Intuitive Analysis of Variance: Effect of Different Data Sets and Presentation Types

Sait Atas, David L. Trumpower, Mehmet Filiz

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Previously, it has been found that students tend to perceive the difference between group means as being more important than the consistency of those differences when engaged in an intuitive analysis of variance (ANOVA) task, even though both are equally important in a formal statistical sense. However, the reason behind this tendency is not known. Therefore, the purpose of the present study was to explore one factor (numerical versus graphical presentation of data sets) that could play a role in students' tendency to place more weight on between group variability. Graduate students enrolled in an introductory level statistics course were shown a series of hypothetical data sets, each comprised of scores from two independent groups. Some of the data sets were presented numerically and others graphically. Students were asked to rate the strength of evidence provided by each data set against a null hypothesis of no difference between the groups, and to provide explanations for their ratings. Results indicate that students place more importance on the magnitude of group mean differences than the consistency of the differences, regardless of whether data are presented numerically or graphically. Implications for teaching ANOVA are discussed
Original languageEnglish
Title of host publicationUnited States Conference on Teaching Statistics 2013
Publication statusPublished - 2013
Externally publishedYes
EventUnited States Conference On Teaching Statistics 2013 - Embassy Suites Hotel & Conference Center, Raleigh, North Carolina, United States
Duration: 16 May 201318 May 2013

Conference

ConferenceUnited States Conference On Teaching Statistics 2013
Country/TerritoryUnited States
CityRaleigh, North Carolina
Period16/05/201318/05/2013

Fingerprint

Dive into the research topics of 'Intuitive Analysis of Variance: Effect of Different Data Sets and Presentation Types'. Together they form a unique fingerprint.

Cite this