PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 12619534

Minimizing sample size when using exploratory factor analysis for measurement.

Kathryn G Sapnas·Richard A Zeller
View original source ↗Export metadata ↓

Abstract

Traditional protocol for the determination of an adequate sample size is power analysis. Such a protocol is not useful when the primary hypothesis focuses on psychometric measurement properties. Traditional psychometrics advises that there should be 10 respondents per item. Both hypothetical and real research examples illustrate the usefulness of subsample analysis in determining that a sample size of at least 50 and not more than 100 subjects is adequate to represent and evaluate the psychometric properties of measures of social constructs. The "10 respondents per item" advice builds a sample size disincentive into the research design; it also represents "sample size overkill." Sample-size overkill occurs when the research design specifies a number of cases needed, which is in excess of the number actually needed for a desired inference.

Explore related subjects

Computer SimulationFactor Analysis, StatisticalHumansLikelihood FunctionsMonte Carlo MethodPsychometricsSample SizeEvaluation StudyJournal Article

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Kathryn G Sapnas, Richard A Zeller. 2002. Minimizing sample size when using exploratory factor analysis for measurement.. https://doi.org/10.1891/jnum.10.2.135.52552

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2025

Generating correlated data for omics simulation.

Simulation of realistic omics data is a key input for benchmarking studies that help users obtain optimal computational pipelines. Omics data involves large numbers of measured features on each sample and these measures are generally correlated with each other. However, simulation too often ignores these correlations, perhaps due to computational and statistical hurdles of doing so. To alleviate this, we describe three approaches for generating omics-scale data with correlated measures which mimic real datasets. These approaches are all based on a Gaussian copula approach with a covariance matrix that decomposes into a diagonal part and a low-rank part. This decomposition allows for extremely efficient simulation, overcoming a hurdle for adoption of past methods. We use these approaches to demonstrate the importance of including correlation in two benchmarking applications. First, we show that variance of results from the popular DESeq2 method increases when dependence is included. Second, we demonstrate that CYCLOPS, a method for inferring circadian time of collection from transcriptomics, improves in performance when given gene-gene dependencies in some circumstances. We provide an R package, dependentsimr, that has efficient implementations of these methods and can generate dependent data with arbitrary marginal distributions, including discrete (binary, ordered categorical, Poisson, negative binomial), continuous (normal), or with an empirical distribution.

Computer Simulation↗
PubMed2017

Addressing current challenges in cancer immunotherapy with mathematical and computational modelling.

The goal of cancer immunotherapy is to boost a patient's immune response to a tumour. Yet, the design of an effective immunotherapy is complicated by various factors, including a potentially immunosuppressive tumour microenvironment, immune-modulating effects of conventional treatments and therapy-related toxicities. These complexities can be incorporated into mathematical and computational models of cancer immunotherapy that can then be used to aid in rational therapy design. In this review, we survey modelling approaches under the umbrella of the major challenges facing immunotherapy development, which encompass tumour classification, optimal treatment scheduling and combination therapy design. Although overlapping, each challenge has presented unique opportunities for modellers to make contributions using analytical and numerical analysis of model outcomes, as well as optimization algorithms. We discuss several examples of models that have grown in complexity as more biological information has become available, showcasing how model development is a dynamic process interlinked with the rapid advances in tumour-immune biology. We conclude the review with recommendations for modellers both with respect to methodology and biological direction that might help keep modellers at the forefront of cancer immunotherapy development.

Computer Simulation↗
PubMed2004

Use of protein charge ladders to study electrostatic interactions during protein ultrafiltration.

Recent studies have demonstrated the importance of electrostatic interactions in membrane systems, but there is still controversy about the underlying phenomena. Protein charge ladders, consisting of a set of chemical derivatives of a given protein that differ by single charge groups, were used to quantify the electrostatic interactions during protein ultrafiltration. Myoglobin charge ladders were generated by acylation, with the different derivatives analyzed simultaneously by capillary electrophoresis. Filtration experiments were performed using polyethersulfone and composite regenerated cellulose membranes, with the membrane charge determined from the streaming potential. As expected, the rejection increased as the protein became more heavily charged due to the increase in electrostatic repulsion. However, the transmission of the weakly charged myoglobin species increased dramatically at very low ionic strength. This increase in transmission was attributed to a shift in pH within the pore caused by hydrogen ion partitioning into the charged membrane. The sieving data were in good agreement with theoretical calculations accounting for the effects of this pH shift on the electrostatic interactions.

Computer Simulation↗