The semantic reactivity of red, blue, and purple: A linguistic analysis of post-election statements made by executive leadership of three public flagship universities

Abstract Examining post-election statements made by UC System, UT-Austin, and UW-Madison executive leadership, this study employs word frequency, collocation, and a three-pronged latent semantic analysis to explicate the associative diction, major concepts, and institutional priorities expressed by said leadership to answer the research question, “What is the purpose and function of a post-election statement made by postsecondary executive leadership?” Results reveal that system-level leaders address broad coalitions, avoid action semantics, and reiterate system principles, whereas institution-level leaders perform reactive semantics while promising campus community action. Implications for campus community members, as well areas of future linguistic research, are addressed.


Introduction
After the Presidential Election on 8 November 2016, many institutions of higher education (IHEs) across the United States witnessed their executive leadership release post-election statements to comment on the implications of a Donald Trump presidency on their respective campus(es). This was certainly the case for the University of California System, led by President Janet Napolitano, the University of Texas at Austin, led by President Gregory Fenves, and the University of Wisconsin at Madison, led by Chancellor Rebecca Blank: all three executive leaders issued post-election statements to their constituencies within one week of election day. Napolitano and the Chancellors of the UC System posted their statement to the UC System website on 9 November (Napolitano & Chancellors of the University of California, 2016), UW-Madison Chancellor Blank posted a statement to her institutionally sponsored blog "Blank's Slate" on the same day (Blank, 2016), and five days later, UT-Austin President Fenves issued a statement via campus-wide email on 14 November (Fenves, 2016).
Situating these three leaders in the historical context of their institution's home provides a fertile soil for the analysis of these leaders' post-election statements. The state of California, as well as the cities of Austin, Texas (and Travis County) and Madison, Wisconsin (and Dane County) have voted overwhelmingly democratic in recent presidential elections, yet the 2016 Presidential Election results saw California vote 61.5/32.2% Democrat/Republican, Texas vote 52.6/43.4% Republican/ Democrat, and Wisconsin 47.9/46.9% Republican/Democrat (Associated Press, 2016). In short, although the UC System, UT-Austin, and UW-Madison share a sociopolitical similitude in their respective local geographies, they lead public flagship universities in three states with radically divergent state-level political climates: "blue," "red," and "purple" states, respectively. These three idiosyncratic relationships between each public flagship and the corresponding political climate in their state rendered the linguistic analysis of each executive leader's post-election statement incredibly timely, relevant, and important. No extant research has addressed post-election statements made by educational leadership of any type at any level. However, an analysis of this content could yield unique insight into how a public flagship leader views their campus political climate, understands the political relationship of their institution and its state, and articulates institution-specific beliefs during a time of sociopolitical turmoil and uncertainty.
As a result, this study focuses on three public flagship leaders in three different sociopolitical climates to answer the question: given diverse political climates, what is the purpose and function of a post-election statement made by postsecondary executive leadership? To answer this question, I performed a word frequency, collocation, and three-pronged latent semantic analysis (LSA) of each executive leader's post-election statement to analyze the associative diction, major concepts, and institutional priorities expressed by said leadership. Granted, an analysis of three statements is limited in its generalizability. However, this is the first study of its kind, and the states of California, Texas, and Wisconsin all proved incredibly idiosyncratic in their presidential voting preferences, yet each state's public flagship area voted similarly. For as divided as the United States is purported to be (Doherty, 2017), it is crucial to linguistically analyze this divide and attempt to provide a foundation for future analysis of official statements from leaders of institutions of higher education in the United States.
I will begin with a review of the literature to situate this study in extant research focused on higher education leadership communication, followed by my conceptual framework rooted in Rorty's (1989) neopragmatism. Then I discuss the LSA methodology (Landauer, Foltz, & Laham, 1998), articulate my findings, elaborate upon those findings, explicate limitations of the study, and outline implications for future research before my conclusion.

