Skip to Main Content
Frequently Asked Questions
Submit an ETD
Global Search Box
Need Help?
Keyword Search
Participating Institutions
Advanced Search
School Logo
Files
File List
brodbeck final.pdf (4.27 MB)
ETD Abstract Container
Abstract Header
The Effect of Readability on Simple Linear Regression
Author Info
Brodbeck, William Joseph
ORCID® Identifier
http://orcid.org/0000-0001-7390-0710
Permalink:
http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1591867761661656
Abstract Details
Year and Degree
2020, Master of Arts (MA), Bowling Green State University, Mathematics/Mathematical Statistics.
Abstract
In this thesis, the effect of readability on simple linear regression (SLR) is investigated. The effect is applicable to SLR because readability errors are included in recorded data, and consequently in the residuals, whenever a sample is taken of a continuous random variable. The readability errors are at a minimum the round off error of the recorded data. However, practical concerns often necessitate larger readability errors. Readability has the effect of dividing a continuous distribution into subsections or intervals. In this thesis, only intervals of equal width and equal area (i.e., equal probability) are evaluated. The minimum number of equal width or equal area intervals for which the effect of readability on SLR may be considered negligible is determined. In addition, a practical minimum of five intervals is also evaluated. The SLR model is modified such that readability errors are treated separately from the residuals. The modified SLR model is called the Readability model. The effect of readability is defined as the difference in the variance of the residuals between the standard SLR model and the Readability model. To assess the appropriateness of the Readability model, a sample with negligible readability errors is created from a continuous random distribution. A second sample is created by introducing readability into the first sample. Linear regressions of the two samples, one without readability errors and one with readability errors, are then compared to determine the effect of readability. In this thesis, samples are taken from Uniform, Beta and Normal distributions. The Readability model is shown to be an appropriate model for evaluating the effect of readability on simple linear regression.
Committee
John Chen, Ph.D. (Advisor)
Craig Zirbel, Ph.D. (Committee Member)
Pages
128 p.
Subject Headings
Mathematics
;
Statistics
Keywords
readability
;
simple linear regression
;
SLR
;
continuous residuals
;
readability model
;
Recommended Citations
Refworks
EndNote
RIS
Mendeley
Citations
Brodbeck, W. J. (2020).
The Effect of Readability on Simple Linear Regression
[Master's thesis, Bowling Green State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1591867761661656
APA Style (7th edition)
Brodbeck, William.
The Effect of Readability on Simple Linear Regression .
2020. Bowling Green State University, Master's thesis.
OhioLINK Electronic Theses and Dissertations Center
, http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1591867761661656.
MLA Style (8th edition)
Brodbeck, William. "The Effect of Readability on Simple Linear Regression ." Master's thesis, Bowling Green State University, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=bgsu1591867761661656
Chicago Manual of Style (17th edition)
Abstract Footer
Document number:
bgsu1591867761661656
Download Count:
171
Copyright Info
© 2020, all rights reserved.
This open access ETD is published by Bowling Green State University and OhioLINK.