Explore chapters and articles related to this topic
Non-Parametric Methods for the Analysis of Repeatedly Measured Data
Published in Gueorguieva Ralitza, Statistical Methods in Psychiatry and Related Fields, 2017
When the 10 observations are ranked (−1, −1, 2, 3, 3, 3, 5.5, 7, 10, 33), the smallest in value is assigned a rank of 1, the second smallest is assigned a rank of 2, and ranking continues until the largest in value is assigned a rank of 10. However, when there are ties, they are assigned the mid-rank. In the example, the smallest value is −1 and there are two observations with this value. Hence, they are assigned the mid-rank of 1 and 2 which is 1.5. The next smallest value is 2 and it is assigned a rank of 3 since there are two observations with smaller values. Then comes 3 which represents a three-way tie, hence, each of these observations is assigned a rank of 5 (the mid-rank of 4, 5, and 6). Ranking continues until the largest observation is assigned a rank of 10 because there are 10 observations. Once ranks are assigned, test statistics are calculated based on these ranks.
Quantitative Methods for Analyzing Experimental Studies in Patient Ergonomics Research
Published in Richard J. Holden, Rupa S. Valdez, The Patient Factor, 2021
Kapil Chalil Madathil, Joel S. Greenstein
Analysis of variance (ANOVA) is used to analyze data with more than two conditions. Nonparametric tests are used to analyze data with non-normal distributions. Nonparametric tests often employ a ranking technique, which generates a data set with the higher scores represented by large ranks and the lower scores by small ranks. The most commonly used nonparametric tests include the Mann–Whitney test, the Wilcoxon signed-rank test, Friedman’s test, and the Kruskal–Wallis test. Cohen et al. (2002) and Cumming and Calin-Jageman (2017) present the details of the statistical analyses that have been introduced here, as well as those for independent t-tests, ANOVA, and nonparametric tests.
Feature extraction and qualification
Published in Ruijiang Li, Lei Xing, Sandy Napel, Daniel L. Rubin, Radiomics and Radiogenomics, 2019
Ranking and feature subset selection are two main operations that can be implemented for filter, wrapper, and embedded methods. A feature ranking based on individual importance of the features will be evaluated by these methods, then the top features that perform well for prediction task could be selected. In some cases, we first get the ranking list and do the subset selection based on the list. While in other cases, we may only get the subset features from the method. Usually, subset selection is supervised, while obtaining a ranking might be supervised or not (Roffo et al. 2015).
A step towards the application of an artificial intelligence model in the prediction of intradialytic complications
Published in Alexandria Journal of Medicine, 2022
Ahmed Mustafa Elbasha, Yasmine Salah Naga, Mai Othman, Nancy Diaa Moussa, Hala Sadik Elwakil
The most effective 26 features were selected based on filter feature selection technique. It is based on ranking the features according to their usefulness in the prediction and evaluating the importance of features based on the properties of data. Then, the output of the selected features is applied to the machine learning algorithm. The filter approach uses a variety of methods, such as Pearson’s correlation coefficient, and mutual information. Pearson’s determines the correlation between the features and the output class, while mutual information shows the amount of information between the feature, and the output class. For feature selection, we first removed the correlated input features, assuming that the features were highly correlated if their correlation coefficient is greater than 0.75. In this step, we removed seven features. Second, we chose only features that are highly correlated with the output, the feature that had no impact on the output was removed. If the correlation coefficient was between −0.05 and 0.05, we neglected it. In this step we removed 15 features. Third, we chose features that had high mutual information with the output. In this step, we removed two features.
Acteoside relieves mesangial cell injury by regulating Th22 cell chemotaxis and proliferation in IgA nephropathy
Published in Renal Failure, 2018
Lu Gan, Xiaozhao Li, Mengyuan Zhu, Chen Chen, Huimin Luo, Qiaoling Zhou
Data were expressed as mean ± standard deviation (SD) or median with minimum and maximum value. Data comparisons were performed using Kruskal–Wallis one-way analysis or Mann–Whitney U-test of variable for ranking. Comparison of urinary protein excretion and Th22 cell expression before and after drug therapy was performed by using repeated measurement ANOVA. Variables in chemotaxis assay and co-culture assay were compared using Students’s t-test or the Wilcoxon signed-rank test. The correlations among variables were determined by calculating the Spearman rank correlation coefficients. p < .05 was set as the statistical significance. The exact p values were expressed as p < .001 if the p values was less than 1 × 1.0−3. Statistical analyses were performed by using SPSS 19.0 software package (International Business Machines Corporation, Chicago, IL, USA).
Impacts of goal setting on engagement and rehabilitation outcomes following acquired brain injury: a systematic review of reviews
Published in Disability and Rehabilitation, 2020
Katri Knutti, Anita Björklund Carlstedt, Rieke Clasen, Dido Green
The level of evidence of single SRs was evaluated using the Oxford Centre for Evidence Based Medicine Levels of Evidence (OCEBM) [29] (see Tables 2 and 3). It is a ranking system to consider the extent of the evidence according to the quality and results of clinical trials and studies. In the OCEBM, level 1 reflects the most robust studies such as SRs of RCTs of high quality and level 5 refers to the lowest quality such as expert opinion without explicit critical appraisal [29].