Research Proposal
Chapter 1: Introduction
Statement of the Problem
The principle of gender indistinguishability in printed literature – which is achieved by citing the surname, usually combined with only the first letter of the forename(s) – initially appears to be completely objective or neutral and therefore desirable for scientific purposes. On the other hand, it is hypothesized that the practice of using non-sex specific, conventional, scientific referencing to make women appear to be indistinct from men does not actually eliminate – nor does it contribute to transforming – the current stereotype of male dominance and prestige, and majority male participation in academics and research. Rather, this convention simply renders invisible women’s participation in academics and research. This study aims to demonstrate that rendering females indistinguishable (from males) or invisible in printed literature is actually counterproductive in terms of improving equal perception of females in academia and in destroying existing negative societal stereotypes about them.
Introduction
Need for Accurate and Straightforward Referencing
A fundamental aspect of any scientific domain is an accurate and uncomplicated referencing system for resources. The question of how to quote a resource is usually viewed as something straightforward, simple and objective. Traditionally, the choice of format for citing of resources has been decided arbitrarily by academic associations such as the American Psychological Association (A.P.A.), among others, who then disseminate these conventions and compel mandatory adherence to them.
All Systems Exist within Constraints of Society
However, the fact that every scientific domain exists within the societal context, and the realization that the format in which a resource is referenced influences and is influenced by societal predispositions, should lead academic societies to reevaluate the appropriateness of the arbitrary resource citing conventions currently in practice. Changes over time in gender roles, in how women are viewed, and in the level of female participation and or recognition in scientific domains may warrant a change in how resources are quoted in order to better reflect this evolution and to continue to progress these trends.
Professional Responsibility to Eliminate Bias
It is researchers’ duty to question whether established referencing formats in current employ are in fact the best ones for meeting the needs of contemporary society and for promoting modern ideals of equality, fairness and impartiality. Every system invented for and exploited by humans exists within human society. Extrapolating from this theme, even a purported ‘impartial’ scientific system may be subject to inherent bias or may simply result in biased outcomes or conclusions when utilized in conjunction with human frailties and prejudices. To safeguard the integrity of our systems and conventions, researchers have a moral obligation to ensure that any system employed in the gathering or compilation of data and resources has minimal influence on (i) the content, (ii) the selection criteria and (iii) the actual selection of these resources. Thus, the onus is on researchers to identify these systemic biases and eradicate them or at least to simply adjust the system to compensate for them in some way. To this end, the contribution and input of female researchers at all stages of the design process of a future referencing system – from rooting out any partiality, to its format and implementation – is not only crucial but will undoubtedly also prove insightful and transformational.
Focus of this Paper
For the purposes of this research paper, the focus will be on the arbitrary customary method in which authors’ names are mentioned in printed scientific publications. Using the American Psychological Association’s (A.P.A.) format as the standard, this is how author references are currently formatted:
- For a resource quoted within a paragraph: Surname (Year of Publication) or (Surname, Year of Publication), e.g.
Thompson (2004) or (Thompson, 2004)
- For resources listed within a bibliography: Surname, Initial Letter of Forename(s). (Year of Publication). Title. e.g.
Thompson, A. (2004). Gentlemanly Orthodoxy: Critical Race Feminism, Whiteness Theory, and the APA Manual. Educational Theory, vol. 54, (issue 1), p. 2.
Since this study is aimed at a Western, English-speaking audience, the focus will be exclusively on the effect of surname-only citation practices for surnames and forenames of Western European origin, particularly those whose origins trace back to the British Isles or names which are in commonly found in present-day Britain. Moreover, more specifically, this study will be constrained mostly to what are known as ‘traditional proper names’ (e.g. Robert in place of Rob; Evelyn in place of Evie) and ‘British’ family names that are of English language etymology or that are in frequent employ in the contemporary English-speaking world (e.g. Smith, Baker, Jones, etc.). In keeping with this choice to utilize proper names, analysis of nicknames will be avoided as much as possible. This decision is taken based on observational evidence which appears to indicate that nicknames may display a higher propensity to become gender neutral or unisex in their current use than traditional proper names. For example:
Table 1. Examples of Unisex Nicknames.
