Statistical hypothesis testing and common misinterpretations: should we abandon p-value in forensic science applications?

Many people regard the concept of hypothesis testing as fundamental to inferential statistics. Various schools of thought, in particular frequentist and Bayesian, have promoted radically different solutions for taking a decision about the plausibility of competing hypotheses. Comprehensive philosophical comparisons about their advantages and drawbacks are widely available and continue to span over large debates in the literature. More recently, controversial discussion was initiated by an editorial decision of a scientific journal [1] to refuse any paper submitted for publication containing null hypothesis testing procedures. Since the large majority of papers published in forensic journals propose the evaluation of statistical evidence based on the so called p-values, it is of interest to expose the discussion of this journal's decision within the forensic science community. This paper aims to provide forensic science researchers with a primer on the main concepts and their implications for making informed methodological choices.

Interpretation of the p value : A national survey study in academic psychologists from Spain

Resumen tomado de la publicaci??n

The F test for comparing two normal variances: correct and incorrect calculation of the two-sided p-value?

This article illustrates that not all statistical software packages are correctly calculating a p-value for the classical F test comparison of two independent Normal variances. This is illustrated with a simple example, and the reasons why are discussed. Eight different software packages are considered.

Approximation of Bayesian predictive p-values with regression ABC

In the Bayesian framework a standard approach to model criticism is to compare some function of the observed data to a reference predictive distribution. The result of the comparison can be summarized in the form of a p-value, and it's well known that computation of some kinds of Bayesian predictive p-values can be challenging. The use of regression adjustment approximate Bayesian computation (ABC) methods is explored for this task. Two problems are considered. The first is the calibration of posterior predictive p-values so that they are uniformly distributed under some reference distribution for the data. Computation is difficult because the calibration process requires repeated approximation of the posterior for different data sets under the reference distribution. The second problem considered is approximation of distributions of prior predictive p-values for the purpose of choosing weakly informative priors in the case where the model checking statistic is expensive to compute. Here the computation is difficult because of the need to repeatedly sample from a prior predictive distribution for different values of a prior hyperparameter. In both these problems we argue that high accuracy in the computations is not required, which makes fast approximations such as regression adjustment ABC very useful. We illustrate our methods with several samples.

Incongruence between test statistics and P values in medical papers

Given an observed test statistic and its degrees of freedom, one may compute the observed P value with most statistical packages. It is unknown to what extent test statistics and P values are congruent in published medical papers. Methods: We checked the congruence of statistical results reported in all the papers of volumes 409–412 of Nature (2001) and a random sample of 63 results from volumes 322–323 of BMJ (2001). We also tested whether the frequencies of the last digit of a sample of 610 test statistics deviated from a uniform distribution (i.e., equally probable digits).Results: 11.6% (21 of 181) and 11.1% (7 of 63) of the statistical results published in Nature and BMJ respectively during 2001 were incongruent, probably mostly due to rounding, transcription, or type-setting errors. At least one such error appeared in 38% and 25% of the papers of Nature and BMJ, respectively. In 12% of the cases, the significance level might change one or more orders of magnitude. The frequencies of the last digit of statistics deviated from the uniform distribution and suggested digit preference in rounding and reporting.Conclusions: this incongruence of test statistics and P values is another example that statistical practice is generally poor, even in the most renowned scientific journals, and that quality of papers should be more controlled and valued

Measurement of the CP-violating phase phi sJ/psi phi using the flavor-tagged decay Bs(0) -> J/psi phi in 8 fb(-1) of p(p)over-bar collisions

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

P-233-Biodiversity of mucosa associated microbiome of Crohn's disease patients living in middle western São Paulo State, Brazil

