Statistical Analysis of High-Level Features from State of the Union Addresses

Statistical Analysis of High-Level Features from State of the Union Addresses

Trevor J. Bihl, Kenneth W. Bauer Jr.
Copyright: © 2017 |Pages: 24
DOI: 10.4018/IJISSC.2017040103
OnDemand:
(Individual Articles)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

A computational political science approach is taken to analyze the State of the Union Addresses (SUA) from 1790 to 2015. While low-level features, e.g. linguistic characteristics, are commonly used for lexical analysis, the authors herein illustrate the utility of high-level features, e.g. Flesch-Kincaid readability, for knowledge discovery and discrimination between types of speeches. A process is developed and employed to exploit high-level features which employs 1) statistical clustering (k-means) and a literature review to define types of speeches (e.g. written or oral), 2) classification methods via logistic regression to examine the validity of the defined classes, and 3) classifier-based feature selection to determine salient features. Recent interest in the SUA has posited that changes in readability in the SUA are due to declining audience capabilities; however, the authors' results show that changes in readability are a reflection of changes in the SUA delivery medium.
Article Preview
Top

Introduction

The State of the Union Address (SUA), a requirement of the office of the Presidency of the United States generally interpreted as annual at the opening of each Congressional session, provides an accounting to Congress of matters deemed important by the President. Ascribed in the Constitution, the SUA has been a constant requirement of the presidency and it thus makes up a large percentage of major presidential speeches (Lim, 2002) and it offers a large set of presidential rhetorical data (Coe & Neumann, 2011). Understanding the SUA can facilitate understanding the presidency, which is of importance in a variety of efforts (Lichtman, 2010). However, the SUA has changed in its form, medium, content, and overall use over time. It has essentially changed from a brief report to a summary of all activities and, more recently, to an agenda of desired actions and public image (Hoffman & Howard, 2006) (Murray, 2011). The SUA has changed as the presidency has evolved, with Fisher (2007) describing the evolution of the presidency due to the invocation of inherent powers. Binkley (1956) further describes the office as moving from a “…Chief Executive of the written Constitution into the Chief Legislator of our unwritten constitution...”

Recent interest has considered reading grade levels of the SUA, as measured by Flesch-Kincaid (FK) reading levels, and noted a decline in SUA readability (Ostermeier, 2010) (Ostermeier, 2011) (Ostermeier, 2012) (Ostermeier, 2013) (Yau, 2013) (Schneiderman, 2012) (Hemingway, 2012) (Schramm, 2014). Borevitz (2012) examined the FK reading level of the SUA, for 1790-2012, and acknowledged a general declining trend in reading level. Ostermeier (2010)-(2013) posited that declining SUA readability was a result of declining American audience attention spans. The conclusions of Ostermeier’s reports were reinforced by opinions of Hemingway (2012) and Schneiderman (2012); The Guardian (2013) similarly expanded upon Ostermeier’s conclusions and reported a headline “The state of our union is…dumber…” and an interactive SUA FK graphical user interface (GUI) which included examples of text from each SUA. Sploid/Gizmodo (Chan, 2014) similarly concluded “Speeches from US presidents have gotten dumber over time.” However, such conclusions oppose the guidance of Orwell (1946) who recommended short and concise statements to aid public understanding. Additionally, the changes in FK could be a result of changes in delivery medium, as posited by Yau (2013), and reflects changes in political consumption, c.f. (Adugu, 2016).

Herein the authors examine relative changes and trends in the SUA using high-level features. All SUAs, from 1790 to 2015, are considered and Joint Addresses are also included as a type of SUA, consistent with Hoffman and Howard (2006). This research presents the use of the FK in SUA analysis in conjunction with other readability metrics (total words, average characters per word, average words per sentence, and presidential post-secondary education) for differentiating between the relative complexities and trends of SUAs. These are used to draw contrasts between written, oral, publicly delivered, televised, and prime-time televised SUAs. A computational political science approach, considered herein as the development and application of analytical, statistical and theoretical methods to the study of political science systems (Weber, Popescu, & Pennacchiotti, 2013) (Zhu, 2010), is used for analysis. Such an approach applies pattern recognition methods, which broadly encompasses: describing, classifying, grouping, categorization and/or clustering data through automated means (Jain, Duin, & Mao, 2000).

Complete Article List

Search this Journal:
Reset
Volume 15: 1 Issue (2024)
Volume 14: 1 Issue (2023)
Volume 13: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 12: 4 Issues (2021)
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing