Chapter 21 Sample Size for Multiple Groups Model
21.1 What is the Sample Size for Multiple Groups Model?
The Sample Size for Multiple Groups Model leverages naturally-occurring biological groupings and group-wise correlation structure to plan a sampling scheme to efficiently investigate whether a free-ranging population is disease free.
21.2 What Question Does it Answer?
Question 1. How many individuals should I test without finding a positive case to make inferences about population-scale disease freedom? The app shows you exactly how many negatives you need to accumulate from each sub administrative area to declare the area free from disease with high confidence.
21.3 Output Details
- Sample size: The number of samples needed from a population to ensure there is a high probability that disease prevalence is at or below 1% or 2%.
21.4 Abbreviated Tutorial
- Click the ‘Run Sample Size for Multiple Groups’ button.
- Once the app opens, click through the app to choose your model parameters (see below).
- Run the model with those desired population parameters.
- Look at the graph to see the number of animals that need to be tested without finding a positive case to ensure there is a high probability that disease prevalence is at or below 1% or 2%.
21.5 Parameters Needed to Execute the Model
Model type: Select ‘Sample Size for Multiple Groups Model’ from the drop-down list, then click the URL link to open and interact with the live software.
Population size: Total number of animals in the population of interest. An integer between 100 and 10,000,000.
Average cluster size: Average cluster size in the population of interest. An integer between 1 (1 animal per cluster) and m (m animals per cluster).
Average correlation: Correlation in disease status between individuals in a cluster. A decimal between 0 (disease status among members is independent) and 0.995 (disease status is nearly perfectly correlated among members).
Sampling scheme: Scheme that you intend to use for sampling.
- Simple Random Sampling: Each subject has an equal chance of being sampled.
Sensitivity of diagnostic test: Sensitivity of the diagnostic test. A decimal between 0 (not sensitive) and 0.999 (perfect sensitivity).
Number of simulations: Number of simulations you would like to conduct. An integer between 1 (1 simulation) and s (s simulations).
21.6 Details on the Theory
Booth JG, Hanley BJ, Hodel FH, Jennelle CS, Guinness J, Them CE, Mitchell CI, Ahmed MS, Schuler KL. 2024. Sample Size for Estimating Disease Prevalence in Free-Ranging Wildlife Populations: A Bayesian Modeling Approach. JABES 29, 438–454. https://doi.org/10.1007/s13253-023-00578-7.
21.7 Code
Code is in Hanley BJ, Booth JG, Hodel FH, JCG Bloodgood, JP Dion, Van de Berg S, Gonzalez-Crespo C, Huang Y, Wang J, Miller LA, Hollingshead NA, Peaslee JL, Schuler KL. 2024. Sample size calculator for declaring a population free of infectious disease (Version 1) [Software]. Cornell University Library eCommons. https://doi.org/10.7298/ka5p-bj90