Literature review
No higher education research exists examining post-election statements made by executive leadership in colleges and universities, and very little literature exists focusing specifically on the communicative responsibilities of university presidents and chancellors at public flagship institutions. Related preliminary foundational work has been performed that explicates executive leadership priorities (Bok, 2014), postsecondary leadership styles (Aldighrir, 2013), framing devices in presidential speeches (Young, 2013), media coverage of crisis communications (DiManno, 2010), communication practices between university presidents and their board of trustees/regents (Legon, Lombardi, & Rhoades, 2013), and presidential roles at independent colleges and universities (Song & Hartley, 2012).
However, these studies do not specifically examine official statements made by presidents and chancellors of public flagship institutions, especially during a national post-Presidential election period. These post-election periods can prove tumultuous and divisive for those on both sides of the political aisle (Doherty, 2017), and this phenomenon has intensified in recent years where the campus has re-emerged as ground zero free-speech related protesting and the presence of hate speech targeting minoritized individuals on college campuses (Jaschik, 2016). Of the existing research focused on communicative roles and responsibilities of postsecondary executive leadership, recent work suggests that effective executive communication at the postsecondary level is associated with several individual characteristics and administrative functions.
First, in a recent study of executive leadership at 10 universities in the United Kingdom, interpersonal communication skills and the ability to negotiate values and beliefs with others were most associated with effective leadership and communication at the postsecondary level (Spendlove, 2007). Proceduralizing these individual characteristics, postsecondary executive leadership must also effectively communicate with the campus and external constituencies regarding an institution's mission and vision, as well as mitigate crises through an exercising of good judgment via clear public relations strategies and honest communication (Pierce, 2011).
Of organizational culture, postsecondary executive leadership often communicates on the behalf of a culture that manages crisis, instead of engaging with reflective practices to foster consensual change. Focusing particularly on the president or chancellor of an institution, pronouncements of excellence or institutional mission hold dual importance: providing a rationale for programmatic development and upholding a standard of self-criticism and performance (Tierney, 1988). Although a post-election statement may not overtly fall under the category of a pronouncement of excellence nor an articulation of an institutional mission, the post-election climate on many college and university campuses across the country constituted a crisis necessitating subsequent management, evidenced by the surge of ethnic and racial harassment experienced by postsecondary students (Jaschik, 2016) and efforts to provide self-care and avenues for civic conversation at postsecondary institutions across the country (Mascarenhas, 2016). Therefore, Tierney's decades-old sentiments are particularly useful in understanding how modern post-election statements issued by an institution's executive leadership may serve as pronouncements of excellence in surviving a crisis or methods to uphold standards of self-criticism and performance in the face of sociopolitical adversity on campus and beyond.
Augmenting the discussion of Tierney's ideology as it relates to post-election statements is Rabovsky and Rutherford's (2016) study of a 2012 survey of United States university presidents. These researchers found that executive leaders at the postsecondary level tend to "network at higher rates with state-level actors in conflictual rather than cooperative environments" (p. 771) and have "shifted toward buffering the organization against negative shocks" as opposed to adopting Tierney's ideology and fostering proactive, reflective practices and consensual change (p. 773). Their results also suggest that university presidents are much more likely to focus their communication on converting skeptical and critical political actors, instead of uniting support from allied constituencies (p. 773).
Moreover, presidents and chancellors of public flagship institutions have unique communication demands, articulated by Douglass, (2016) The New Flagship University: Changing the Paradigm from Global Ranking to National Relevancy. Therein, Douglass implies a paradoxical relationship between the charter language of many public flagship institutions and the growing pressure to globalize the mission and scope of said institutions, claiming that a "range of public engagement and active involvement in local economies" were essential, foundational roles of public flagships in California,Texas,and Wisconsin (p. 34), yet these public flagships also embrace a responsibility related to the matriculation of international students, while drawing attracting faculty and student talent from an increasingly international pool (p. 44). As a result, public flagship presidents and chancellors may feel compelled to communicate in a fashion that speaks to local, national, and international audiences simultaneously, while representing a public flagship institution chartered to represent their respective state's multifarious interests (Douglass, 2016). Given Tierney's (1988) communication implications rooted in organizational theory, Rabovsky and Rutherford's (2016) insight regarding external relations and university leadership, and Douglass (2016) articulation of the communication paradox facing current public flagship university presidents and chancellors, the examination of post-election statements made by these leaders seems an especially timely and valuable contribution to the field.

