scholarly journals Twenty steps towards an adequate inferential interpretation of p-values

Author(s):  
Norbert Hirschauer ◽  
Sven Grüner ◽  
Oliver Mußhoff ◽  
Claudia Becker

We suggest twenty immediately actionable steps to reduce widespread inferential errors related to “statistical significance testing.” Our propositions refer first to the theoretical preconditions for using p-values. They furthermore include wording guidelines as well as structural and operative advice on how to present results, especially in research based on multiple regression analysis, the working horse of empirical economists. Our propositions aim at fostering the logical consistency of inferential arguments by avoiding false categorical reasoning. They are not aimed at dispensing with p-values or completely replacing frequentist approaches by Bayesian statistics.

2018 ◽  
Author(s):  
Norbert Hirschauer ◽  
Sven Grüner ◽  
Oliver Mußhoff ◽  
Claudia Becker

We suggest twenty immediately actionable steps to reduce widespread inferential errors related to “statistical significance testing.” Our propositions refer first to the theoretical preconditions for using p-values. They furthermore include wording guidelines as well as structural and operative advice of how to present results, especially in multiple regression analysis. Our propositions aim at fostering the logical consistency of inferential arguments by avoiding false categorical reasoning. They are not aimed at dispensing with p-values or completely replacing frequentist approaches by Bayesian statistics.


2017 ◽  
Author(s):  
Norbert Hirschauer ◽  
Oliver Mußhoff ◽  
Claudia Becker ◽  
Sven Grüner

We suggest twenty immediately actionable steps to reduce widespread inferential errors related to “statistical significance testing.” Our propositions refer first to the theoretical preconditions for using p-values. They furthermore include wording guidelines as well as structural and operative advice of how to present results, especially in multiple regression analysis. Our propositions aim at fostering the logical consistency of inferential arguments by avoiding false categorical reasoning. They are not aimed at dispensing with p-values or completely replacing frequentist approaches by Bayesian statistics.


2019 ◽  
Vol 239 (4) ◽  
pp. 703-721 ◽  
Author(s):  
Norbert Hirschauer ◽  
Sven Grüner ◽  
Oliver Mußhoff ◽  
Claudia Becker

Abstract We suggest twenty immediately actionable steps to reduce widespread inferential errors related to “statistical significance testing.” Our propositions refer to the theoretical preconditions for using p-values. They furthermore include wording guidelines as well as structural and operative advice on how to present results, especially in research based on multiple regression analysis, the working horse of empirical economists. Our propositions aim at fostering the logical consistency of inferential arguments by avoiding false categorical reasoning. They are not aimed at dispensing with p-values or completely replacing frequentist approaches by Bayesian statistics.


2018 ◽  
Author(s):  
Norbert Hirschauer ◽  
Sven Grüner ◽  
Oliver Mußhoff ◽  
Claudia Becker

We suggest twenty immediately actionable steps to reduce widespread inferential errors related to “statistical significance testing.” Our propositions refer first to the theoretical preconditions for using p-values. They furthermore include wording guidelines as well as structural and operative advice of how to present results, especially in multiple regression analysis. Our propositions aim at fostering the logical consistency of inferential arguments by avoiding false categorical reasoning. They are not aimed at dispensing with p-values or completely replacing frequentist approaches by Bayesian statistics.


2019 ◽  
Author(s):  
Norbert Hirschauer ◽  
Sven Grüner ◽  
Oliver Mußhoff ◽  
Claudia Becker

We suggest twenty immediately actionable steps to reduce widespread inferential errors related to “statistical significance testing.” Our propositions refer first to the theoretical preconditions for using p-values. They furthermore include wording guidelines as well as structural and operative advice on how to present results, especially in research based on multiple regression analysis, the working horse of empirical economists. Our propositions aim at fostering the logical consistency of inferential arguments by avoiding false categorical reasoning. They are not aimed at dispensing with p-values or completely replacing frequentist approaches by Bayesian statistics.


2019 ◽  
Vol 12 (1) ◽  
pp. 420-423
Author(s):  
Prapada Watcharanat ◽  
Prasong Tanpichai ◽  
Ravee Sajjasophon

Purpose: This research aims to study the relationship between perception of elderly’s health and health behaviors in Nakhon Nayok province, Thailand Methods: This research was a cross-sectional study. The questionnaire was used to collect the data. This research was conducted in Nakhon Nayok province. The sample size was 270 which applied Taro Yamane's formula at a significant level 0.05. The descriptive statistics was implemented to describe the variables by presenting the frequency, percentage, mean and standard deviation. Furthermore, multiple regression analysis was applied to analyze the relationship between perception of elderly’s health and health behaviors. The statistical significance was considered to reject Hypothesis-null at < 0.05. Results: From a total of 270 people, more than 58.22% of the elderly perceived that they had moderate health conditions. Most elderly had congenital diseases (62.2%). The multiple regression analysis results showed that health status perception and health status perception when compared to their cohort related significantly to health behavior. Conclusion: The government should support the elderly on participation, trust, engagement, and cultural concern of the people in the community, which can contribute to promoting the physical, mental and social condition of the elderly.


Author(s):  
Dragana Tomašević ◽  
Dragana Gašević ◽  
Marija Vranješ

Abstract: Research question: The purpose of this paper is to analyze the factors that influence consumers' intentions when buying foreign brands, and then the purchase itself. Motivation: The business environment is experiencing an extensive change and it has become very demanding for companies to retain customers as well as market share. Companies must learn to adapt to changes in order to survive in the market, and therefore need to be aware of the needs of their customers and their purchasing decisions. The rise in consumer awareness has led consumers to opt for well-known and high-quality foreign brands. Hence, if companies hope to outperform their competitors, they need to persuade consumers to purchase their products and brands. Idea: The main objective of the research is to be based on previous research in this field to find out whether perceived quality, prestige, perceived value and environmental impact affect the intention of domestic consumers to buy foreign brands, and consequently the purchase of these foreign brands. Data: The survey was conducted on the territory of the Republic of Serbia by means of a questionnaire and included 158 respondents. Data were collected from the respondents electronically during a period of about one month. Tools: Statistical analyses of all collected data (utilizing linear correlation, multiple regression analysis, T-test of sample independence and ANOVA) were used to draw conclusions about the relationships between the observed variables and to determine differences in the mean values between individual categories of respondents in terms of buying foreign brands. Findings: Multiple regression analysis confirmed the statistically significant influence of perceived quality and perceived value on the intention of consumers in Serbia to buy foreign brands. Other factors do not have a statistically significant effect on consumer intention. Also, a simple regression analysis revealed a statistically significant influence of consumer intention on the act of purchase. At the very end of the paper, it was determined that statistically significant differences regarding the purchase of foreign brands existed when respondents were observed by gender and education levels, while age did not have a statistical significance. Contribution: This research has a scientific and practical contribution in terms of collecting information about the factors that influence the consumers` intension to buy foreign brands.


2021 ◽  
Vol 17 (3) ◽  
pp. 66-75
Author(s):  
I.V. Tepla

Relevance. Birth weight is one of the main indicators of intrauterine fetal development. The condition of the placenta reflects the influence of the maternal environment on the fetus. It is important to find out which placental growth parameters significantly affect the mass of dichorionic diamniotic (DHDA) twins. Objective: to investigate the relationship between indicators of placental growth and birth weight of dichorionic diamniotic twins. Materials and methods. One hundred thirty-five DCDA twin pairs were studied during 2016-2020: 68 pairs were opposite-sex, 32 – same-sex females and 35 – same-sex males. The gestational age ranged from 29 to 39.5 weeks, averaging 36,4 ± V1,74 weeks. The influence of such parameters as maximum and minimum placental diameters, as well as perimeter, area, circularity (Circ), roundness, thickness, mass (PM), volume (V), displacement of the umbilical cord insertion site from the placental disc centre along its maximum (XCD) and the minimum axis (YCD). Results. A strong positive correlation was found between PM and birth weight in the whole cohort of DCDA twins (r = 0.7059, p < 0.0001). It was strongest in the same-sex female group (r = 0.7886, p < 0.0001), then in the opposite-sex pairs (r = 0.7093, p < 0.0001), slightly weaker – in the same-sex male couples (r = 0.6065, p < 0.0001). The birth weight and PM correlated with almost all indicators that characterize the placental shape. The strongest correlation was with V (rV_BW = 0.66, p < 0.0001; rV_PM = 0.59, p < 0.0001) and the area*Circ (rS*Circ_BW = 0.64; p < 0.0001; rS*Circ_PM = 0.62; p < 0.0001). The multiple regression analysis of the effect of different morphometric parameters of the placentas on the birth weight has demonstrated the statistical significance of the following parameters: XCD, PM, V, YCD and Circ. The general regression model for the whole DCDA group: birth weight = 2242,3 + 0,8*МП + V – 10,0* XCD (R2 = 0,7002). Similar equations were calculated for different types of DCDA tweens. Thus, morphometric measures account for the 67-85% variances of the birth weight of the DCDA tweens. Conclusion. In the DCDA pairs, placental mass strongly positively correlated with birth weight. Both indicators are in strong relationship with V and their area*Circ. The closer the placental shape is to the circle, the higher its functional ability. Deterioration of the latter may be due to the far fixation of the umbilical cord from the center of the placental disc. The displacement along the maximum axis has a stronger impact comparing to the displacement relative to the minimum axis. The multiple regression analysis has revealed that statistical significance had XCD and PM or V in the whole types of twins. Proposed models demonstrate that 67-85% of the variability of birth weight may be accounted for by the three or four simple placental measures 


Sign in / Sign up

Export Citation Format

Share Document