Background: The intestinal microbiome (IM) has extensively been studied in the search for a link of bacteria with the cause of Crohn`s disease (CD). The association might result from the action of a specific pathogen and/or an eventual imbalance in bacterial species composition of the gut. The innumerous virulence associated markers and strategies described for adherent and invasive Escherichia coli (AIEC) have made them putative candidate pathogens for CD. IM of CD patients shows dysbiosis, manifested by the proliferation of bacterial groups such as Enterobacteriaceae and reduction of others such as Lactobacillus and Bifidobacterium. The augmented bacterial population comprising of commensal and/or pathogenic organisms super stimulates the immune system, triggering the inflammatory reactions responsible for the clinical manifestations of the disease. Considering the role played by IM in CD and the multiple variables influencing its species composition, resulting in differences among populations, the objective of this study was to determine the bacterial biodiversity in the mucosa associated microbiome of CD patients from a population not previously subject to this analysis, living in the middle west region of Sao Paulo state. Methods: A total of 4 CD patients and 5 controls subjects attending the Botucatu Medical School of the Sao Paulo State University (UNESP) for routine colonoscopy and who signed an informed consent were included in the study. A number of 2 biopsies, one from the ileum and other from any part of the terminal colon, were taken from each subject and immediately frozen at -70[degrees]C until DNA purification. The bacterial biodiversity was assessed by next generation (ion torrent) sequencing of PCR amplicons of the ribosomal DNA 16S V6 region (16S V6 rDNA). The bacterial identification was performed at the genus level, by alignment of the generated DNA sequences with those available at the ribosomal database project (RDP) website. Results: The overall DNA sequence output was based on an average number of 526,427 reads per run, matching 50 bacterial genus 16SrDNA sequences available at the RDB website, and 22 non matching sequences. Over 95% of the sequences corresponded to taxa belonging to the major phyla: Firmicutes, Bacterioidetes, Proteobacteria and Actinobacteria. Irrespective of the intestinal site analyzed, no case-control differences could be observed in the prevalence of Actinobacteria and Firmicutes. The prevalence of Proteobacteria was higher (40%) in the biopsies of control subjects as compared to that of DC patients (16%). For Bacterioidetes, the higher prevalence was observed among DC patients (33% as opposed to 14,5% in controls). The significance for all comparisons considered a p value < 0,05 in a Chi2 test. No mucosal site specific differences could be observed in IM comparisons of CD and control subjects. Conclusions: The rise in the number of Bacterioidetes observed here among CD patients seems to be in agreement with most of studies published thus far. Yet, the reduction in the number of Proteobacteria along with an apparently unaltered population of Actinobacteria and Firmicutes, which include the so called "beneficial" organisms Bifidobacterium and Lactobacillus were rather surprising. These data suggest that the analyses on the role of IM in CD should consider the multiple variables that may influence its species composition.

A common variant of PNPLA3 (p.I148M) is not associated with alcoholic chronic pancreatitis

Background Chronic pancreatitis (CP) is an inflammatory disease that in some patients leads to exocrine and endocrine dysfunction. In industrialized countries the most common aetiology is chronic alcohol abuse. Descriptions of associated genetic alterations in alcoholic CP are rare. However, a common PNPLA3 variant (p.I148M) is associated with the development of alcoholic liver cirrhosis (ALC). Since, alcoholic CP and ALC share the same aetiology PNPLA3 variant (p.I148M) possibly influences the development of alcoholic CP. Methods Using melting curve analysis we genotyped the variant in 1510 patients with pancreatitis or liver disease (961 German and Dutch alcoholic CP patients, 414 German patients with idiopathic or hereditary CP, and 135 patients with ALC). In addition, we included in total 2781 healthy controls in the study. Results The previously published overrepresentation of GG-genotype was replicated in our cohort of ALC (p-value <0.0001, OR 2.3, 95% CI 1.6–3.3). Distributions of genotype and allele frequencies of the p.I148M variant were comparable in patients with alcoholic CP, idiopathic and hereditary CP and in healthy controls. Conclusions The absence of an association of PNPLA3 p.I148M with alcoholic CP seems not to point to a common pathway in the development of alcoholic CP and alcoholic liver cirrhosis.

