The reporting of p values, confidence intervals and statistical significance in Preventive Veterinary Medicine (1997–2017)

Background Despite much discussion in the epidemiologic literature surrounding the use of null hypothesis significance testing (NHST) for inferences, the reporting practices of veterinary researchers have not been examined. We conducted a survey of articles published in Preventive Veterinary Medicine, a leading veterinary epidemiology journal, aimed at (a) estimating the frequency of reporting p values, confidence intervals and statistical significance between 1997 and 2017, (b) determining whether this varies by article section and (c) determining whether this varies over time. Methods We used systematic cluster sampling to select 985 original research articles from issues published in March, June, September and December of each year of the study period. Using the survey data analysis menu in Stata, we estimated overall and yearly proportions of article sections (abstracts, results-texts, results-tables and discussions) reporting p values, confidence intervals and statistical significance. Additionally, we estimated the proportion of p values less than 0.05 reported in each section, the proportion of article sections in which p values were reported as inequalities, and the proportion of article sections in which confidence intervals were interpreted as if they were significance tests. Finally, we used Generalised Estimating Equations to estimate prevalence odds ratios and 95% confidence intervals, comparing the occurrence of each of the above-mentioned reporting elements in one article section relative to another. Results Over the 20-year period, for every 100 published manuscripts, 31 abstracts (95% CI [28–35]), 65 results-texts (95% CI [61–68]), 23 sets of results-tables (95% CI [20–27]) and 59 discussion sections (95% CI [56–63]) reported statistical significance at least once. Only in the case of results-tables, were the numbers reporting p values (48; 95% CI [44–51]), and confidence intervals (44; 95% CI [41–48]) higher than those reporting statistical significance. We also found that a substantial proportion of p values were reported as inequalities and most were less than 0.05. The odds of a p value being less than 0.05 (OR = 4.5; 95% CI [2.3–9.0]) or being reported as an inequality (OR = 3.2; 95% CI [1.3–7.6]) was higher in the abstracts than in the results-texts. Additionally, when confidence intervals were interpreted, on most occasions they were used as surrogates for significance tests. Overall, no time trends in reporting were observed for any of the three reporting elements over the study period. Conclusions Despite the availability of superior approaches to statistical inference and abundant criticism of its use in the epidemiologic literature, NHST is substantially the most common means of inference in articles published in Preventive Veterinary Medicine. This pattern has not changed substantially between 1997 and 2017.

Measurement of Cosmetic Purchasing Decisions Using Brand Image and Consumer Trust

This study aimed to measure the purchasing decisions of cosmetics using the brand image and consumer trust. This type of research is quantitative research and uses a sample of 70 respondents. The terms of the instrument test include the validity and reliability of the data. The analysis requirements used normal, homogeneous, linear, and regression significance tests. The data were analyzed using structural equations. The research findings show that brand image affects consumer trust, brand image affects purchasing decisions, and consumer trust also affects purchasing decisions.

Digestibility Coefficients and Energy in Alfalfa Hay from Chimborazo and Tungurahua

This study was part of a project (Grasas By Pass), funded by Ciencias Pecuarias ESPOCH. Samples from Chimborazo and Tungurahua were analyzed. The statistical analysis included descriptive analysis and significance tests, including Tukey test. The following results of ranges were found in the first phase: humidity: 7.79% (± 0.06) – 9.99% (± 0.08); ashes: 9.16% (± 0.14) – 10.50% (± 0.01); protein: 18.73% (± 0.10) – 21.37% (± 0.52); crude fiber: 31.64% (± 0.26) – 41.69% (± 0.11); ethereal extract: 1.36% (± 0.22) – 2.13% (± 0.50); and ELN: 28.30% (± 0.09) – 34.46% (± 0.24). The results from the second phase were: F.D.N.: 41.39% (± 1.32) – 51.77% (± 0.01); F.D.A.: 37.36% (± 0.26) – 45.47% (± 1.04); and L.D.A.: 7.85% (± 0.48) – 9.30% (± 0.39). In vivo digestibility tests were developed in metabolic cages with six sheep per treatment for 11 days. The following results were found: Dry matter digestibility coefficient (C.D.M.S.): 60.78% – 66.42%; organic matter (C.D.M.O.): 61.97% – 68.24%; protein (C.D.P.B.): 74.35% – 80.23%; crude fiber (C.D.F.C): 43.73% – 59.39%; ethereal extract (C.D.E.E.): 26.56% – 51.76%; and E.L.N. (C.D.ELN): 67.26% – 81.96%. Finally, the Van Es equations were applied to estimate the values of E.M., E.Nl., E.Ng. and N.D.T. The digestibility and energy coefficients for the hay from Chimborazo - San Juan were higher, and there were lower values for Tungurahua – Cunchibamba. Keywords: chemical composition, digestibility coefficient, useful energy. RESUMEN Como parte del proyecto financiado Grasas By Pass, de la Facultad de Ciencias Pecuarias ESPOCH; se analizaron muestras procedentes de Chimborazo y Tungurahua; el análisis estadístico incluyó análisis descriptivo y pruebas de significancia según Tukey P < 0.0001, en la primera etapa se realizó el análisis proximal que reportó rangos de: humedad 7,79% (±0,06) –9,99% (±0,08), cenizas 9,16% (±0,14) – 10,50% (±0,01), proteína 18,73% (±0,10) – 21,37% (±0,52), fibra cruda 31,64% (±0,26) – 41,69% (±0,11), extracto etéreo 1,36% (±0,22) – 2,13% (±0,50) y E.L.N. 28,30% (±0,09) – 34,46% (±0,24). La segunda fase analizó: F.D.N. con valores de 41,39% (±1,32) – 51,77%(±0,01), F.D.A. 37,36% (±0,26) – 45,47% (±1,04) y L.D.A. 7,85% (±0,48) – 9,30% (±0,39). Se desarrolló las pruebas de digestibilidad In vivo en jaulas metabólicas con 6 ovinos por tratamiento durante 11 días, encontrándose rangos para el coeficiente de digestibilidad de la materia seca (C.D.M.S.) desde 60,78% – 66,42%; materia orgánica (C.D.M.O.) de 61,97% a 68,24%, proteína (C.D.P.B.) de 74,35% a 80,23%, fibra cruda (C.D.F.C) desde 43,73% hasta 59,39%, extracto etéreo (C.D.E.E.) 26,56% a 51,76%, y E.L.N. (C.D.ELN) desde 67,26% hasta 81,96%. Finalmente se aplicó las ecuaciones de Van Es para estimar los valores de E.M., E.Nl., E.Ng. N.D.T. Siendo más elevados los coeficientes de digestibilidad y energía para los henos de Chimborazo - San Juan y valores más bajos para Tungurahua- Cunchibamba. Palabras claves: composición química, coeficiente de digestibilidad, energía útil.

