Related Experiment Video
Updated: Aug 19, 2026

Long-term Behavioral and Reproductive Consequences of Embryonic Exposure to Low-dose Toxicants
Published on: March 6, 2018
Effect of different sampling designs on outcome of endocrine disruptor studies
B A Elswick1, F Welsch, D B Janszen
1Chemical Industry Institute of Toxicology, 6 Davis Drive, 27709, Research Triangle Park, NC 27709, USA. elswick@ciit.org
Abstract:
In this article, we demonstrate how sampling strategy can influence the outcome of endocrine disruptor studies. In a study of the weak xenoestrogen bisphenol A (BPA), possible treatment-related effects on ventral prostate (VP) fresh weight were found in rat offspring at 6 months of age when only one or two male pups were randomly selected from each litter. In subsequent BPA and di-n-butyl phthalate studies, large intralitter variability of this specific end point was apparent when the VP weights from entire litter complements were examined. We modeled the effects of sampling 1, 2, or 3 pups from each litter using the full-litter complement data. When one pup was randomly selected, a substantial percentage of incorrect conclusions about the presence or absence of treatment effects occurred. These statistical modeling analyses raise significant concern about the selection of one pup per litter for highly variable end points.
Related Concept Videos
Sampling Plans
Random sampling is a method where each member of the population has an equal chance of being selected for the sample. It involves selecting individuals randomly, often using random number generators or lottery-type methods. For example, when analyzing the properties of a...
Bioavailability Study Design: Single Versus Multiple Dose Studies
Bioequivalence Experimental Study Designs: Completely Randomized and Randomized Block Designs
Bioequivalence Experimental Study Designs: Repeated Measures, Cross-Over, Carry-Over, and Latin Square Designs
Toxicity Testing in Animals
Study Designs in Epidemiology
Observational studies are those where the researcher does not intervene but rather observes natural variations. They include cross-sectional, cohort, and case-control studies.