| Boy’s Forename | Girl’s Forename | Frequent Nickname (Unisex) |
| Samuel | Samantha | Sam |
| Alexander | Alexandra | Alex |
| Andrew | Andrea | Andy |
| Patrick | Patricia | Pat |
| Christopher | Christina | Chris |
Background
Logic behind Challenging Surname-only Referencing
It is worth underscoring that in the majority of cultures, and certainly in English-speaking and European-derived cultures, the way a female is distinguished from a man by name is mainly in terms of the forename(s) she is assigned at birth. In European and English-speaking cultures certain forenames are used for males, certain are used for females and a small number of forenames are unisex or gender neutral. Representative of this is that, according to Gender Checker (www.GenderChecker.com), of the 102,000 unique given names found in the United Kingdom in 2011, 45% were male only, 48% were female only, and 7% were unisex. This particularity – of the first given name being the unique marker of femininity in naming – is supported by the fact that the tradition throughout many centuries was for females to ‘take on’ the name of their father or husband, conserving merely their given names for specific identification purposes. Even in contemporary society this tradition persists and thrives: the majority of women in modern Britain and much of the English-speaking world change their family name to that of their husband upon marriage. Furthermore, by default the children of the union all carry merely their father’s last name, the mother’s family name being completely absent as an officially inherited name. What this means is that family names still have a greater association with maleness and masculinity than with femaleness and femininity. Males are still the owners and bestowers of family names. Notwithstanding this observation, standard quotation systems give the impression that they deem surname-only references to be the most objective or neutral citation format available. It may even be accurate to claim that one of the main principles in modern resource referencing systems may be the indistinguishability of female researchers or authors from male authors – an abstruseness of gender which is accomplished by using surname-only references. At first glance, this lack of differentiation may appear to be both justifiable and desirable. Contrariwise, it shall soon be made clear that family names carry within them specific indicators of maleness that cannot be easily eradicated or ignored and whose role should not be downplayed in terms of their psychological impact and the extent to which they contribute to readers’ eventual gender assumptions about the author carrying each surname.
Origins of Assumptions of Maleness from Surnames
With respect to the origins of family names, last names or surnames, many simply refer to the original bearer of the name, they are simply the first name of the ancestor in question. E.g. Thomas; Arnold; Henry; William, Cameron, Phillip, Adam. Such last names have been adapted from their role as a male’s first name and passed from father to children [patrilineally] as the family name. Other surnames mention the original bearer of the name: they contain within them the acknowledgement that the bearer being the son of someone. For example:
Table 2. Examples of Surnames Referring to a Familial Relationship
| Surname | Refers to | Meaning |
| Benson | Ben / Ben’s | Ben’s son |
| Ericson | Eric / Eric’s | Eric’s son |
| Peterson | Peter / Peter’s | Peter’s son |
| Williamson | William / William’s | William’s son |
| Williams | William’s | of / belonging to William |
| Phillips | Phillip’s | of / belonging to Phillip |
| Adams | Adam’s | of / belonging to Adam |
Unfortunately, females did not hold much status is terms of naming, which is why there are no last names formed by combining ‘daughter’ with the father’s name. Furthermore, males were usually the ones who traditionally held professions and trades and who were the main ‘breadwinners’ or financial support for the family. Concerning this, where the family name is derived from the word for a trade or occupation, the occupation would have referred to that of the male or males in the family or household. This is why we find the surname ‘Taylor’ but not ‘Seamstress’. Other commonplace last names indicative of their original bearer’s profession are: Smith, Porter, Miller, Mason, Cook, Barber, Fisher. Throughout history and to some extent today, men were regarded as the ‘head of the household’. Where the surname originated from a place name, it would be indicative of the male household head’s place of origin (or the unmarried female’s father’s place of origin). E.g. Brooke; Greene; Banks; Townsend, meaning Town’s end; French; Holland; Moore (i.e. the Moor); English; Scott; etc. Following on from the practice of the man being the head of the family, males were deemed the ‘bearers’ of the family name (or surname, as we often refer to it today). A woman was and still is usually called “the mistress of [her husband’s family name]” or ‘Mrs’. For this reason, women took – and to great extent still take – the last name of their spouse. Whereas, the reverse occurrence – where a man takes his wife’s family name and deletes his own surname from his name – is exceedingly rare. Most present-day men are forcefully opposed to this idea and at most might concede that both parties take on a hyphenation of their combined last names, e.g. Smith-Jones. In addition to this, the default is for children of the union to inherit their father’s surname. It is important to point out that this reason alone – that surnames are regarded as owned by men and passed on by men – suffices to explain why it is more usual to call males by their last name only, and females by their first name only, since this is the only name which is likely to be consistent for the female throughout her lifetime, the only name she truly ‘owns’ by the current social standards and practices. Notwithstanding, as women assume greater individual renown, and gain prestige and acclaim independent (from their male work partners, fathers and spouses) within professional and academic realms, society may now be getting closer to accepting the idea of matrilineal naming for children. Even so, this latter is not the purport of this study.
