Estimating Subgroup Differences in Staffing Research When the Selection Mechanism Is Unknown: A Response to Li’s Case IV Correction

2018 ◽  
Vol 23 (2) ◽  
pp. 367-384 ◽  
Author(s):  
Dustin A. Fife ◽  
Jorge Mendoza ◽  
Eric Day ◽  
Robert Terry
Keyword(s):  
Monte Carlo Simulation ◽  
Monte Carlo ◽  
Group Membership ◽  
Group Differences ◽  
Selection Mechanism ◽  
Range Restriction ◽  
Subgroup Differences ◽  
Cohen’S D ◽  
Selection Decisions ◽  
Cohen's D

When estimating subgroup differences on incumbents, range restriction may bias estimates. Bobko, Roth, and Bobko recognized this problem and developed a Case II and Case III correction for Cohen’s d. Subsequently, Li developed a Case IV correction, which seeks to estimate group differences on a predictor using only incumbent data but must assume that group membership (e.g., ethnicity) plays no role in selection decisions. In this paper, we extend Li’s correction and relax this assumption. In addition, this new correction allows for the estimation of subgroup differences on both the criterion and predictor. Using Monte Carlo simulation, we study the performance of both estimators under situations where Li’s assumptions are violated and demonstrate that this new procedure almost always outperforms Li’s Case IV correction and does so with greater precision. We also provide R code to assist applied researchers in using these corrections.

scholarly journals An Improved Model for Evaluating Change in Randomized Pretest, Posttest, Follow-Up Designs

Methodology ◽  
2012 ◽  
Vol 8 (3) ◽  
pp. 97-103 ◽  
Author(s):  
Constance A. Mara ◽  
Robert A. Cribbie ◽  
David B. Flora ◽  
Cathy LaBrish ◽  
Laura Mills ◽  
...  
Keyword(s):  
Monte Carlo Simulation ◽  
Monte Carlo ◽  
Group Differences ◽  
Primary Research ◽  
Control Groups ◽  
Proposed Model ◽  
Research Questions ◽  
Improved Model ◽  
And Control

Randomized pretest, posttest, follow-up (RPPF) designs are often used for evaluating the effectiveness of an intervention. These designs typically address two primary research questions: (1) Do the treatment and control groups differ in the amount of change from pretest to posttest? and (2) Do the treatment and control groups differ in the amount of change from posttest to follow-up? This study presents a model for answering these questions and compares it to recently proposed models for analyzing RPPF designs due to Mun, von Eye, and White (2009) using Monte Carlo simulation. The proposed model provides increased power over previous models for evaluating group differences in RPPF designs.

scholarly journals Statistical Permutation Test Reveals Progressive and Region-Specific Iron Accumulation in the Thalami of Children with Aspartylglucosaminuria

2020 ◽  
Vol 10 (10) ◽  
pp. 677
Author(s):  
Viljami Sairanen ◽  
Anna Tokola ◽  
Ritva Tikkanen ◽  
Minna Laine ◽  
Taina Autti
Keyword(s):  
Multivariate Regression ◽  
Permutation Test ◽  
Test Group ◽  
Iron Accumulation ◽  
Group Differences ◽  
P Value ◽  
Lysosomal Storage ◽  
Cohen’S D ◽  
Cohen's D ◽  
Eventual Death

Aspartylglucosaminuria (AGU) is a rare lysosomal storage disorder causing developmental delay, intellectual disability, and eventual death. A distinct feature in AGU is iron accumulation within the thalamus. Our aim is to demonstrate that susceptibility-weighted images (SWI) could be used as an MRI biomarker to evaluate the response within the AGU population to newly evolving treatments. SWI from 16 patients with AGU and 16 age-matched controls were used in the analysis. Thalamic volume with an iron accumulation was identified using a permutation test. Group differences were investigated for both the complete thalamus and the iron accumulation regions. Group-wise age correlation within these volumes were assessed with analysis of variance and multivariate regression. We found a statistically significant and large difference (p-value = 0.01, Cohen’s D = 0.97) for the whole thalamus comparison and an even greater difference in the iron accumulation regions (p-value < 0.01, Cohen’s D = 3.52). Furthermore, we found strong evidence for iron accumulation as a linear function of age with R2 = 0.65 only for AGU. The statistical analysis of SWI provides tools for assessing the degree of iron accumulation. This method could be used to study the response to treatments, in that a successful treatment would be expected to result in a decline in iron accumulation.

scholarly journals Effect Size Guidelines, Sample Size Calculations, and Statistical Power in Gerontology

2019 ◽  
Vol 3 (4) ◽  
Author(s):  
Christopher R Brydges
Keyword(s):  
Individual Differences ◽  
Sample Size ◽  
Effect Size ◽  
Effect Sizes ◽  
Group Differences ◽  
Cohen’S D ◽  
Pearson's R ◽  
Sample Size Calculations ◽  
Cohen's D ◽  
Pearson’S R

