ERIC Number: EJ1094191
Record Type: Journal
Publication Date: 2016-Mar
Abstractor: As Provided
Reference Count: 25
Confidence Intervals for Effect Sizes: Applying Bootstrap Resampling
Banjanovic, Erin S.; Osborne, Jason W.
Practical Assessment, Research & Evaluation, v21 n5 Mar 2016
Confidence intervals for effect sizes (CIES) provide readers with an estimate of the strength of a reported statistic as well as the relative precision of the point estimate. These statistics offer more information and context than null hypothesis statistic testing. Although confidence intervals have been recommended by scholars for many years, these statistics are often not reported. This may be partially due to the complexity of calculating confidence intervals for many statistics. Bootstrap resampling can be used to easily estimate confidence intervals around almost any type of point estimate. The aim of this paper is to demonstrate this methodology using real-world data and to develop several simple principles around this methodology to guide readers in appropriate application.
Descriptors: Computation, Statistical Analysis, Effect Size, Sampling, Statistical Inference, Correlation, Regression (Statistics)
Dr. Lawrence M. Rudner. e-mail: firstname.lastname@example.org; Web site: http://pareonline.net
Publication Type: Journal Articles; Reports - Descriptive
Education Level: N/A
Authoring Institution: N/A