Given that patrilineal naming history intricately associates last names with males, it is unsurprising to discover that dissociating surnames from men and from concepts of manhood and maleness may be more complicated to achieve than simply spreading its scope to include women. On the surface, this may function, but, as various studies have shown, (Atir and Ferguson, 2018; Knobloch-Westerwick, Glynn and Huge, 2013) subconsciously, in spite of everything, readers may be assuming authors quoted by last name only are men (whether or not they are) and in practice far fewer women are referred to in conversation by surname only as compared to men. Resulting from this, the employment of the family name alone as a means of address for women has not been entirely successful in its spread. In other words, family names have not been transformed to incorporate concepts of womanhood and femaleness. This suggests that another tactic should be undertaken in order to transform the current preconceptions about perceived surname ‘ownership’ from males to females. One path forward may be to make females’ presence more visible, and not indistinct to that of males since, for historical and metacognitive reasons, the male presence will always be perceived as manifest within a family name.
Synopsis of What Surnames Bring to Mind
In summary, there are concrete linguistic, historic, connotative and social markers [metacognitive markers], as mentioned above, that are contained within last names themselves and or found in the ways in which surnames are utilized in English-speaking societies which allude to a male presence in most surnames. If one takes for granted that last names already automatically denote a male presence on some level, the only way to change this perception of maleness may be to make the female presence obvious. Then, simply including the first names of all authors may go a long way towards accomplishing this goal because for many English names, the forename is the unique marker of the feminine sex. Nevertheless, this goal may also be achieved in other ways. Another means of accomplishing this may be to include Mr. and Ms. (the latter of which is the neutral civil status term for addressing women) within all bibliographical references.
Note that ‘Ms.’, pronounced [miz], refers to a woman of any civil status. This may be a woman who has never been married (more precisely ‘Miss’), or to a married woman (more precisely ‘Mrs.’), or to a divorced woman. The use of ‘Ms.’ has come into favor in modern times because, like ‘Mr.’ for men, it may be used to represent a woman of any civil status. In this fashion, it does not necessitate one investigating or divulging the woman’s civil status – something that many consider personal and moreover irrelevant, especially in the professional sphere – in order to address the woman directly. Any ‘Miss’ may be called ‘Ms.’ without offense and any ‘Mrs.’ may be called ‘Ms.’ without offense, any female Dr. may be called ‘Ms.’ without offense. Since a female doctor was and is a woman and will always be one regardless of their professional status as a doctor, there should also be no offense received or perceived in calling a female Dr. ‘Ms.’ or a male Dr. ‘Mr.’.
Details of Proposed Changes
Full Name Referencing is More Accurate
Requiring that all authors be referenced by their full names and not just by their surname and the initials of their forenames is the simpler and easier tack for changing the current name citing convention. However, it should be underlined that this is in fact the less efficient pathway for tackling the gender discrimination problem because despite the fact that full names simultaneously allow for more accuracy in terms of author identification, it does not take into consideration authors whose names do not following English language naming conventions, e.g. Li, Shunlai [male]; Ho, Fang [female]. Most English-speakers would be unable to identify the sex of the authors with either of the two names quoted above, one of which is Mandarin Chinese and the other is Cantonese Chinese. For example, compare the formats for referencing the following names.:
Table 3. Comparison of Conventional versus Full Name Referencing and Formal Address
| Conventional Referencing | Referencing by Full Name | Referencing by Full Name and by Formal Address |
| Lewis, L. | Lewis, Lily | Lewis, Lily (Ms.) |
| Lloyd, H. | Lloyd, Harrison | Lloyd, Harrison (Mr.) |
| Marshall, N. | Marshall, Naomi | Marshall, Naomi (Ms.) |
| Martin, Z. | Martin, Zachary | Martin, Zachary (Mr.) |
| Mason, Y. | Mason, Yvonne | Mason, Yvonne (Ms.) |
| Mills, E. | Mills, Emma | Mills, Emma (Ms.) |
| Mitchell, P. | Mitchell, Patrick | Mitchell, Patrick (Mr.) |
| Moore, G. | Moore, Grace | Moore, Grace (Ms.) |
| Murphy, K. | Murphy, Kieran | Murphy, Kieran (unisex, viz. Mr.) |
| Shaw, J. | Shaw, Jaden | Shaw, Jaden (unisex, viz. Ms.) |
| Smith, A. | Smith, Alfred | Smith, Alfred (Mr.) |
| Spencer, Y. | Spencer, Yohan | Spencer, Yohan (Mr.) |
‘A. Matthews’ could be anyone from Aaron Matthews to Azaria Matthews. Native English-speakers will be familiar with most of the popular first names and be able to immediately identify the sex of each person referenced from them, unless the given name is unisex. In connection with this, the table above demonstrates that simply including the full names of all authors goes a long way towards permitting identification of the sex of most authors through their given names, provided they have English names. However, to make the author’s gender even more explicit, the reference could include the address ‘Mr.’ or ‘Ms.’. For all that, there is no reason why both these methods could not be employed simultaneously to further elucidate identity and gender.