Conceptual framework
Bathe's Janua Linguarum (1611) and Comenius' Orbus Pictus (1657) examined word frequency to teach second languages and incorporate audio and visual techniques in children's language education, respectively. As a topic of theoretical research, word frequency was popularized in the twentieth century by Zipf, a quantitative linguist who articulated Zipf's Law (Zipf, 1935): in each text, few words occur very often and many others occur rarely considering the context and purpose of that text. Since his contribution, Zipf's Law has been exhaustively used as a conceptual framework to study a wide range of social science issues such as income distribution of companies in relation to their geography (Okuyama, Takayasu, & Takayasu, 1999), size distribution of cities (Gabaix, 1999), and gene expression (Furusawa & Kaneko, 2003).
Applying Zipf's law to this study, the more frequently a public flagship leader uses a word for a specific purpose (i.e. articulating institutional values in a divisive, post-election climate), the more powerful that word is believed to be by its author and the more powerful that word is to influence the syntax and semantics (meaning) of subsequent text. In short, words matter: they reveal values, beliefs, and thought processes. In higher education research specifically, a gap in the literature exists, necessitating a word frequency analysis of post-election statements issued by postsecondary executive leadership, especially that of public flagship institutions.
Collocation is the habitual co-occurrence (or mutual selection) of lexical items (Crystal, 2010). For example, the words "heavy traffic" are much more likely to be collocated in the English language than "weighty traffic" or "substantial traffic," even though the denotations (dictionary definitions) and connotations (associative diction) of each root word have separate similar, if not identical, meanings. Extant research suggested that institutional mission statements issued by postsecondary executive leadership can reveal the current tone and climate of an institution, along with the values and beliefs that produce institutional goals and culture (Wilson, Meyer, & McNeal, 2012). Examining the collocation of terms in post-election statements may explicate a public flagship university's attitude toward a recent election and subsequent political landscape of a campus and/or local region. Furthermore, studies of institutional mission statements have found that universities are frequently shortening the length of their mission statements to reflect a more straightforward, intelligible set of institutional goals and beliefs (Kiley, 2011), possibly rendering post-election statements made by postsecondary executive leadership even more terse in size and scope given their potentially controversial and political nature. Therefore, a collocation analysis of post-election statements seems appropriate and timely. https://doi.org/10.1080/2331186X.2017.1391927 Finally, Landauer, Foltz, and Laham (1998) define LSA thus: … a fully automatic mathematical and statistical technique for extracting and inferring relations of expected contextual usage of word in passages of discourse. … it takes as its input only raw text parsed into words defined as unique character strings and separated into meaningful passages or samples such as sentences or paragraphs. (p. 263) Upon assigning key terms a numeric value, LSA produces word-word, word-passage, and passagepassage relation similarity statistics in a given corpus (collection of texts), which then informs a reader of a given word's or passage's relative representation of the whole in mathematical terms. Given the philosophical and political nature of a post-election statement, an LSA is especially appropriate, as representative passages formed by LSA can be interpreted as "episodes," which can inform an audience's perception of the importance and relevance of a statement within the context of other statements, forming a corpus (p. 261). Furthermore, LSA remedies the paradoxical inductive property of learning known as the "poverty of input" or the "insufficiency of evidence," when interlocutors acquire more knowledge than appears to be readily available in a communicative experience (p. 262), such as a post-election statement, effectively removing reader bias and providing a statistical analysis of a text.
As a result, applying LSA to post-election statements may reveal more insight into the potential knowledge gleaned from such statements that are not explicitly referenced therein. Additionally, LSA does not examine human speech characteristics, such as tone or rate, and only surmises the reflexive meaning of text in self-contained documents, compared to a corpus of representative text, known as a semantic space (Landauer, Foltz, & Laham, 1998). As a result, the LSA of a post-election statement treats words and passages as currency of equal value across text in the same corpus, despite the disparate environments in which these statements realize their origin: an objective, quantitative, comparative analysis of text is achieved.
Framing this examination of word frequency, collocation, and LSA is Rorty (1989), who spoke of the nature of truth as incontrovertibly framed by language, inseparable from the human mind, validated by the activities of human beings whose agency is solely determined by their linguistic resources and abilities. For decades, American universities have attempted to adjust their intrinsic properties to become more equitable and inclusive (i.e. Texas HB 588, or the "Top 10% Rule"), yet without examining the language used by universities to communicate with internal and external constituencies, universities cannot be surprised when they continue to suffer from the same equity and inclusivity issues (Astin & Oseguera, 2004) in large part due to an inability to recruit and retain students from underrepresented groups (Demaris & Kritsonis, 2007), especially at predominantly white institutions (PWIs), such as many public flagship universities (Palmer, Maramba, & Holmes, 2012).
Here, there is an evaluative disconnect between the purported goals of equity and inclusiveness of postsecondary institutions and the manner in which these concepts are analyzed, examined, and explicated to internal and external constituencies by said institutions: until this point, this disconnect has not been linguistically quantified. Rorty (1989) and the neopragmatic tradition, however, strongly believe that words are used strategically and to achieve a purpose. Postsecondary institutions and their strategies and purposes can only be defined by the language they employ and audiences they choose to rhetorically address. Rortian neopragmatics could help inform the studies of these postsecondary access and equity issues by framing these difficulties as linguistic obstacles remedied by linguistic analysis and explication of statements made by executive leadership: the voices of postsecondary institutions.

Design
Word frequency and collocation analysis is a quantitative linguistics method recently used by lexicographers (Nurmukhamedov, 2016; Siepmann, 2005; Zhao-Ming, 2014) to build understanding and establish usage standardization of a given language: these analyses seem appropriate for the examination of post-election statements, given all statements were issued in text format by leaders of public flagships institutions. I used Landauer and Dumais (1997) to guide the LSA process; further detail on the three-pronged LSA approach appears below.

