Home
Resource
Understanding graphs with two independent variables
PDF
PDF format is widely accepted and good for printing.
Plug-in required
PDF-1(7.48 MB)
Citation & Export
View Usage Statistics
Staff View

Citation & Export
Hide

Simple citation

Cooper, Jennifer L.. Understanding graphs with two independent variables. Retrieved from https://doi.org/doi:10.7282/T3CC10FH

Export

Click here for information about Citation Management Tools at Rutgers.

Statistics
Hide

Description

TitleUnderstanding graphs with two independent variables
NameCooper, Jennifer L. (author); Gelman, Rochel (chair); Chapman, Gretchen (internal member); Singh, Manish (internal member); Rips, Lance (outside member); Rutgers University; Graduate School - New Brunswick
Date Created2010
Other Date2010 (degree)
SubjectPsychology, Graphic methods, Statistics, Variables (Mathematics)
Extentxx, 288 p. : ill.
DescriptionAdults are not necessarily competent users of graphs with two independent variables, despite the frequency of this representational format. The three tasks in this thesis address the impact of interpretation statements and graph patterns. Interpretation statements were based on the statistical effects – simple effects, main effects, and interactions. Graph patterns were systematically varied based on a novel classification scheme of graphs with two IVs. I suggest that the complexity of a graph’s data pattern depends on the consistency of the simple effects’ directions and magnitudes. In the first study, undergraduates constructed graphs based on statements about data patterns. Errors reflected a misunderstanding of how two IVs could be combined and represented graphically. When the experimental group had graph-relevant information added (variable labels spatially located on axes), the ability to represent the relationships among the IVs significantly increased. The ability to satisfy the constraints imposed by the statements was not affected. Adding labels specifically targeted skills relevant to graphical literacy. Transfer to a third trial was stronger for those of higher math abilities. The second study focused on the effect of an introductory statistics course. Overall, undergraduates performed well on statements describing the simple effects of the IVs. However, even though they improved from Time 1 to Time 2 for interaction statements, performance on statements about main effects and interactions still showed considerable room for improvement. In the third study, repeated trials of the 20 patterns proposed by the simple effects consistency model established that the proposed classification scheme addresses additional sources of variability in reasoning with graphs (i.e., sources not captured by traditional classification schemes). As the complexity level of the data pattern increased, performance (based on accuracy and RT) decreased, with parallel impacts on performance for each IV’s complexity. This suggests that participants responded to conceptual differences among the levels, as the graph’s perceptual characteristics vary based on the IV. Further development of a model organizing graph patterns will allow investigation of the interplay between the statement and graph pattern. In turn, this can lead to greater understanding of the graphical reasoning processes and improvements in graphical literacy.
NotePh.D.
NoteIncludes bibliographical references
NoteIncludes vita
Noteby Jennifer L. Cooper
Genretheses, ETD doctoral
Persistent URLhttps://doi.org/doi:10.7282/T3CC10FH
Languageeng
CollectionGraduate School - New Brunswick Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.
Version 8.5.5
Rutgers University Libraries - Copyright ©2024