheses (effect sizes) that are most compatible with our data, given our statistical model. We already have the means for doing so, because we routinely present compatibility (“confidence”) intervals covering these hypotheses.

Antibacterial Effect of Disinfectants and Antiseptics

Disinfectants and antiseptics are substances or solvents which play a key role inthe control of infections. Their wide functionality is applied in almost all human activities. Theseagents can have antiviral, antibacterial and antifungal effects. Disinfectants are synthetic agents thebiocidal effect of which destroys microorganisms in the environment, while antiseptics are appliedlocally, primarily on skin and/or mucosae. The morphological differences between Gram (+) andGram (–) bacteria raise questions regarding the biocidal effects of antibacterial agents againstdifferent types of infectious agents. The aim of our study is to evaluate the biocidal effect ofdisinfectants and antiseptics available on the market. The study uses a reference strain ofEscherichia coli and a laboratory isolate of Staphylococcus aureus. The analysis was conductedusing the agar well diffusion method. Results show that sodium hypochlorite, hydrogen peroxideH2O2 and mouthwashing water demonstrate strong bactericidal effect on both Gram (+) and Gram(–) bacteria. The bath cleaner gel, which is considered effective against Escherichia coli,demonstrated effectiveness against Staphylococcus aureus instead. The dishwashing gel,ethacridine lactate, silver water and lavandula water show intermediate or zero effect againstEscherichia coli and Staphylococcus aureus.

The Deterrence and Prevention of Cyber Conflict

For all its apparent complexity, deterrence is built upon a relatively straightforward premise: that fear of adverse consequences can serve to prevent unacceptable behaviour. In traditional deterrence theory, particularly that which matured during the Cold War, prevention could be achieved both by fear of punishment and by fear of failure: by punitive measures that would be undertaken in response to a transgression and intended to inflict some pain, damage, or loss upon the transgressor; or by a denial posture intended to impress upon an adversary that the complexities of the defences will be technologically insurmountable and/or the costs of doing so unbearably high. Fresh thinking is required to make deterrence relevant to the cyber era. The challenges of cyber warfare are described here as the ‘four zeros’: ‘zero day’ vulnerability to novel malware; ‘zero source’ or non-attribution of an attack; ‘zero effect’ or covert, non-detectable attack; and ‘zero intent’ or lack of clarity as to an adversary’s intentions. These challenges can be met by allowing deterrence to function in different ways as circumstances allow or require: punitive, constructive, and protective deterrence.

Estimation of regional polygenicity from GWAS provides insights into the genetic architecture of complex traits

The number of variants that have a non-zero effect on a trait (i.e. polygenicity) is a fundamental parameter in the study of the genetic architecture of a complex trait. Although many previous studies have investigated polygenicity at a genome-wide scale, a detailed understanding of how polygenicity varies across genomic regions is currently lacking. In this work, we propose an accurate and scalable statistical framework to estimate regional polygenicity for a complex trait. We show that our approach yields approximately unbiased estimates of regional polygenicity in simulations across a wide-range of various genetic architectures. We then partition the polygenicity of anthropometric and blood pressure traits across 6-Mb genomic regions (N = 290K, UK Biobank) and observe that all analyzed traits are highly polygenic: over one-third of regions harbor at least one causal variant for each of the traits analyzed. Additionally, we observe wide variation in regional polygenicity: on average across all traits, 48.9% of regions contain at least 5 causal SNPs, 5.44% of regions contain at least 50 causal SNPs. Finally, we find that heritability is proportional to polygenicity at the regional level, which is consistent with the hypothesis that heritability enrichments are largely driven by the variation in the number of causal SNPs.

Observed effects of education on intelligence in longitudinal data may be spurious: a formal model and reanalysis

Effects of education on intelligence are controversial. Earlier studies of longitudinal data have observed positive associations between level of education and a later measurement of intelligence, when statistically controlling for an earlier measurement of intelligence, and furthermore that this association is stronger among individuals with lower pre-education intelligence. Here we challenge the interpretation that these observations reflect a causal effect of education. We develop and analyze a mathematical model in which education is assumed to have zero effect on intelligence, showing that precisely the observed pattern of results arises as a statistical artefact due to measurement errors. Fitting our model to a dataset used in a prior study, we show that observed associations between education and intelligence are closely replicated in simulated data generated by our model. Thus, our reanalysis indicates that additional higher education does not cause an increase in intelligence. We discuss how positive findings in studies of policy changes and school-age cutoff are limited to basic education and may not generalize to higher education.

The role of journalism in combating fake news: the case of the last day of the 2019 election campaign in Portugal

This article reflects on the role of journalism in the deconstruction of fake news propaganda that came out in the media on the last day of the 2019 parliamentary election campaign in Portugal. We collected news items carried by the Portuguese media and contextualised this media coverage with regard to the impact of disinformation on confidence in the news with the help of data collated as part of the Digital News Report project. We found that journalistic scrutiny, aided by the characteristics of the Portuguese media system, might have contributed to a zero effect of this fake news on the election results, unlike what happened in elections in other countries, such as the United States, United Kingdom and Brazil.

Effects of tDCS on the attentional blink revisited: A statistical evaluation of a replication attempt

The attentional blink (AB) phenomenon reveals a bottleneck of human information processing: the second of two targets is often missed when they are presented in rapid succession among distractors. A recent study by London & Slagter (Journal of Cognitive Neuroscience, 33, 756 to 768, 2021) showed that the size of the AB can be changed by applying transcranial direct current stimulation (tDCS) over the left dorsolateral prefrontal cortex (lDLPFC). Although AB size at the group level remained unchanged, the effects of anodal and cathodal tDCS were negatively correlated: if a given individuals AB size decreased from baseline during anodal tDCS, their AB size would increase during cathodal tDCS, and vice versa. Here, we attempted to replicate this finding. Like London & Slagter, we found no group effects of tDCS, but also no longer found a significant negative correlation. We present a series of statistical measures of replication success, all of which confirm that both studies are not in agreement. First, the correlation here is significantly smaller than a conservative estimate of the original correlation. Second, the difference between the correlations is greater than expected due to sampling error, and our data are more consistent with a zero effect than with the original estimate. Finally, the overall effect when combining both studies is small and not significant. Our findings thus indicate that the effects of lDPLFC tDCS on the AB are less substantial than observed by London & Slagter (2021). Although this should be quite a common scenario, null findings can be difficult to interpret and are still under represented in the brain stimulation and cognitive neuroscience literatures. An important auxiliary goal of this paper is therefore to provide a tutorial for other researchers, to maximize the evidential value from null findings.

Do People Respond to the Mortgage Interest Deduction? Quasi-Experimental Evidence from Denmark

Using a major reform that scaled back the mortgage interest deduction for middle- and high-income households in Denmark, we study how tax subsidies affect housing decisions. We present four main findings. First, the mortgage deduction has a precisely estimated zero effect on homeownership for high- and middle-income households. Second, the mortgage deduction has a clear effect on housing demand at the intensive margin, inducing homeowners to buy larger and more expensive houses. Third, the deduction has sizeable effects on household financial decisions, inducing them to increase indebtedness. Finally, the reduction of the tax subsidy lowered equilibrium house prices. (JEL G21, G51, H24, K34, R21, R31)