Explore chapters and articles related to this topic
Nonparametric Methods
Published in Marcello Pagano, Kimberlee Gauvreau, Heather Mattie, Principles of Biostatistics, 2022
Marcello Pagano, Kimberlee Gauvreau, Heather Mattie
The Kruskal-Wallis test is an extension of the Wilcoxon rank sum test which can be used to compare three or more independent populations. It is the nonparametric counterpart to the one-way analysis of variance, but does not require that the underlying populations be normally distributed or that their variances be equal. Like the Wilcoxon rank sum test, it does assume that the k populations being compared all have the same basic shape. The Kruskal-Wallis test evaluates the null hypothesis that the medians of the k populations are identical. The alternative hypothesis is that at least one of the population medians differs from one of the others.
New Statistical Designs for Clinical Trials of Immunomodulating Agents
Published in Thomas F. Kresina, Immune Modulating Agents, 2020
Whereas cohorts of three to six patients may be appropriate for roughly determining the MTD, there is no reason to believe that they are sufficient for determining the relationship between dose and biological response. If one has treated n patients in each of k dose levels, one question of interest is whether biological responses for all dose groups are equivalent A widely used nonparametric test for this purpose is the Kruskal-Wallis test. Nonparametric tests are useful for continuous endpoints that are not normally distributed. Biological endpoints are often of this type. The computations involved in the Kruskal-Wallis test are described in statistical texts [4].
Mathematical Concepts and Statistics
Published in Sarah Armstrong, Barry Clifton, Lionel Davis, Primary FRCA in a Box, 2019
Sarah Armstrong, Barry Clifton, Lionel Davis
With more than two groups, multiple comparisons are made and there is a 5% chance of obtaining statistical significance by chance alone if the above tests are used For normally distributed data, therefore, analysis of variance (ANOVA) is used insteadFor non-normally distributed data of more than two groups, the Kruskal–Wallis test is used
The effect of robot-assisted walking in different modalities on cardiorespiratory responses and energy consumption in patients with subacute stroke
Published in Neurological Research, 2023
Ahmet Mert Sayın, Neslihan Duruturk, Birol Balaban, Süleyman Korkusuz
The data to be obtained in the study were analyzed using the SPSS version 25 package statistical computer program. All data were reported as mean and standard deviations. It was performed with ANOVA, a repeated measure, to assess differences between walking tests, including (walking conditions) and (stroke group, control group) subject factors. Walking conditions and group were considered the main factors in this analysis, so comparison between walking conditions was made by including all subjects in the two groups. Group comparisons were made by including all the group walks. The level of significance for the ANOVA analysis was determined as p ≤ 0.05. If ANOVA gives non-parametric results, Kruskal–Wallis test is applied. Considering the cardiopulmonary health and parameters of the patient after robotic rehabilitation in subacute stroke individuals, the number of samples was determined for each group in order to observe the effect of robotic rehabilitation with 80% power and 5% error, and the different BWS and GF modalities in it, on the cardiopulmonary functions of individuals with subacute stroke. It was calculated as 10 participants for each group [13].
The potential value of serum GP73 in the ancillary diagnosis and grading of liver cirrhosis
Published in Scandinavian Journal of Clinical and Laboratory Investigation, 2023
Chen Hui-ling, Huang Kang-ming, Zhao Yu, Deng Yin-han, Du Huang, Xiao Shu-ping, Chen Hong-bin
SPSS 23.0 software was used to statistically process the data. The one-sample Kolmogorov–Smirnov test was used to analyze whether each quantitative variable conformed to a normal distribution (if p > 0.05, it conformed to a normal distribution), and the quantitative data conforming to a normal distribution are presented as the mean and standard deviation (x̅ ± s). The t test was used for comparisons between the two groups, and one-way analysis of variance (ANOVA) was used for multisample single-factor comparisons. The quantitative data that did not conform to a normal distribution are presented as the median and interquartile range [M (P25–P75)]. The Mann–Whitney U test was used to compare two groups, and Kruskal–Wallis test was used to compare multiple groups. The qualitative data are presented as frequencies and rates, and differences between groups were analyzed by the chi-square (χ2) test or Fisher’s exact method.
Prediction of a reliable method for the estimation of central corneal thickness in diabetic patients with and without diabetic retinopathy
Published in Expert Review of Ophthalmology, 2022
Anita Syla Lokaj, Gazmend Kaçaniku, Kelmend Spahiu, Faruk Semiz
In the case of uncorrected vision, group A patients had the most impaired vision with a mean of 0.19 (SD ± 0.16, range 0.04–0.7), while there was no significance in the control group (mean 0.51; SD ± 0.25, range 0.1–1.0). Using the Kruskal–Wallis test, the study obtained a significant difference between groups A and B (P < 0.001) and groups A and C (P < 0.001). Similarly, in the case of corrected vision, group A patients had the most impaired vision even after correction with a mean of 0.32 (SD ± 0.24, range 0.05–1.0), while the vision after correction of those in groups B (mean 0.71; SD 0.21, range 0.3–1) and C showed no significant difference (mean 0.71; SD ± 0.22, range 0.3–1). Using the Kruskal–Wallis test, a significant difference was observed in the degree of vision after the best correction between groups A and B (P < 0.001) and groups A and C (P = 0.0301).