Analysis of the causes and effects of cyclist-pedestrian accidents in biggest Polish cities

Traffic incidents between pedestrians and cyclists result in an incomparably smaller number of victims (injured and killed) than accidents between unprotected traffic participants and other vehicles. However, such incidents cannot be underestimated, as in most cases they take place on elements of infrastructure designed for pedestrians and cyclists, and thus negatively affect the sense of safety of users in places where they should not only feel safe but also comfortable. This paper presents an analysis of such traffic incidents, aimed at recognizing the share of pedestrians and cyclists as perpetrators and also victims of accidents. Three research hypotheses were examined: that the type of infrastructure and also light and weather conditions influences the structure of perpetrators and victims, that the proportion of accidents between pedestrians and cyclists caused by pedestrians is increasing, and that the proportion of victims of accidents between pedestrians and cyclists in the pedestrian group is de-creasing. Analyses were performed based on the number of traffic incidents involving cyclists and pedestrians in the six largest Polish cities, registered in the Police Accident and Collision Records System. A total of 1,702 incidents involving 1,034 injured and killed people in years 2007-2018 were considered. Each traffic incident was considered individually, taking into account changes in perpetrator and victim proportions depending on the type and purpose of the infrastructure and external conditions, as well as variability of these proportions over subsequent years. The tools of mathematical statistics were used, including – among others - significance tests for independent proportions and Chi-squared test for trend. On the basis of the conducted statistical analyses, all research hypothesis were proved. It also confirmed that although the proportions are changing, there are still much more traffic incidents are caused by cyclists, but more victims are in group of pedestrians. The results of the research confirm the need to take action to develop effective mechanisms of mutual interaction between pedestri-ans and cyclists. Especially in view of the growing bicycle traffic in Polish cities.

Improved procedures and computer programs for equivalence assessment of correlation coefficients

The correlation coefficient is the most commonly used measure for summarizing the magnitude and direction of linear relationship between two response variables. Considerable literature has been devoted to the inference procedures for significance tests and confidence intervals of correlations. However, the essential problem of evaluating correlation equivalence has not been adequately examined. For the purpose of expanding the usefulness of correlational techniques, this article focuses on the Pearson product-moment correlation coefficient and the Fisher’s z transformation for developing equivalence procedures of correlation coefficients. Equivalence tests are proposed to assess whether a correlation coefficient is within a designated reference range for declaring equivalence decisions. The important aspects of Type I error rate, power calculation, and sample size determination are also considered. Special emphasis is given to clarify the nature and deficiency of the two one-sided tests for detecting a lack of association. The findings demonstrate the inappropriateness of existing methods for equivalence appraisal and validate the suggested techniques as reliable and primary tools in correlation analysis.

Effect Size and Effect Uncertainty in Organizational Research Methods

To understand and communicate research findings, it is important for researchers to consider two types of information provided by research results: the magnitude of the effect and the degree of uncertainty in the outcome. Statistical significance tests have long served as the mainstream method for statistical inferences. However, the widespread misinterpretation and misuse of significance tests has led critics to question their usefulness in evaluating research findings and to raise concerns about the far-reaching effects of this practice on scientific progress. An alternative approach involves reporting and interpreting measures of effect size along with confidence intervals. An effect size is an indicator of magnitude and direction of a statistical observation. Effect size statistics have been developed to represent a wide range of research questions, including indicators of the mean difference between groups, the relative odds of an event, or the degree of correlation among variables. Effect sizes play a key role in evaluating practical significance, conducting power analysis, and conducting meta-analysis. While effect sizes summarize the magnitude of an effect, the confidence intervals represent the degree of uncertainty in the result. By presenting a range of plausible alternate values that might have occurred due to sampling error, confidence intervals provide an intuitive indicator of how strongly researchers should rely on the results from a single study.

The Effect of Preprocessing Techniques, Applied to Numeric Features, on Classification Algorithms’ Performance

It is recognized that the performance of any prediction model is a function of several factors. One of the most significant factors is the adopted preprocessing techniques. In other words, preprocessing is an essential process to generate an effective and efficient classification model. This paper investigates the impact of the most widely used preprocessing techniques, with respect to numerical features, on the performance of classification algorithms. The effect of combining various normalization techniques and handling missing values strategies is assessed on eighteen benchmark datasets using two well-known classification algorithms and adopting different performance evaluation metrics and statistical significance tests. According to the reported experimental results, the impact of the adopted preprocessing techniques varies from one classification algorithm to another. In addition, a statistically significant difference between the considered data preprocessing techniques is demonstrated.