Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Should we take measurements at an intermediate design point?

A Gelman1

  • 1Department of Statistics, Columbia University, New York, NY, 10027, USA. gelman@stat.columbia.edu

Biostatistics (Oxford, England)
|August 23, 2003
PubMed
Summary

For estimating linear treatment effects, optimal designs typically use extreme points. However, for nonlinear dose-response relationships, avoiding intermediate points is often best unless sample sizes are very large.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Computer-Aided Detection of Respiratory Sounds in Bronchial Asthma Patients Based on Machine Learning Method.

Sovremennye tekhnologii v meditsine·2023
Same author

A simple explanation for declining temperature sensitivity with warming.

Global change biology·2021
Same author

A child with resistant Kawasaki disease successfully treated with anakinra: a case report.

BMC pediatrics·2017
Same author

B Cell-Activating Transcription Factor Plays a Critical Role in the Pathogenesis of Anti-Major Histocompatibility Complex-Induced Obliterative Airway Disease.

American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons·2016
Same author

Targeting low-arsenic groundwater with mobile-phone technology in Araihazar, Bangladesh.

Journal of health, population, and nutrition·2007
Same author

Effect of ozone pretreatment on fish storage life at low temperatures.

Journal of food protection·2005

Area of Science:

  • Biostatistics
  • Experimental Design
  • Environmental Health

Background:

  • Optimal experimental design for linear treatment effects involves allocating units to extreme design points.
  • Nonlinear dose-response relationships may necessitate intermediate design points for accurate estimation.
  • The trade-off exists between efficiently estimating linear effects and detecting nonlinearity.

Purpose of the Study:

  • To determine if gathering data at intermediate design points is beneficial when a nonlinear dose-response relationship is possible.
  • To evaluate whether the gains from detecting nonlinearity outweigh the loss in efficiency for estimating linear effects.

Main Methods:

  • The study analyzes the optimal allocation of units in experimental designs considering both linear and nonlinear treatment effects.
  • It compares designs with and without intermediate data points under varying sample sizes and nonlinearity assumptions.

Main Results:

  • Under reasonable assumptions of nonlinearity, designs without intermediate measurements are generally superior for moderate sample sizes.
  • Detecting nonlinearity in dose-response relationships is challenging with moderate sample sizes, favoring simpler designs.
  • The efficiency loss in estimating linear effects often outweighs the benefits of intermediate points for moderate sample sizes.

Conclusions:

  • For estimating treatment effects, especially in contexts like pest control for asthma, focusing on extreme design points is often the most efficient strategy.
  • Intermediate design points are only advantageous for detecting nonlinearity when sample sizes are sufficiently large.
  • The findings suggest prioritizing robust linear effect estimation over potentially undetectable nonlinear effects in many practical scenarios.

Related Experiment Videos