Sample
This research study depended solely on the text of post-election statements publicly issued by executive leadership at three public flagship institutions: the University of California System, the University of Texas at Austin, and the University of Wisconsin at Madison. These three institutions and their statements were selected for this reason using the following criteria.
First, I wanted to examine the post-election statements made by executive leadership of public flagship institutions because of the election results in their three states, respectively. As previously stated, California, Texas, and Wisconsin each represent Democratic (blue), Republican (red), and bipartisan (purple) voting outcomes, evidenced by the statistics referenced in the introduction section of this study (Associated Press, 2016).
Second, as articulated in the literature review section of this paper, university presidents and chancellors at public flagship institutions face distinct communicative challenges that postsecondary leaders of other institution types simply do not encounter. As previously stated, the three public flagship institutions examined in this study are physically situated in historically Democratic areas, yet these public flagship institutions were founded upon charters that express representation of the state collectively a variety of issues. Compounding this communicative paradox is the modern trend for public flagships to become increasingly more globalized, yet also aspire to serve local and state constituencies. This paradox rendered the analysis of post-election statements issued by executive leaders of these three public flagship institutions especially interesting.
Finally, the dissemination method of each post-election statement was curious. To begin, the UC System and UW-Madison saw their executive leaders issue statements immediately after the election-9 November-whereas UT-Austin's executive leadership issued a statement on 14November, a full six days after the election occurred. Continuing, none of these executive leaders issued an official verbal statement as of 1 December 2016. Instead, UC System President Janet Napolitano and the Chancellors of the UC System decided to issue their statement on their system-wide website, UT-Austin President Gregory Fenves issued his statement through a campus-wide email, and UW-Madison Chancellor Rebecca Blank issued her statement on her institutionally sponsored blog: all three were issued using text only, as no images or audio accompanied their text. These three forms of media-and their usage by public flagship executive leadership-represents a wealth of future research possibility. Also, the length of each statement was interesting: Napolitano's statement was 161 words in length, Fenves' was 725 words, and Blank's was 503 words. These differences rendered the analysis of post-election statements issued by the UC System, UT-Austin, and UW-Madison especially intriguing.
For these reasons, I felt it appropriate to include these three post-election statements in this study's sample to provide a foundation for future work in the linguistic analysis of official statements made by leaders of postsecondary institutions.

Data collection
Each post-election statement was issued by an executive leader at a public flagship institution, and therefore all post-election statements were made public and required no special permission or public records request. First, the text was extracted from each origin source and uploaded into a simple text editor (Microsoft Word) to clean the text of articles (e.g. a, an, the), conjunctions (e.g. and, because, for), and prepositions (e.g. in, on, between) per Landauer and Dumais (1997). Then, the text of each statement and the corpus was uploaded into two separate software suites: Oleander Solutions' Readability Studio and the University of Colorado at Boulder's LSA website (lsa.colorado.edu). https://doi.org/10.1080/2331186X.2017.1391927

Data analysis procedures
The text of each post-election statement and corpus was analyzed using two software applications: Oleander Solutions' Readability Studio and CU-Boulder's LSA website (lsa.colorado.edu).
First, I used Readability Studio to perform word frequency and collocation calculations based on the input text. Once the text was uploaded, the software isolated the most frequently used words in each statement (three separate inputs or texts), and the most frequently used words in the corpus (one input, one text). This procedure allows for the reader to compare and contrast the most frequently used words in a given text and the corpus as a whole. The results of this isolation allowed me to strategically target frequently used words to perform a collocation analysis of these words, which can be found in Table 1. All Readability Studio files (.rsp) are available upon request from the author.
Second, I used the CU-Boulder LSA software website and their "General Reading, up to 1st year college" document (CU-Boulder, 2015, para. 1) as the common semantic space in which to analyze the semantics of the three post-election statements and the corpus (Landauer & Dumais, 1997). I used this particular semantic space because of its audience-appropriateness and the nature of a post-election statement: these statements are not technical reports, medical journals, or any other type of idiosyncratic publication requiring highly technical jargon and phraseology. This semantic space served as the comparative text to measure the similarity between post-election statements and the similarity of semantic utility between the post-election statements and a document meant for "General Reading" of an individual with a first-year postsecondary reading ability.
Then, each statement and the corpus was uploaded into the "Near Neighbors" application-first excluding terms which did not appear more than three times-to measure LSA similarity of the top five terms of each statement and the top five terms in comparison to the sample semantic space. Next, each statement was uploaded into the "One-to-Many" application to measure each text's representation of the corpus. Finally, each statement was uploaded into the "Matrix Comparison" application to measure the semantic similarity between each statement. The LSA results are included in Table 2. All LSA files (.htm) are available from the author upon request.

Word frequency and collocation analysis
A word frequency and collocation analysis of each post-election statement can be found in Table 1.
Beginning with the UC System, the most frequently used words in the post-election statement were "university" (6 incidents), "California" (4), "our" (4), "principles" (3), and "we" (3). These results are affected by the fact that the UC System statement was the shortest (161 words) of the three post-election statements analyzed in this study. Of these most frequently used words, "university" and "California" were most frequently collocated with each other and the word "community." Furthermore, the words "principles" and "we" were most frequently collocated with the words "university" and "California." For UT-Austin, the most frequently used words in the post-election statement were "we" (12), "our" (10), "campus" (8), "members" (7), and "they" (7). These results are affected by the fact that the UT-Austin statement was the longest (725 words) of the three post-election statements analyzed in this study. Of these most frequently used words, "we" and "campus" were most frequently collocated with the word "university." Also, the words "our" and "they" were most frequently collocated with the word "students." "They" was also frequently collocated with the term "targeted," and the word "campus" was frequently collocated with the word "ideas," which represented curious outliers in the data: these will be addressed in the discussion section.
For UW-Madison, the most frequently used words in the post-election statement were "we" (13), "our" (10), "community" (8), "campus" (7), and "diversity" (4). Of these most frequently used words, "we," "our," and "diversity" were frequently collocated with the word "community." Also, the words "our" and "diversity" were frequently collocated with the word "campus." The word "diversity" was an outlier in the data, as this word was not frequently used or collocated in any other post-election statement in this sample.
Of the corpus, the most frequently used words were "we" (28), "our" (24), "university" (15), "campus" (15), and "community" (15). Again, these results are affected by the various lengths of each statement as it represented the corpus. The words "we," "university," and "campus" were frequently collocated with the word "community," and the word "campus" was most frequently collocated with the word "climate" in the overall corpus.