Abstract Background and Objectives Researchers typically use Cohen’s guidelines of Pearson’s r = .10, .30, and .50, and Cohen’s d = 0.20, 0.50, and 0.80 to interpret observed effect sizes as small, medium, or large, respectively. However, these guidelines were not based on quantitative estimates and are only recommended if field-specific estimates are unknown. This study investigated the distribution of effect sizes in both individual differences research and group differences research in gerontology to provide estimates of effect sizes in the field. Research Design and Methods Effect sizes (Pearson’s r, Cohen’s d, and Hedges’ g) were extracted from meta-analyses published in 10 top-ranked gerontology journals. The 25th, 50th, and 75th percentile ranks were calculated for Pearson’s r (individual differences) and Cohen’s d or Hedges’ g (group differences) values as indicators of small, medium, and large effects. A priori power analyses were conducted for sample size calculations given the observed effect size estimates. Results Effect sizes of Pearson’s r = .12, .20, and .32 for individual differences research and Hedges’ g = 0.16, 0.38, and 0.76 for group differences research were interpreted as small, medium, and large effects in gerontology. Discussion and Implications Cohen’s guidelines appear to overestimate effect sizes in gerontology. Researchers are encouraged to use Pearson’s r = .10, .20, and .30, and Cohen’s d or Hedges’ g = 0.15, 0.40, and 0.75 to interpret small, medium, and large effects in gerontology, and recruit larger samples.

scholarly journals On the Reproducibility and Interpretability of Group-Level Task-fMRI Results

2020 ◽  
Author(s):  
Alex Bowring
Keyword(s):  
Monte Carlo ◽  
Monte Carlo Simulations ◽  
Effect Size ◽  
Fmri Data ◽  
Group Level ◽  
Confidence Sets ◽  
Cohen’S D ◽  
3D Monte Carlo ◽  
Cohen's D ◽  
The Brain

In this thesis, we aim to address two topical issues at the forefront of task-based functional magnetic resonance imaging (fMRI). The first of these is a growing apprehension within the field about the reproducibility of findings that make up the neuroimaging literature. To confront this, we assess how the choice of software package for analyzing fMRI data can impact the final group-level results of a neuroimaging study. We reanalyze data from three published task-fMRI studies within the three most widely-used neuroimaging software packages -- AFNI, FSL, and SPM -- and then apply a range of comparison methods to gauge the scale of variability across the results. While qualitatively we find similarities, our quantitative assessment methods discover considerable differences between the final statistical images obtained with each package. Ultimately, we conclude that exceedingly weak effects may not generalize across fMRI analysis software. In the second part of this work we shift our attention to the analytical methods applied for fMRI inference. Here, we seek to overcome limitations with the traditional statistical approach, where for sufficiently large data sizes current methods determine universal activation across the brain, rendering the results as uninterpretable. We extend on a method proposed by (Sommerfeld et al., 2018; SSS) to develop spatial Confidence Sets (CSs) on clusters found in thresholded raw blood-oxygen-level-dependent (BOLD) effect size maps. The CSs give statements on the locations where raw effect sizes exceed, and fall short of, a purposeful non-zero threshold. We propose several theoretical and practical implementation advancements to the original method formulated in SSS, delivering a procedure with superior performance in sample sizes as low as N = 60. We validate the method with 3D Monte Carlo simulations that resemble fMRI data. We then compute CSs for the Human Connectome Project (HCP) working memory task contrast images, illustrating the brain regions that show a reliable %BOLD for a given %BOLD threshold. In the final part of this thesis, we develop the CSs to operate on standardized Cohen's d effect size images. We derive the statistical properties of the Cohen's d estimator to motivate three algorithms for computing Cohen's d CSs, including a novel method based on normalizing the distribution of Cohen's d. With intensive 3D Monte Carlo simulations, we find that two of these methods can be effectively applied to fMRI data. We compute Cohen's d CSs on the HCP data, and by comparing the CSs with results obtained from a standard testing procedure, exemplify the improved localization of effects that can be gained by using the Confidence Sets.

The Utility of Personnel Selection Decisions

2018 ◽  
Vol 17 (4) ◽  
pp. 172-182 ◽  
Author(s):  
Jisoo Ock ◽  
Frederick L. Oswald
Keyword(s):  
Monte Carlo Simulation ◽  
Monte Carlo ◽  
Model Selection ◽  
Personnel Selection ◽  
Selection Models ◽  
Compensatory Model ◽  
Hurdle Models ◽  
Selection Decisions ◽  
Cost Efficient ◽  
The Cost

Abstract. Compensatory selection is generally more reliable than multiple-hurdle selection. Yet, practitioners may lean toward multiple-hurdle models, because administering an entire predictor battery to every applicant can be time-consuming, labor-intensive, and costly. Using Monte Carlo simulation, we considered some specific cases to illustrate, in terms of selection utility and the cost-reliability tradeoff between compensatory and multiple-hurdle selection models. Results showed that compensatory model selection produced a higher level of expected criterion performance in the selected applicant subgroup, and a higher overall selection utility in most conditions. The simulation provides researchers and practitioners with a practical illustration of the tradeoff between reliable (compensatory) versus cost-efficient (multiple-hurdle) selection models – one that can inspire the exploration of other scenarios and tradeoffs.

