As defined by Eckert and McConnell-Ginet, “A community of practice is an aggregate of people who come together around mutual engagement in an endeavor. . . practices emerge in the course of this mutual endeavor” (Meyerhoff and Strycharz 429). We chose to focus our study on the University of Alaska Anchorage community of practice because of our various roles and connections to the university itself. We each have attended the University of Alaska Anchorage a minimum of four years and have found the dynamics among its members to be both beneficial and problematic. In the same manner, we collectively found the social implications of gendered speech patterns to be problematic, and a relatively new area of study. Our goal is to contribute to the discourse of gendered speech patterns by focusing on the University of Alaska Anchorage (UAA) as our community of practice, and assess the use of gendered terms within three characteristics among members of the community.

For the purpose of this study, we have focused on gendered terms within our chosen community of practice. Specifically, we examined gender neutral pronouns and gender specific pronouns when used in regard to profession. For this inquiry gender neutral pronouns are limited to the following words: the profession name with no gender identifying factors (for example, mail carrier), person, people, and singular they, their and them. Gender specific pronoun words are confined to profession names with gender identifying factors (for example, mailman), he, him, his, she, her and hers. It is important to note that during the course of this project we are subscribing to the social definition of gender, wherein the notion of gender is distinguished from biological sex by the social implications and expectations associated with the concept. It is also important to note that upon graphing the data, the profession titles are graphed separately due to the manner of our interview questions.

In beginning this study, we believed that there would be a general trend for masculine pronouns to be the default. As our study evolved and our different variables were outlined, we believed that there would be a general trend among women and younger individuals to use more gender neutral and feminine terms overall, while men and older individuals would tend to use more masculine terms.
 
Literature Review
 
Speculating on our findings, we found the following articles to support our assumptions. Anne Curzan’s Gender Shifts in the History of English provides valuable insight into how our language’s grammatical gender system transitioned to natural gender over time. Among many things, Curzan addresses Modern English’s use of the “generic he,” which some consider to be sexist. This is important for our research because we suspect that many of our study participants truly are using generic he or she as a gender neutral, and others are simply defaulting to their own personal experiences.

Lauren Posey’s “Gender” from the Critical Quarterly discusses the many definitions and sub-definitions of the term gender and the contentions that come from those who believe it is a social construct and those who believe it is biological. This is a valuable resource for our study because it is necessary for us to define gender in regard to our gender analysis section, so it is clear what boundaries we are working within.

Lisa Maxwell Arter’s dissertation “Children’s Perceptions of Gender as Studied Through Pronoun Use” discusses a topic very similar to ours: she examines the ways in which children perceive masculine and feminine roles in society, as well as the parameters of their subconscious gender labelling. Her results suggest that a child’s native language itself has limited influence on the pronouns they use, and that children are particularly accepting of gender equality. This provides a foundation for our hypothesis that younger individuals will be more likely to accommodate gender neutral pronouns and resist stereotyping professions by gender.

Marie Gustafsson Sendén’s article “‘She’ and ‘He’ in News Media Messages: Pronoun Use Reflects Gender Biases in Semantic Contexts” from the journal Sex Roles suggests that news media uses masculine pronouns nine times as frequently as feminine pronouns, and in an overall more neutral or positive light — while feminine pronouns are often associated with more negative contexts. She concludes that men are represented as the norm in media, and that this media influences the assumptions and stereotypes of consumers. This conclusion supports our own study, offering additional explanation for the ways our stereotypes are formed and for the prevalence of masculine pronouns.

Susanne Wagner’s article “‘Gendered’ Pronouns in English Dialects - A Typological Perspective” discusses the use of gendered pronouns in non-standard English dialects. The study looks at how two communities in particular have adapted their own dialect use of gendered pronouns into the forms used within standard English. The variables that the study looks at are similar to our own variables and it focuses on two particular communities where ours focuses on one, but the characteristics are also similar among the communities chosen.
 
Methodology
 
A thorough collection of demographic information is necessary for successful data analysis. To do this, we started our audio recordings with preliminary background questions. These included name, age, role at the university, how long the participant attended UAA, major (or role focus), what gender they identified with, how long they lived in Alaska and where, and if they had a profession outside of school. Any personal information that could potentially be used to identify a participant is kept anonymous and only used to examine the interview data in a more comprehensive manner.

To assess the chosen characteristics, discussed below, we elected to adapt a set of questions written by Dr. Clare Dannenberg. Each of the ten questions in the set asks the interviewee to describe a short story centering around an individual of a particular profession. The professions are as follows: healthcare professional, mail carrier, server, flight attendant, police officer, lawyer, trash collector, librarian, firefighter, and soldier. The questions omit any reference to gender, allowing the interviewee to fill in the space with their own assumptions, providing us with their (typically) unconscious biases. Following the final question, we went through each profession and asked the interviewee, “Is a [profession] a man, woman, or either?” to elicit their conscious biases.
 