Latent semantic analysis
A three-pronged LSA of each post-election statement and the corpus can be found in Table 2.

Statement
Length of statement (words)
The "One-to-Many" LSA rendered the UW-Madison post-election statement most semantically representative of the corpus (0.94), even though the UW-Madison statement was not the longest statement analyzed in this study. The UC System post-election statement was least semantically similar to the corpus (0.72), and UT-Austin held the middle ground (0.89). Finally, the "Matrix Comparison" LSA rendered the UW-Madison and UT-Austin post-election statements most semantically similar (0.73), whereas the UC System statement was least semantically similar to UW-Madison (0.63) and UT-Austin (0.66).

Statement
Near-neighbor analysis One-to-many Matrix comparison

Discussion
My research question-What is the purpose and function of a post-election statement made by postsecondary executive leadership?-was answered, in part, from the themes which emerged from the series of linguistic analyses.
First, judging from the mere length of each statement, it could be rationalized that because the UC System President-Janet Napolitano-was speaking on behalf of the entire UC System and its many Chancellors, the statement was purposefully composed in a terse fashion. However, like Chancellor Blank of UW-Madison, the UC System released their statement immediately following the election, although Blank's solo-authored statement was over three times as long. Here, differences between a solo-authored leadership statement and a collaboratively authored leadership statement emerge: although Napolitano's and Blank's statements were released on the same day, they were published at much different lengths and through different media sources (the UC System website and Blank's institutionally sponsored blog), casting an entirely different ethos onto each document. For instance, the UC System's rationale for publishing a system-wide statement instead of each UC campus leader composing their own statement could be tied to the US Presidential election results: all cities that house a UC System campus voted Democratic (Associated Press, 2016), and perhaps Napolitano's statement was meant to serve as an assertion of unity despite unfavorable election results in many UC System areas. However, it does not seem that the UC System mission statement informed the post-election statement, as nowhere in the UC System mission statement is there mention of representation of or service to the state of California (Regents of the University of California, 2017), instead choosing to broadly address society and public service without mentioning specific stakeholders.
Similarly, by deciding to speak as the leader of UW-Madison without co-authoring a statement with any other UW System Chancellor or the system's Board of Regents, Chancellor Blank's statement seemingly sidestepped UW-Madison's guiding principle, the Wisconsin Idea. The origins of the Wisconsin Idea are attributed to former UW President Charles Van Hise, who asserted that, "I shall never be content until the beneficent influence of the University reaches every family of the state" (Board of Regents of the University of Wisconsin System, 2017, para. 2). Like California, nearly all cities that house UW System campuses voted Democratic, but the state of Wisconsin voted Republican (Associated Press, 2016). As a result, Blank's statement-because of its solo-authorship and positioning on her UW-Madison blog instead of the UW System website-could have been written to reflect only the local politics on her campus. In no uncertain terms, Madison, Wisconsin's preferred US Presidential candidate did not match that of the state, and perhaps Blank's statement was written to reflect that sociopolitical disconnect between the state and its public flagship.
Despite the similar publication timing, the LSA revealed that the UC System statement and the UW-Madison statement were semantically similar (0.63), but not to the extent of semantic similarity of Blank's solo-authored statement and UT-Austin President Gregory Fenves' solo-published statement (0.73). Blank's rhetoric focused more on "community" than did Fenves who chose to focus more on "students"-here is where the outliers in Fenves' statement are important. Upon review of each statement, UT-Austin's President Fenves tied the notion of "targeting" to "students" twice in his statement, while also asserting that, "Some heard the rhetoric of the campaign and fear they could be targeted because of their race, ethnicity, national origin, religion, gender, disability or sexual orientation" (Fenves, 2016, para. 1). Here, even though the LSA rendered Fenves' statement the most "student" centered, Fenves' rhetoric was metacognitive: he spoke about the dangers of the spoken word, supported by the finding that "campus" and "ideas" were collocated along with "students" and "targeted." Although Blank's and Fenves' semantics were similar, Blank focused on the impact of the election on the community, while Fenves focused on the impact that individuals could have on each other, namely students who may be victims of rhetorical targeting.