scholarly journals Effect Size Interpretation, Sample Size Calculation, and Statistical Power in Gerontology

2019 ◽  
Author(s):  
Christopher Brydges
Keyword(s):  
Individual Differences ◽  
Sample Size ◽  
Effect Size ◽  
Sample Size Calculation ◽  
Effect Sizes ◽  
Group Differences ◽  
Cohen’S D ◽  
Pearson's R ◽  
Cohen's D ◽  
Pearson’S R

Background and Objectives: Researchers typically use Cohen’s guidelines of Pearson’s r = .10, .30, and .50, and Cohen’s d = 0.20, 0.50, and 0.80 to interpret observed effect sizes as small, medium, or large, respectively. However, these guidelines were not based on quantitative estimates, and are only recommended if field-specific estimates are unknown. The current study investigated the distribution of effect sizes in both individual differences research and group differences research in gerontology to provide estimates of effect sizes in the field.Research Design and Methods: Effect sizes (Pearson’s r, Cohen’s d, and Hedges’ g) were extracted from meta-analyses published in ten top-ranked gerontology journals. The 25th, 50th, and 75th percentile ranks were calculated for Pearson’s r (individual differences) and Cohen’s d or Hedges’ g (group differences) values as indicators of small, medium, and large effects. A priori power analyses were conducted for sample size calculations given the observed effect size estimates.Results: Effect sizes of Pearson’s r = .12, .20, and .32 for individual differences research and Hedges’ g = 0.16, 0.38, and 0.76 for group differences research were interpreted as small, medium, and large effects in gerontology. Discussion and Implications: Cohen’s guidelines appear to overestimate effect sizes in gerontology. Researchers are encouraged to use Pearson’s r = .10, .20, and .30, and Cohen’s d or Hedges’ g = 0.15, 0.40, and 0.75 to interpret small, medium, and large effects in gerontology, and recruit larger samples.

Electron Scattering in Solids

1990 ◽  
Vol 48 (2) ◽  
pp. 4-5
Author(s):  
Ryuichi Shimizu ◽  
Ze-Jun Ding
Keyword(s):  
Monte Carlo Simulation ◽  
Monte Carlo ◽  
Angular Distribution ◽  
Elastic Scattering ◽  
Electron Scattering ◽  
Cross Sections ◽  
Expansion Method ◽  
Electron Excitation ◽  
Partial Wave Expansion ◽  
Backscattered Electrons

Monte Carlo simulation has been becoming most powerful tool to describe the electron scattering in solids, leading to more comprehensive understanding of the complicated mechanism of generation of various types of signals for microbeam analysis.The present paper proposes a practical model for the Monte Carlo simulation of scattering processes of a penetrating electron and the generation of the slow secondaries in solids. The model is based on the combined use of Gryzinski’s inner-shell electron excitation function and the dielectric function for taking into account the valence electron contribution in inelastic scattering processes, while the cross-sections derived by partial wave expansion method are used for describing elastic scattering processes. An improvement of the use of this elastic scattering cross-section can be seen in the success to describe the anisotropy of angular distribution of elastically backscattered electrons from Au in low energy region, shown in Fig.l. Fig.l(a) shows the elastic cross-sections of 600 eV electron for single Au-atom, clearly indicating that the angular distribution is no more smooth as expected from Rutherford scattering formula, but has the socalled lobes appearing at the large scattering angle.

Determination of Stoichiometry of GaAs Component in (Gaas)Ge Epitaxially Grown Thin Films by Electron Microprobe X-Ray Analysis

1990 ◽  
Vol 48 (2) ◽  
pp. 264-265
Author(s):  
D. R. Liu ◽  
S. S. Shinozaki ◽  
R. J. Baird
Keyword(s):  
Thin Film ◽  
Thin Films ◽  
Monte Carlo Simulation ◽  
Monte Carlo ◽  
Compound Semiconductors ◽  
Energy Dispersive Analysis ◽  
X Ray ◽  
Metastable Alloys ◽  
Lower Voltage

The epitaxially grown (GaAs)Ge thin film has been arousing much interest because it is one of metastable alloys of III-V compound semiconductors with germanium and a possible candidate in optoelectronic applications. It is important to be able to accurately determine the composition of the film, particularly whether or not the GaAs component is in stoichiometry, but x-ray energy dispersive analysis (EDS) cannot meet this need. The thickness of the film is usually about 0.5-1.5 μm. If Kα peaks are used for quantification, the accelerating voltage must be more than 10 kV in order for these peaks to be excited. Under this voltage, the generation depth of x-ray photons approaches 1 μm, as evidenced by a Monte Carlo simulation and actual x-ray intensity measurement as discussed below. If a lower voltage is used to reduce the generation depth, their L peaks have to be used. But these L peaks actually are merged as one big hump simply because the atomic numbers of these three elements are relatively small and close together, and the EDS energy resolution is limited.

Sign in / Sign up

Export Citation Format

Share Document