Univariate selection: The effects of size of correlation, degree of skew, and degree of restriction

Roth

Organizational Research Methods

2001

Pearson product moment correlations are often corrected for statistical artifacts such as range restriction and unreliability. Formulas have long existed to make such corrections. However, other effect size estimates are rarely corrected for these artifacts, in spite of the fact that there is an established mathematical link between the correlation and some effect size estimates. Correlations and other effect sizes are therefore vulnerable to the same artifacts. The authors take a common effect size estimate, the standardized mean difference between two groups, and derive (and reaffirm in one instance) correction formulas suitable for use with this statistic. It is demonstrated how these formulas might substantially increase the precision of estimates and decisions made within organizational research and practice, whenever correction factors can be appropriately estimated.

Section: Correction Of D For Direct Range Restrictionmentioning

confidence: 99%

Correcting the Effect Size of d for Range Restriction and Unreliability

Roth

Organizational Research Methods

2001

Applied Psychological Measurement

“…The violation of particular interest is linearity, because univariate corrections have been found to be sensitive to violations of linearity but robust to violations of homoscedasticity (Greener & Osbum, 1979). Further, although Lawley (1943-44) relaxed distributional assumptions for use of the univariate and multivariate correction formulas, a skewed distribution, which is a common data condition, can affect linearity (Brewer & Hills, 1969).…”

mentioning

confidence: 99%

Explanations for Accuracy of the General Multivariate Formulas in Correcting for Range Restriction

Held

Foley

1994

Univariate and multivariate corrections for range restriction were compared using Navy applicant scores on the Armed Services Vocational Aptitude Battery (ASVAB). Two Navy school ASVAB selector composites were used separately and together to simulate three selection situations for nine selection ratios (SRs). The selectors then were used as predictors. Composite validities (the ASVAB Mechanical Comprehension test was the criterion) were corrected using the univariate and the general multivariate formulas. In general, multivariate corrections were more accurate than univariate corrections, notably when the univariate explicit selection variable was negatively skewed and correction violations (linearity and homoscedasticity) did not offset each other. Multivariate correction accuracy was attributed to the inclusion of variables with adequate distributional properties, the compensatory effects of regression weights, and the related psychometric principle that differentially weighting a large number of correlated predictor variables has little impact on a multiple correlation. Index terms: correction formulas, explicit selection, incidental selection, multivariate correction formulas, restriction in range, validation studies.Since Pearson (1903) developed the correction formulas for restriction in range and Thorndike

“…We have used 3 different feature selection methods (univariate selection [59,60], feature importance [59,[61][62][63], and correlation matrix with heatmap [59,64]) with our domain knowledge to select the 10 most important features from the extended data sets. From univariate selection [60] and feature importance [61][62][63], we obtained 10 important features with their score from each approach (Multimedia Appendices 2 and 3, respectively). We also prepared an important correlation matrix (Multimedia Appendix 4) with heatmap [64] for the features.…”

Section: Feature Selectionmentioning

confidence: 99%

Informing Developmental Milestone Achievement for Children With Autism: Machine Learning Approach

et al. 2021

Background Care for children with autism spectrum disorder (ASD) can be challenging for families and medical care systems. This is especially true in low- and- middle-income countries such as Bangladesh. To improve family–practitioner communication and developmental monitoring of children with ASD, mCARE (Mobile-Based Care for Children with Autism Spectrum Disorder Using Remote Experience Sampling Method) was developed. Within this study, mCARE was used to track child milestone achievement and family sociodemographic assets to inform mCARE feasibility/scalability and family asset–informed practitioner recommendations. Objective The objectives of this paper are threefold. First, it documents how mCARE can be used to monitor child milestone achievement. Second, it demonstrates how advanced machine learning models can inform our understanding of milestone achievement in children with ASD. Third, it describes family/child sociodemographic factors that are associated with earlier milestone achievement in children with ASD (across 5 machine learning models). Methods Using mCARE-collected data, this study assessed milestone achievement in 300 children with ASD from Bangladesh. In this study, we used 4 supervised machine learning algorithms (decision tree, logistic regression, K-nearest neighbor [KNN], and artificial neural network [ANN]) and 1 unsupervised machine learning algorithm (K-means clustering) to build models of milestone achievement based on family/child sociodemographic details. For analyses, the sample was randomly divided in half to train the machine learning models and then their accuracy was estimated based on the other half of the sample. Each model was specified for the following milestones: Brushes teeth, Asks to use the toilet, Urinates in the toilet or potty, and Buttons large buttons. Results This study aimed to find a suitable machine learning algorithm for milestone prediction/achievement for children with ASD using family/child sociodemographic characteristics. For Brushes teeth, the 3 supervised machine learning models met or exceeded an accuracy of 95% with logistic regression, KNN, and ANN as the most robust sociodemographic predictors. For Asks to use toilet, 84.00% accuracy was achieved with the KNN and ANN models. For these models, the family sociodemographic predictors of “family expenditure” and “parents’ age” accounted for most of the model variability. The last 2 parameters, Urinates in toilet or potty and Buttons large buttons, had an accuracy of 91.00% and 76.00%, respectively, in ANN. Overall, the ANN had a higher accuracy (above ~80% on average) among the other algorithms for all the parameters. Across the models and milestones, “family expenditure,” “family size/type,” “living places,” and “parent’s age and occupation” were the most influential family/child sociodemographic factors. Conclusions mCARE was successfully deployed in a low- and middle-income country (ie, Bangladesh), providing parents and care practitioners a mechanism to share detailed information on child milestones achievement. Using advanced modeling techniques this study demonstrates how family/child sociodemographic elements can inform child milestone achievement. Specifically, families with fewer sociodemographic resources reported later milestone attainment. Developmental science theories highlight how family/systems can directly influence child development and this study provides a clear link between family resources and child developmental progress. Clinical implications for this work could include supporting the larger family system to improve child milestone achievement.