Estimating Effect Sizes From Graphs Using Electronic Calipers: An Analysis of the Reliability and Accuracy of a Data-Gathering Method

Abstract

Valuable data that would strengthen meta-analyses are often presented in graphs without reported means and standard deviations. This indicates potential data may not be used, and the true state of knowledge about the investigative question is not accurately represented. This paper describes and evaluates a method for extracting estimated effect sizes from graphic presentations. Two studies were conducted to assess the reliability and accuracy of using electronic calipers to estimate effect sizes from bar and line graphs. The first study looked at the reliability of effect size estimates derived from measurements taken from published graphs showing changes in one repetition maximum strength. The second study assessed the accuracy of effect size estimates computed from graphs that were constructed from known means and standard deviations. The first study demonstrated very high levels of test-retest and inter-rater reliability for the effect size estimates. The second study showed a close correspondence between the effect sizes estimated from the graphs and the known effect sizes used to construct the graphs. Thus, using electronic calipers to estimate effect sizes from graphs produces results that are accurate and reliable. Meta-analysts can confidently use this methodology to include results that have been presented only graphically.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 01, 2010
Accession Number
ADA520853

Entities

People

  • Amanda C. Barnard
  • Ross R. Vickers

Organizations

  • Naval Health Research Center

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Accuracy
  • Amino Acids
  • Body Composition
  • Computer Programs
  • Computers
  • Data Analysis
  • Data Science
  • Department Of Defense
  • Information Science
  • Measurement
  • Muscle Fibers
  • Muscles
  • Reliability
  • Spreadsheet Software
  • Standards
  • Statistics
  • Word Processors

Fields of Study

  • Education

Readers

  • Climatology
  • Systems Analysis and Design
  • Vision Science/Vision Psychology/Cognitive Neuroscience.

Technology Areas

  • Microelectronics