It is then interesting to note that the city of Austin, Texas and its county-Travis-did not vote as the state of Texas did: Austin and Travis County voted firmly Democratic, whereas the overall state of Texas voted firmly Republican (Associated Press, 2016). Here, both Blank's and Fenves' local area voted one way, but their states voted the other. And like the Wisconsin Idea, the UT System mission statement makes a direct reference to service to state: "The mission of The University of Texas System is to improve the human condition in Texas" (The University of Texas System, 2017, para. 1). Perhaps this is why Blank's and Fenves' statements were so semantically similar: both were attempting to speak on behalf of their flagships without speaking on behalf of other system institutions or their states because of the sociopolitical differences witnessed by the US Presidential election results. Furthermore, Blank's focus on "community" may have been a precautionary measure to produce a sense of togetherness on campus, whereas Fenves' cautioned against hateful rhetoric and xenophobia perpetrated against individual campus "members": these two public flagship leaders were semantically similar but ideologically dissimilar.
Therefore, it is arguable that solo-authored leadership statements (e.g. post-election statements) may be composed fundamentally differently than a collaboratively authored statement meant to represent multiple leadership perspectives and contexts (every Chancellor of each UC System institution). Supporting this theme of differentiation between solo-authored and collaboratively authored leadership statements, the word frequency and collocation analysis of each statement and the corpus were telling. Both solo-authored statements (UT-Austin and UW-Madison) featured the words "we" and "our" as the two most frequently used, yet the collaboratively published statement (UC System) featured "our" and "we" as the third and fifth most frequently used words. Furthermore, each solo-authored statement focused on an individual campus, evidenced by the word "campus" being the third and fourth most frequently used word in the UT-Austin and UW-Madison statement. Instead of individual campuses and individuals, the UC System statement semantically focused on the system as a whole, citing the importance of the UC System "mission," collocating this term with the word "our," meant to represent the entire UC System instead of an individual campus. In these cases, it is clear that the solo-authored leadership statements were composed to address the concerns of individual campuses in Madison and Austin, whereas the collaboratively authored leadership statement was composed to encompass the concerns of California in general.
In addition, both UT-Austin and UW-Madison's statement contained frequently collocated verbs-"will" for UT-Austin and "value" for UW-Madison-a whereas the UC System statement frequently used the noun "principles" and collocated the noun "mission" with the pronoun "our" in lieu of speaking to any particular action. Possibly, these semantic strategies speak to the potential freedom granted to individual campus leaders (e.g. UT-Austin's President and UW-Madison's Chancellor) to encourage and perform action on their respective campuses, whereas the UC System statement semantically deferred to its mission instead of speaking to action. However, the word frequency, collocation, and LSA make it clear that action-through the frequency and collocation of verbs-was prioritized by the institution-level leaders and not the system-level leader in their respective postelection statements.
Also curious was the use of "community" in each statement. The UC System statement most frequently collocated the word "community" with "university" and "California," yet UT-Austin most frequency collocated the word "community" with "members." Similarly, UW-Madison frequently used the word "community" and frequently collocated "community" with "we," "our," and " diversity." Here, both institution-level leaders and their solo-authored statements reveal much about the attitudes toward "community" held by these institution-level leaders juxtaposed with the system-level leader in this study. For instance, UT-Austin's statement focused on the concerns of individual "members" of the community, supported by the high frequency of the word "members" and the collocation of the words "responsibilities" with "students" and "targeted" with "students." Moreover, UT-Austin collocated "campus"-a collective noun, physical place, or surrogate for a term like "community"-with "ideas," which can be held by groups or individuals. In addition, UT-Austin's statement was the only one to frequently use three pronouns-"we," "our," and "they"-possibly meant to speak to individual members on the UT-Austin "campus" instead of the state of Texas or the UT System. Similarly, UW-Madison collocated "our"-a possessive, collective pronoun-with "campus," semantically connecting people to a specific place or group of people. Here, both President Fenves and Chancellor Blank focused on their local campus communities but defined them differently. Furthermore, President Fenves' semantics reacted in defense of possible hateful rhetoric on UT-Austin's campus, while Chancellor Blank's semantics reacted in a way to bring her campus community together.
