China
Africa
Explore chapters and articles related to this topic
Conditioned Allergic Rhinitis: A Model For Central Nervous System And Immune System Interaction In Ige-Mediated Allergic Reactions
Published in Husband Alan J., Behaviour and Immunity, 2019
M. Gauci, A.J. Husband, M.G. King
Since the assumption of homogeneity of variance was violated, nonparametric statistics were used to analyse TE activity. A Kruskal-Wallace analysis of variance showed that the difference in TE activity on CD1 between groups was significant (χ2 = 8.903, d.f=2, p < 0.05). Nonparametric multiple comparisons35 revealed that TE activity was greater in Groups 1 and 3 compared with Group 2 (p<0.05) (Fig.l). A similar conclusion was reached for TE activity on CD2 (χ2 = 14.098, d.f. = 2, p<0.001) (Fig.l). Therefore, comparison of TE activity distinguished between groups challenged with allergen (Groups 1 and 3) and control Group 2 (given PBS) on both conditioning days. On test day, a significant difference was also apparent in TE activity between groups (χ2 = 6.321, d.f.=2, p<0.05). Application of Nemenyi’s35 nonparametric multiple comparisons revealed that enzyme activity was greater in Group 3 compared with Group 2 (p<0.05). Unlike the results for CD1 and CD2, however, the difference in TE activity between Group 1 and Group 2 on test day failed to reach significance (Fig.l). The large variance contributed by Group 1 TE activity on test day discussed above may account for this lack of significance.
Approaches to quantitative data analysis
Published in Louis Cohen, Lawrence Manion, Keith Morrison, Research Methods in Education, 2017
Louis Cohen, Lawrence Manion, Keith Morrison
Non-parametric data are those which make no assumptions about the population, usually because the characteristics (numerical parameters) of the population are unknown. Parametric data assume knowledge of the characteristics of the population, in order for inferences to be able to be made securely; inferential statistics are premised on a normal, Gaussian curve of distribution, as, for example, in reading scores, in order to be able to generalize to the wider population (though Wright (2003, p. 128) suggests that normal distributions are actually rare). In practice this distinction means the following: nominal and ordinal data are often considered to be non-parametric, whilst interval and ratio data are often considered to be parametric data (unless, for example, the data are skewed). The distinction is important, as, for the four scales of data, the consideration of which statistical test to use is dependent on the kinds of data: it is often incorrect to apply parametric statistics to non-parametric data, though it is possible to apply non-parametric statistics to parametric data if those data do not conform to the curve of distribution, being skewed or unevenly distributed. Statistics for parametric data tend to be more powerful than those for non-parametric data, though such power is bought at the price of, for example, conformity to the normal curve of distribution and random samples. Non-parametric data are often derived from questionnaires and surveys (though these can also include parametric data), whilst parametric data tend to be derived from experiments and tests (e.g. examination scores).
Interpreting inferential statistics in quantitative research
Published in John Maltby, Glenn A. Williams, Julie McGarry, Liz Day, Research Methods for Nursing and Healthcare, 2014
John Maltby, Glenn A. Williams, Julie McGarry, Liz Day
Statisticians have noted that if scores on a variable (e.g. anxiety among the general population, with few people having high and low levels) show a normal distribution, then we have a potentially powerful statistical tool. This is because we can then begin to be certain about how scores will be distributed for any variable; that is, we can expect many people’s scores to be concentrated in the middle and a few to be concentrated at either end of the scale. However, even if scores are not normally distributed, there are still ways and means of accommodating to this and we will explore this further when we look at the distinction between parametric and non-parametric statistics. By having an awareness of the role of normal or skewed distributions in quantitative studies, nurse researchers can be more confident in comparing differences in samples of patients or looking at relationships between a type of treatment and health outcomes. This confidence has come through being able to examine the data collected and analysed through inferential statistics by having a full understanding of the concept of probability.
The impact of the BEL intervention on levels of motivation, engagement and recovery in people who attend community mental health services
Published in Scandinavian Journal of Occupational Therapy, 2023
Mona Eklund, Kristine Lund, Elisabeth Argentzell
Conventional non-parametric statistics were used; the Mann–Whitney test to analyse differences between independent samples, Wilcoxon’s test for related samples, and Spearman correlations for associations between variables. As a way of analysing change in DC motivation, we applied a method based on standard deviations (SD) [43]. The span for change in motivation was set at one standard deviation in either direction from the mean for the respective motivation variable. Jacobson and Traux [43] preferred two SD, but also stated that one SD could be enough to detect a meaningful change among severely ill groups, for example, persons with psychosis. We thus used the baseline SD for each of the motivation variables to form groups of change. Those who at 16 weeks had a score smaller than the baseline value minus the SD formed a decreased group; those with scores within one SD in either direction formed an unchanged group; and those who had a score larger than the baseline value plus one SD formed an increased group. These groups were used for descriptive purposes and to further explore change in DC motivation. The Jonckheere-Terpstra test was used for the latter purpose. Since the study was underpowered, we found it justified to perform some of the analyses on the two groups together, as further specified below in the result section. The software used for computations was the IBM SPSS version 28 [44].
Adherence to Mediterranean-Style Dietary Pattern and Risk of Bladder Cancer: A Case-Control Study in Iran
Published in Nutrition and Cancer, 2023
Melika Hajjar, Marzieh Pourkerman, Arezoo Rezazadeh, Faisel Yunus, Bahram Rashidkhani
Statistical analysis was performed using SPSS 21.0 software (IBM, NY, USA). The Kolmogorov–Smirnoff test was used to examine the normality of the distribution of the variables. The Student’s t-test was used to compare the mean values of cases and controls, and the Analysis of Variance (ANOVA) to compare means of more than two groups for normally distributed variables. The nonparametric statistics, including the Mann–Whitney U test was used for variables that were not normally distributed. Chi-square test was performed for categorical variables. Age, sex, total energy intake and smoking were determined as covariates and all models were adjusted for these confounders. The variables level of BMI, physical activity, education, dietary supplement use, clinical phenotype (NMIBC and MIBC), family history of BC and other factors were initially included in all models but since the results did not change appreciably, these variables were not included in the final models. MSDPS were analyzed as quartiles. The quartiles were defined based on the distribution in the control population. Binary logistic regression was implemented to estimate odd ratios (ORs) and 95% confident intervals (CIs) adjusted for multiple covariates in different models. All statistical tests were 2-tailed and the level of significance was set at p = 0.05.
The Impact of Occlusion Therapy on Amblyopia Success Outcomes
Published in Journal of Binocular Vision and Ocular Motility, 2022
Initially, a Kolmogorov-Smirnov test was conducted to determine data normality. Given that the data did not meet the assumptions of normality needed for parametric statistical analyses, non-parametric statistics were performed for all analyses. Regression analyses were performed to determine predictive variables of treatment success and outcome VA. All variables included in Tables 1 and 2 were used in constructing the models. Only significant predictors were used in constructing the final regression model. Age at treatment initiation, classification of amblyopia, severity of amblyopia, and treatment duration were analyzed as they were found to be significant in previous literature. Both linear regression and generalized linear models found them insignificant in relation to success and outcome VA in the study population.22,28–33