Diagnostic Performance and Additional Value of Elastosonography in Focal Breast Lesions: Statistical Correlation between Size-Dependant Strain Index Measurements, Multimodality-BI-RADS Score, and Histopathology in a Clinical Routine Setting

Objective. To evaluate the diagnostic benefit of real-time elastography (RTE) in clinical routine. Strain indices (SI) for benign and malignant tumors were assessed. Methods. 100 patients with 110 focal breast lesions were retrieved. Patients had mammography (MG), ultrasound (US), and, if necessary, MRI. RTE was conducted after ultrasound. Lesions were assessed with BI-RADS for mammography and ultrasound. Diagnosis was established with histology or follow-up. Results. SI for BI-RADS 2 was 1.71 ± 0.86. Higher SI (2.21 ± 1.96) was observed for BI-RADS 3 lesions. SI of BI-RADS 4 and 5 lesions were significantly higher (16.92 ± 20.89) and (19.54 ± 10.41). 31 malignant tumors exhibited an average SI of 16.13 ± 14.67; SI of benign lesions was 5.29 ± 11.87 (P value <0.0001). ROC analysis threshold was >3.8 for malignant disease. Sensitivity of sonography was 90.3% (specificity 78.5%). RTE showed a sensitivity of 87.1% (specificity 79.7%). Accuracy of all modalities combined was 96.8%. In BI-RADS 3 lesions RTE was able to detect all malignant lesions (sensitivity 100%, specificity 92.9%, and accuracy 93.9%). Conclusions. RTE increased sensitivity and specificity for breast cancer detection when used in combination with ultrasound.

Weather variability, tides, and Barmah Forest Virus Disease in the Gladstone region, Australia

In this study we examined the impact of weather variability and tides on the transmission of Barmah Forest virus (BFV) disease and developed a weather-based forecasting model for BFV disease in the Gladstone region, Australia. We used seasonal autoregressive integrated moving-average (SARIMA) models to determine the contribution of weather variables to BFV transmission after the time-series data of response and explanatory variables were made stationary through seasonal differencing. We obtained data on the monthly counts of BFV cases, weather variables (e.g., mean minimum and maximum temperature, total rainfall, and mean relative humidity), high and low tides, and the population size in the Gladstone region between January 1992 and December 2001 from the Queensland Department of Health, Australian Bureau of Meteorology, Queensland Department of Transport, and Australian Bureau of Statistics, respectively. The SARIMA model shows that the 5-month moving average of minimum temperature (β = 0.15, p-value < 0.001) was statistically significantly and positively associated with BFV disease, whereas high tide in the current month (β = −1.03, p-value = 0.04) was statistically significantly and inversely associated with it. However, no significant association was found for other variables. These results may be applied to forecast the occurrence of BFV disease and to use public health resources in BFV control and prevention.

Cardiac state diagnosis using higher order spectra of heart rate variability

Heart rate variability (HRV) refers to the regulation of the sinoatrial node, the natural pacemaker of the heart, by the sympathetic and parasympathetic branches of the autonomic nervous system. Heart rate variability analysis is an important tool to observe the heart's ability to respond to normal regulatory impulses that affect its rhythm. A computer-based intelligent system for analysis of cardiac states is very useful in diagnostics and disease management. Like many bio-signals, HRV signals are nonlinear in nature. Higher order spectral analysis (HOS) is known to be a good tool for the analysis of nonlinear systems and provides good noise immunity. In this work, we studied the HOS of the HRV signals of normal heartbeat and seven classes of arrhythmia. We present some general characteristics for each of these classes of HRV signals in the bispectrum and bicoherence plots. We also extracted features from the HOS and performed an analysis of variance (ANOVA) test. The results are very promising for cardiac arrhythmia classification with a number of features yielding a p-value < 0.02 in the ANOVA test.