On the contrary, the UC System frequently used more broad terms such as "university" and "California" and most frequently collocated "our" with "mission" and "we" with "university," instead of referencing groups of people or "ideas," "responsibilities," and "diversity" held or embodied by individual campus stakeholders as the UT-Austin and UW-Madison statements did. However, it is interesting to note how important the "mission" was to the UC System in their statement, however a mentioning of the state of California or its many campuses is absent from their actual mission statement (Regents of the University of California, 2017). Here, the linguistic analyses point to the institution-level statements serving a purpose to unite a campus community by connecting people to places instead of abstract ideas, whereas the system-level statement spoke about their constituency as united by a "mission" and "principles." Semantically, the solo-authored statements of Chancellor Blank and President Fenves spoke much more to individuals and their sense of belonging on their respective college campuses than did the UC System collaboratively authored statement.
Finally, the LSA revealed that the UT-Austin statement could be perceived by the general public as the most student-centered, evidenced by the "Near Neighbor Analysis": "students" and "student" were the two most semantically similar terms between the UT-Austin statement and the "General Reading" semantic space provided by the LSA website. Furthermore, the UW-Madison statement most spoke to collective priorities, evidenced by the most semantically similar words featured in the post-election statement: "community," "concerns," "our," "civic," and "ourselves." In speaking to collective priorities, UW-Madison's statement was most semantically representative of the corpus (0.94) given the "One-to-Many" LSA, followed closely by UT-Austin's (0.89), yet as previously mentioned, UT-Austin semantically focused on students, not a "community" in the "Near Neighbor" LSA. Coupling these findings with the fact that UT-Austin's and UW-Madison's statement were more semantically similar (0.73) than separately compared to the UC System's statement (0.66) and (0.63), another theme emerges: the institution-level statements semantically equivocate students as members of a collective community or campus, whereas the system-level statement defined the entire system as belonging to a collective community. Again, the post-election sociopolitical climate in each state could help explain these differences. The UC System's many local campus areas voted firmly Democratic, yet UW-Madison and UT-Austin's local campus areas voted firmly Democratic while their states-Wisconsin and Texas-voted Republican (Associated Press, 2016). Perhaps Chancellor Blank and President Fenves chose to focus on campus community members because of their sociopolitical similitude given the election results. Inversely, President Napolitano's collectively authored statement addressed broader constituencies that reflected both the UC System mission statement (Regents of the University of California, 2017) and the voting preferences for UC System campus areas (Associated Press, 2016).
This semantic "zooming out" by the system-level leader in this study is potentially problematic for individuals on the many UC System campuses, as the "issues" expressed by UT-Austin and the "concerns" expressed by UW-Madison were not semantically addressed by the UC System in their collectively authored statement. The word frequency and collocation analyses along with the three-pronged LSA collectively emphasize the fact that the UC System spoke about a collective of universities, not individual groups of people or people themselves, potentially placing the priorities of the system ahead of the priorities of the people operating within that system or people holding sociopolitical beliefs counter to that of the system. The inverse is true about UT-Austin and UW-Madison's statements: these two post-election statements semantically addressed individuals and the groups to which they belong to a much greater degree than the UC System's statement. This divergence rests with the semantics the UC System used to define their audience. It is likely that UW-Madison's statement was most representative of the corpus because it spoke to university priorities and community needs-as did the UC System statement-yet was more semantically similar to UT-Austin's statement because of its pronoun-centered diction and emphasis on verb usage, instead of the UC System's echoing of "principles" and "mission" which do not specifically speak to any campus-level constituency. Here, Chancellor Blank and President Fenves used their platforms to address the immediate concerns of their campus community instead of broadly addressing the entire state university system or the state itself as their system mission statements do.
The platforms of the post-election statement also support the notion that the institution-level leaders chose to speak to their campus community, whereas the system-level leader addressed the entirety of the system. Napolitano and the UC System chose to publish their collectively authored statement on the UC System website (Napolitano & Chancellors of the University of California, 2016) and frequently referred to the California community as connected to the mission of the UC System. Inversely, Chancellor Blank and President Fenves chose institution-specific modes of communication-Blank's institutionally sponsored blog (Blank, 2016) and Fenves' campus-wide email (Fenves, 2016)-and frequently used diction that referred to their specific campus members (students, faculty) as comprising the entirety of their campus community. In this study, there was a clear difference between not only the semantics of the statements but the avenue of communication between the system-level leader and the two institution-level leaders in how to communicate with a campus community after a controversial and divisive election. The implications of this finding will be discussed later in this study.