Findings
 
For the analysis section of our paper we have decided to separate the data into three different characteristics: age, gender, and university role. We have analyzed all of the data for the use of gendered speech patterns within each of these characteristics for both the perceived gender of each individual profession considered, and for patterns regarding gendered pronouns in relaxed speech.

For age, we have divided the data into five sets separated by increments of ten years, beginning at the age of 15, and ending at the age of 69. As for gender, we separated the data into two groups based off of the subject’s provided responses: male and female, (for future more in-depth studies, a third group “other” could be added). In the same manner, for university role we categorized the data into three sections: professor, teaching assistant (TA), and student. For this study, the term “professor” covers a wide range of instructors from adjunct and term professors to tenured or tenure track professors currently instructing at the university.

For the purposes of our study, we have divided neutral pronouns into two categories: the first includes any reference to the neutral position title provided to the interviewee within the question (e.g. “mail carrier,” “firefighter,” or “server”) which we will refer to as “profession”; the second includes any other pronoun or referent that does not connote a specific gender (e.g. “they,” “the person”) which we will refer to as “neutral.” We have made this distinction to address the potential difference in impetus to use each kind, which is the same reason we avoided introducing any pronouns, neutral or gendered, within the questions or the additional prompts during the interviews: we would not be able to tell whether the interviewee was using a particular pronoun because the interviewee had introduced it in the question, or they had chosen it themselves.

For gender, this study leans on the social definition of gender, that is, gender as a social construct and sex as biological, as defined by Posey when she says, “The social definition was brought about and reinforced as a fully conscious effort to make English reflect that how a person self-identifies in relation to social factors can be completely separate from biological sex. Other results of the same idea shift include the widening use of gender-neutral third-person pronouns in the singular; for example, ‘they’ instead of either ‘he’ or ‘she’” (95).

We have chosen to focus on these three different characteristics based on the speculation that each characteristic directly influences the data. As an example of generational change, the age of the subject reflects the perceived social norms attached to gendered speech patterns in which they grew up in and thus will influence the data accordingly; our hypothesis being that the older an individual is, the less likely they are to use gender neutral terms because of the currently shifting social standards. We speculated the gender of each subject to influence the use of gendered pronouns, theorizing that an individual is more likely to use the gendered pronoun in which they associate themselves with during relaxed speech. For university role our hypothesis regards education level; the longer an individual has been a part of the academic community the more likely they are to use gender neutral terms in relaxed speech, and more likely to associate each profession as gender neutral rather than assigning a given profession a particular gender.

In order to address each hypothesis, we have divided the rest of the paper into four sections, one for each of the characteristics identified, and a conclusion. Each characteristic section outlines the initial hypothesis in more detail, provides the dependent and independent variables, analyzes the data for use of gender pronouns in relaxed speech and the perceived gender of each profession, and provides our results in both graphs and written format. In addition to our analyses based on the numbers of our results, we also include a number of observations based on interesting paratextual phenomena. We conclude our report with methodological notes, speculating on what we could have done differently, how the results may have been skewed, and how the study could be expanded for future analysis.
 
Gender
 
Among our participants, we had eight individuals who identified as men, and nine who identified as women. All participants were responsible for providing their own gender identifications. In relation to the gender of the subject, we hypothesized that their use of gender pronouns would reflect the gender they identify with. For the most part, this is not what we found. Although the data does reflect that those who identify as males are more likely to use a masculine pronoun in relaxed speech, those who identify as women on the other hand are more likely to use a neutral pronoun. Although these gender categories reflect usage of gender pronouns, there were two anomalies found in the data reflecting the gender perception of the professional in the scenario rather than the gender of the subject; the flight attendant and the librarian. The flight attendant is to be considered an anomaly because about 80% of males interviewed used a feminine pronoun in this scenario. As for the librarian, both males and females were more likely to use the feminine pronouns in relation to this profession, matching the general perception of the individual in the professional position. Graphs are provided below reflecting this data.

Age
 
Among our participants, we had six individuals who fell between the age range of 15-25, three individuals between 26-36, one between 37-47, four who fell between 48-58, and only one individual whom fell between the age range of 59-69. In relation to the age of our interviewee, we hypothesized that their use of gendered pronouns would be more likely to be feminine or neutral the younger they were. However, the data does not support this. Rather, choice in pronoun roughly follows a bell curve, in which, of our five age categories, outlined above, the youngest and oldest groups are both more likely to use masculine pronouns, while the middle three use some form of neutral identifier; the second and third groups are more likely to use pronouns reflecting their gender perception of the profession, and the fourth group is most likely to use neutral pronouns. Professions in which pronoun use stands out in particular for are trash collector and soldier: the two youngest age categories were more significantly more likely to refer to trash collectors by masculine pronouns, while the oldest age category was significantly more likely to refer to soldiers by masculine pronouns. In the case of the latter observation, this may be due to the fact that many positions within the United States military were closed to women until very recently, resulting in an exclusively male population for portions of the military.

Additionally, while age did not seem to impact choice of pronoun as much as gender or position within the university, we noticed that those who fell into the higher age categories were more likely to identify as the individual with the profession for part or all of their story, such as by speaking in first person or explicitly stating they were that character (See graph below).

University Role
 
Among our participants, we had seven students, three teaching assistants, and seven professors. All of these individuals came from various fields of study, though the majority fell within the humanities. In relation to the position of our interviewee within the university, we hypothesized that their use of gendered pronouns would be more likely to be neutral the higher their position was within the university. This hypothesis is supported by the data. Students at the university were more likely to use masculine pronouns across the board, while teaching assistants were more likely to use neutral pronouns and professors were more likely to refer to subjects by professional titles. Among those who taught at the university, those who had spent more time teaching were more likely to refer by profession. Overall, the results of pronoun use for the university role map closely with those for age; however, several key distinctive features are outlined below.

When we asked our interviewees the final set of review questions — whether each profession could be a man, woman, or either — to gauge their conscious views, professors were significantly more likely to answer with “either;” 71% percent of the answers provided by professors were “either,” while this was only 56% for teaching assistants and 57% for students.

During the course of our interviews, a handful of our participants at various points noted without prompting from the interviewer that they were using gendered terminology or were intentionally attempting to balance use of gendered characters or subvert gender norms. All of the participants who verbally acknowledged this behavior were professors, suggesting that pronoun use is an issue that is salient to professors.

Moreover, teaching assistants or younger teachers were exceptional in that their stories were more likely to display some elements of gender fluidity: this category was the only one to refer to subjects within a single story as more than one gender — to switch genders for one character during the course of a story or to identify the character as a particular gender and introduce a second character of the opposite gender who acted equally alongside the first. This category was also more likely to use two or more types of pronoun in general when they were referring to their main character.
 
Conclusion
 
After analyzing our data, our initial hypothesis was partially supported. We had anticipated that women and younger interviewees would be more likely to use feminine or neutral pronouns, as opposed to men and older interviewees being more likely to use masculine or gendered pronouns. Additionally, while we expected that masculine pronouns would be more likely overall, we believed that use of neutral or professional forms would increase with rank within the university.

Masculine pronouns were, in fact, used much more often than either feminine or neutral pronouns, though our variables of age, gender, and position did not seem to significantly affect likelihood to use masculine or feminine pronouns in particular. Unlike with masculine and feminine pronouns, use of neutral pronouns was significantly impacted by our variables: gender and position significantly affect the likelihood of using neutral or professional pronouns.

In conducting our research, there are a number of features that we found difficulty with. One of the major hurdles we ran into was inconsistency of data. In having three interviewers with no specified frontloading other than the text of the questions themselves, the length and style of the interview varied both between interviewers and between interviews. For example, in some interviews additional prompts were asked to further a participant’s story, while in other interviews they were not. As such, some interviews spoke for two or more minutes on a single prompt, while other interviewees spoke for much less than a minute. As longer periods of talking provides greater opportunity for pronoun use, this may have artificially inflated certain interviewees’ results.

Additionally, the majority of participants noted without prompting that for at least one or more of their answers they were basing the subject of their story off of a particular person that they knew, or on the types of people they had seen perform those professions in the past. This tendency for participants to filter their stories through their experiences also has an impact on the pronouns they use, and thus the results we received.

Finally, a notable phenomenon that we noticed while interviewing was that when a participant was prompted with a probing question that related to the feelings of the subject, the participant, regardless of gender, age, or position, was significantly more likely to use gendered pronouns.

Should a follow-up study be conducted, we would be interested in looking at this behavior in more depth. Additionally, we believe it would be beneficial to create a script for interviewers to read that provided instructions for the participant as well as possible prompts for the interviewer to ask. Finally, it would be desirable to find a larger sample size, as well as a more balanced sampling.
 
Bibliography
 
Curzan, Anne. 2003. Gender Shifts in the History of English. Cambridge: Cambridge University Press.
Gustafsson Sendén, Marie. 2015. Introducing a Gender-Neutral Pronoun in a Natural Gender Language: The Influence of Time on Attitudes and Behavior. Frontiers in Psychology 6. 896.

Posey, Lauren. 2016. Gender. Critical Quarterly 58. 94-96.

Meyerhoff, Miriam, and Strycharz, Anna. 2013. Communities of Practice. The Handbook of Language Variation and Change 2. 428-447.

Thomas, Tracy. 2014. Gendered Pronouns: Just Stop It. Gender and The Law. 11.

Wagner, Susanne. 2004. ‘Gendered’ Pronouns in English Dialects: A Typological Perspective. Mouton de Gruyter.

Arter, Lisa Maxwell. 2011. "Children’s Perceptions of Gender as Studied through Pronoun Use." ProQuest Dissertations Publishing.

Gustafsson Sendén, Marie, et al. 2014. "“She” and “He” in News Media Messages: Pronoun use Reflects Gender Biases in Semantic Contexts." Sex Roles 72.1: 40-9.

---