자료유형  

 학위논문

Control Number  

0016935832

International Standard Book Number  

9798380596275

Dewey Decimal Classification Number  

150

Main Entry-Personal Name  

Hoisington-Shaw, Kathryn Jewell.

Publication, Distribution, etc. (Imprint  

[S.l.] : The Ohio State University., 2023

Publication, Distribution, etc. (Imprint  

Ann Arbor : ProQuest Dissertations & Theses, 2023

Physical Description  

1 online resource(155 p.)

General Note  

Source: Dissertations Abstracts International, Volume: 85-04, Section: B.

General Note  

Advisor: Pek, Jolynn.

Dissertation Note  

Thesis (Ph.D.)--The Ohio State University, 2023.

Restrictions on Access Note  

This item must not be sold to any third party vendors.

Summary, Etc.  

요약In social science research, statistical power is emphasized as the "gold standard" to justify sample size in hopes that it promotes transparency in study design, addresses concerns about the replicability of findings, and limits questionable research practices. However, statistical power can result in a sample size that is unattainable for researchers to collect, and the practical uses of power are often misinterpreted beyond their scope. This dissertation consists of three projects that address statistical power and its limitations, as well as investigates alternative methods to justify sample size. The first project is a meta-science study that examines elements of study design and statistical analyses used in research published in Psychological Science. Results indicate the most common type of effect size used as an input in power analysis is sourced from previously collected data. However, use of point estimated effect sizes does not account for sampling variability and can result in estimates that are imprecise. Therefore, the second project in this dissertation empirically evaluates the performance of several sample size determination methods that expand upon classical power analysis by accounting for sampling variability of the effect size. In general, the sample sizes calculated from these modern methods have the potential to be unrealistically large for researchers to collect, which could deter from their use. To address this issue, the third project moves away from statistical power and focuses on using the sequential probability ratio test (SPRT) for study design and analysis instead. Two new extensions of SPRT are proposed that also account for sampling variability of the effect size. Evaluations of these approaches suggest that they offer a beneficial alternative to power analysis, especially for researchers that need to limit sample size. 

Subject Added Entry-Topical Term  

Psychology.

Subject Added Entry-Topical Term  

Statistics.

Subject Added Entry-Topical Term  

Quantitative psychology.

Index Term-Uncontrolled  

Statistical power

Index Term-Uncontrolled  

Sample size determination

Index Term-Uncontrolled  

Sequential testing

Index Term-Uncontrolled  

Meta-science

Index Term-Uncontrolled  

Sampling variability

Added Entry-Corporate Name  

The Ohio State University Psychology

Host Item Entry  

Dissertations Abstracts International. 85-04B.

Host Item Entry  

Dissertation Abstract International

Electronic Location and Access  

로그인을 한후 보실 수 있는 자료입니다.

Control Number  

joongbu:643975