U.S. flag

An official website of the United States government, Department of Justice.

Just Science Podcast: Special Release: Just Postmortem Interval Estimation Research

NCJ Number
251041
Date Published
July 2017
Author(s)
Jeffrey Wells; Lynn LaMotte
Agencies
NIJ-Sponsored
Publication Type
Research (Applied/Empirical), Report (Technical Assistance), Report (Study/Research), Report (Grant Sponsored), Program/Project Description, Interview
Grant Number(s)
2016-MU-BX-K110
Annotation

This is the online audio of the NIJ-sponsored podcast of the Just Science interview with Dr. Jeffrey Wells and Dr. Lynn LaMotte regarding their work in providing adequate statistical power for postmortem interval (PMI) estimation.

Abstract

LaMotte is Professor of Biostatistics at LSU Health-New Orleans. For 25 years, he has collaborated with Wells, an Associate Professor in the Department of Biological Sciences at Florida International University, in efforts to address the statistical issues and problems associated with estimates of the statistical accuracy in PMI estimates. LaMotte and Wells have been developing methods for calculating statistical confidence limits about a PMI estimate based on either continuous quantitative or categorical data. The examples they present are from forensic entomology, but the approach is suitable for any postmortem variable. They extended and adapted the time-tested statistical method of inverse prediction (IP) to the PMI estimation. In their Just Science presentation, they show how IP, using categorical data, can be done by reading a table. Regarding quantitative data, they show how inverse prediction of PMI can be performed using statistical analysis software already widely available for general linear mixed models. They also show how flexible models using polynomal spines can be fit for both the means and variance-covariance matrices, as well as how to use dummy variables over a grid of values of x to get the p-values required for confidence sets automatically. They note that Just Science attendees familiar with mixed models and their applications will be able to implement these methods in standard statistical packages.

Date Created: July 28, 2017