Limitations
This study was chiefly limited to the focus of post-election statements made by three executive leaders of public flagship universities in the United States: conclusions cannot be drawn about private, for-profit, or community college institutions or institutions in other countries, nor can my findings be generalized to encompass all public flagship leaders and system-level leaders.
Also, the variance in post-election statement length affected the word frequency, collocation, and LSA of each statement and its relation to the corpus. Furthermore, textual analysis was limited to the text of the post-election statement alone, as institutions can express their beliefs and attitudes about the election through other media (e.g. promotional videos, marketing materials, social media): these media were not included in the data collection. However, these limitations represent areas of further research and analysis.

Implications and conclusion
Although limited to three post-election statements, this study found that the purpose and function of a post-election statement made by three different postsecondary leaders varied depending on how each leader defined a campus community and how best to directly communicate with their campus community while operating as part of a state-wide university system. The linguistic analyses of these statements revealed three critical themes of a post-election statement.
First, every leader defined their campus communities in different ways. The UC System statement spoke about the overarching university system and connected it to the state of California, while the UT-Austin statement spoke about a campus community as comprised of individual members and UW-Madison articulated a campus community as one collective entity. This is especially important to note consider the sociopolitical climate in each state shortly after the election: the UC System statement maintained California's "blue" unity despite unfavorable "red" election results, the UT-Austin statement made clear that campus community members may feel threatened or marginalized given the radical voting differences between the "blue" city of Austin and the "red" state of Texas, and the UW-Madison statement urged the campus community to bond together as the state of Wisconsin witnessed a "purple" election result. Therefore, future research could address how higher education leaders address their campus communities after traumatic, divisive events such as the 2016 Presidential Election to learn how these leaders define their campus community and prognosticate the complications of such an event.
Second, the data in this study demonstrate that system-level mission statements did not influence campus-level executive rhetoric. Although the Wisconsin Idea and the Mission for the University of Texas System reference each system's state as being a beneficiary of the public flagship's labor, neither Wisconsin's public flagship leader or Texas' public flagship leader used semantics to address their state. Again, the election results may have constrained how each of these leaders could address their campus community without speaking for a state that did not vote according to the public flagship's local area. Subsequently, future research could expand Morphew and Hartley's (2006) and Wilson et al.'s (2012) studies of institutional mission statements and connect these works to the rhetoric of institutional leaders. Surely, the mission statement of a public flagship university should articulate its goals and values, but if these goals and values are not expressed in the rhetoric of the public flagship's executive leader, the importance of the flagship's mission statement may be compromised. This work could also embrace the mission statements and executive rhetoric of other types of institutions, such as community colleges, private institutions, and the for-profit sector.
Finally, although "we" and "our" were the two most frequently used words in the corpus of statements, it is clear that the institution-level leaders prioritized action that can be taken to mitigate a crisis and the system-level leader did not, hearkening back to Tierney's (1988) organizational theory ideology. The linguistic analyses employed in this study partially supported Tierney's claim that university leaders are much more reactive than proactive in the promise of action through semantics: in this study, the system-level statement only reiterated their mission statement, while both institutional-level statements spoke to action. This is especially important when discussing the hateful, divisive experiences felt by some on college campuses shortly after the election (Jaschik, 2016). For instance, Chancellor Blank's statement included a word-for-word articulation of UW-Madison's campus statement on diversity, equity, and inclusion: her statement was much more semantically focused on action than the UC System's statement was, even though the UC System statement also included a word-for-word articulation of "UC's Principles Against Intolerance." This articulation of a mission or values statement is not a call to action. The "Near Neighbor" LSA is most telling in this regard: UT-Austin semantically prioritized "students," UW-Madison semantically prioritized "community," and the UC System semantically prioritized "California." Simply put, the data in this study demonstrate that post-election statements made by the two public flagship leaders articulated a course of action to protect and uphold their standards of their idiosyncratic notions of campus community than did their system-level counterpart. Implied in these themes is a call to all members of a campus community: students, faculty, staff, and all educational stakeholders of a public flagship institution should call upon their executive leaders to articulate specific courses of action to remedy any campus community concerns, issues, and inequities. A well-articulated mission statement is wonderful, especially if it accurately and honestly reflects the priorities and goals of a given institution, yet that mission statement cannot perform action. People act. And too often, these actions are in response to a crisis, not in prevention of it. It is equally important that all educational stakeholders demand that their system-level and institutional-level leaders use their words to communicate the values and beliefs of their respective institutions to the best of their abilities, but more importantly, those words need to address individuals and the groups to which they belong, along with the specific action that a leadership consortium plans to take given a traumatic, divisive event such as the 2016 Presidential Election. Future studies must semantically analyze system-level and institutional-level statements-of any kind, not merely post-election statements issued by public flagships-to ensure that postsecondary leaders of all levels not only support the mission and vision of an institution but also promote action to make all college campuses truly inclusive, diverse, and safe. Semantic analyses help quantify the qualitative, which can lead university leaders to reflect upon not only their words, but their actions as well.
Ultimately, this study is not meant to measure which post-election statement was semantically superior: it was meant to analyze how three post-election statements were written, to whom they were addressed, and which values they conveyed. As Carl Jung once articulated, "You are what you do, not what you say you'll do." This quote best surmises this study and the necessary call to action issued by countless educational stakeholders to their executive leaders across the country. Regardless of the color of the state-blue, red, or purple-leaders of public flagship institutions owe their students action beyond semantics. Now, it is entirely possible that these three post-election statements will ease tension and work to remedy some of the communicative issues facing university leaders as articulated by Tierney (1988), Rabovsky and Rutherford (2016), and Douglass (2016). It is also entirely possible that these statements will help assuage divisive political climates on each university's respective campus(es) and general communities.
Yet, semantics without action renders a leadership plan little more than words.

Funding
The author received no direct funding for this research.

Citation information
Cite this article as: The semantic reactivity of red, blue, and purple: A linguistic analysis of post-election statements made by executive leadership of three public flagship universities, Zachary Wayne Taylor, Cogent Education (2017), 4: 1391927.