Related Experiment Video
Updated: Jan 20, 2026
06:03
Studying Visual Awareness and Motion-Induced Blindness
Published on: April 30, 2023
7.7K
Overdispersion models for correlated multinomial data: Applications to blinding assessment
V Landsman1,2, D Landsman3, C S Li4
1Institute for Work and Health, Toronto, Ontario, Canada.
Statistics in Medicine
|August 29, 2019
Summary
This study introduces a multinomial overdispersion model for correlated data, estimating a Global Blinding Index using generalized estimating equations (GEE). GEE estimators performed well, even with few clusters, outperforming other methods in clinical trials and meta-analyses.
Area of Science:
- Statistics
- Biostatistics
- Correlated Data Analysis
Background:
- Overdispersion models are well-studied for normal and binomial data but less so for multinomial data.
- Correlated multinomial data present unique analytical challenges in clinical trials and meta-analyses.
Purpose of the Study:
- To develop and evaluate a multinomial overdispersion model for correlated data.
- To introduce and estimate a Global Blinding Index using generalized estimating equations (GEE).
- To assess the performance of GEE estimators under various conditions, including small cluster numbers and high intraclass correlation.
Main Methods:
- Specification of a multinomial overdispersion model defining the first two moments of the outcome.
- Application of generalized estimating equations (GEE) for parameter estimation.
- Simulation studies and comparisons with inverse-variance weighted and maximum-likelihood estimators.
Main Results:
- GEE estimators for the Global Blinding Index demonstrated satisfactory performance, even with a small number of clusters.
- Inverse-variance weighted estimators performed poorly, particularly with high intraclass correlation coefficients.
- The proposed GEE method proved robust across different simulation scenarios.
Conclusions:
- The developed multinomial overdispersion model and GEE approach are effective for analyzing clustered multinomial data.
- The GEE method offers a reliable alternative to other estimators, especially in challenging data situations.
- Findings provide practical guidance for analyzing clustered multinomial data in clinical trials and meta-analyses.
Related Concept Videos
02:07Blind Procedures
13.1K
Ideally, the people who observe and record the children’s behavior are unaware of who was assigned to the experimental or control group, in order to control for experimenter bias. Experimenter bias refers to the possibility that a researcher’s expectations might skew the results of the study. Remember, conducting an experiment requires a lot of planning, and the people involved in the research project have a vested interest in supporting their hypotheses. If the observers knew which...
13.1K
06:03Motion-induced Blindness
7.7K
Source: Laboratory of Jonathan Flombaum—Johns Hopkins University
One thing becomes very salient after basic exposure to the science of visual perception and sensation: what people see is a creation of the brain. As a result people may fail to see things, see things that are not there, or see things in a distorted way.
To distinguish between physical reality and what people perceive, scientists use the term awareness to refer to what people perceive. To study awareness, vision scientists...
One thing becomes very salient after basic exposure to the science of visual perception and sensation: what people see is a creation of the brain. As a result people may fail to see things, see things that are not there, or see things in a distorted way.
To distinguish between physical reality and what people perceive, scientists use the term awareness to refer to what people perceive. To study awareness, vision scientists...
7.7K
01:23Correlation of Experimental Data
480
Dimensional analysis simplifies complex physical problems and guides experimental investigations, but it does not provide complete solutions. It identifies the dimensionless groups that influence a phenomenon, but experimental data is needed to establish the specific relationships and validate theoretical predictions.
For example, a spherical particle moving through a viscous fluid experiences drag. Dimensional analysis shows that the drag force depends on the particle's diameter, velocity,...
For example, a spherical particle moving through a viscous fluid experiences drag. Dimensional analysis shows that the drag force depends on the particle's diameter, velocity,...
480
14:51Inattentional Blindness
14.6K
Source: Laboratory of Jonathan Flombaum—Johns Hopkins University
We generally think that we see things pretty well if they are close by and right in front of us. But do we? We know that visual attention is a property of the human brain that controls what parts of the visual world we process, and how effectively. Limited attention means that we can't process everything at once, it turns out, even things that might be right in front of us.
In the 1960s, the renowned cognitive psychologist...
We generally think that we see things pretty well if they are close by and right in front of us. But do we? We know that visual attention is a property of the human brain that controls what parts of the visual world we process, and how effectively. Limited attention means that we can't process everything at once, it turns out, even things that might be right in front of us.
In the 1960s, the renowned cognitive psychologist...
14.6K
02:20Correlations
35.8K
Correlation means that there is a relationship between two or more variables (such as ice cream consumption and crime), but this relationship does not necessarily imply cause and effect. When two variables are correlated, it simply means that as one variable changes, so does the other. We can measure correlation by calculating a statistic known as a correlation coefficient. A correlation coefficient is a number from -1 to +1 that indicates the strength and direction of the relationship between...
35.8K
02:31Diffuse Correlation Spectroscopy to Assess Cerebral Blood Flow in a Mouse Model
721
In this video, we demonstrate the diffuse correlation spectroscopy technique to measure cerebral blood flow in a mouse model of mild traumatic brain...
721