Full Names and Titles of Respect are Most Precise
Undoubtedly, quoting the full name and putting the formal male or female address form or titles of respect following the name (or in brackets after it) is the most precise way to refer to someone and will make gender clear regardless of the linguistic origin of the surnames and forenames. In other words, the best practice might well be to reference full names and to add ‘Mr.’ and ‘Ms.’ to all references. This solution actually resolves the question of how sex would be identified in the case of first names of foreign, non-English language, origin. Using the formal male and female status neutral salutation terms of ‘Mr.’ and ‘Ms.’ thus appears to resolve those cultural issues. For example:
- For within a paragraph: Forename Surname (Year of Publication) or (Forename Surname, Year of Publication)
Ms. Michelle L. Dion, Ms. Jane Lawrence Sumner and Ms. Sara McLaughlin (2018)
(Ms. Michelle L. Dion, Ms. Jane Lawrence Sumner & Ms. Sara McLaughlin, 2018)
- For the bibliography: Surname, Forenames Mr./Ms. (Year of Publication). Title.
Admittedly, pertaining to the question of the choice of name referencing systems to employ, it may indeed be quite difficult to determine what (if any) citation style would be truly neutral in terms of gender preconceptions or prejudices. Furthermore, the most objective format may not necessarily be the best format for transforming gender stereotypes, as we have seen with the A.P.A. referencing convention. Nonetheless, it is imperative that we make a concerted effort to right existing inequalities and strive to make a future choice of citation system as unbiased and fair as possible.
Chapter 2: Literature Review
Use of References
In the reference list for this paper, sources could be classed into three different categories. The first category is that of ‘principal references’. These are papers published in scientific journals within the last seven years that have been subjected to peer-review. The second category is that of ‘important older references’. This category of sources is similar to the first in that these are peer-reviewed, scientifically published papers. Its dissimilarity from the preceding category arises in that these references are significantly older – some dating back almost twenty-five years. Despite the age of these references, as a consequence of the dearth of materials available on this paper’s research topic, these older references have become integral to this current research paper. The third and last category of references is that of ‘additional references’. These are mostly online and public data sources covering demographical population statistics and dealing with the frequency of surnames and forenames. Such sources also served as a cornerstone for this research work since without this data it would be difficult if not impossible to compose a survey to assess the public’s perception of gender through names. Sources of all three of the classifications used in this study have been amalgamated and may be found within the References section, arranged in alphabetical order.
Patterns Observed in Published Studies
Men and Women are Similarly Biased
The most surprising pattern revealed in studies investigating persons’ tendency to assume male authorship for published works was that women and men behave similarly in terms of preferring men over women in terms of their choice of works to cite. In the research study by Dion, Lawrence Sumner and McLaughlin (2018), it was not clear to what extent the women and men surveyed may have been aware (or unaware) of the sex of the authors they were citing. With current referencing conventions i.e. where it requires some effort to discover the sex of the author of a source, it is quite possible that the authors’ genders were largely unknown to those citing their works. This might explain why both men and women writers displayed similar rates of quoting other male and female authors. Another noteworthy trend revealed from studies is that gender favoritism occurs at all levels of society and within every single social interaction, regardless of whether those acting or interacting are male or female. For this reason, it may be difficult to eliminate or diminish unfair practices caused by gender bias in academia. Three different studies support this conclusion: one by Kaatz and Carnes (2014); one by Savigny (2014); and another by Oreskes (1996). Looking at this claim from the perspective of daily life, one could take the instance where there are two equally qualified candidates for a position: one of which is a woman and the other is a man. Unfortunately, in many cases female voters may be predisposed to vote for the man. Women are therefore responsible for voting men into power or voting against women. This occurs as a consequence of the women who are voting not reflecting on the impact of their actions in the big picture. They respond to and satisfy their subconscious (though conditioned) feelings of greater comfort in having a man lead which result from their familiarity with such an arrangement from having seen it repeatedly throughout their lifetime. Some may argue that this is just personal preference, even so, this is how gender prejudice against females is also perpetrated by women: because women too have been brought up seeing and internalizing unequal structures and feel most at ease with that with which they are accustomed. Accordingly, women are no less susceptible to acting out on these ingrained prejudices than are men.
Issues Raised in Existing Literature
The study by Dion, Lawrence Sumner and McLaughlin (2018) raised the issue that making women distinguishable from men could be counterproductive because it could facilitate discrimination towards women. If it is obvious at a glance which author is a man and which is a woman, writers could choose to cite men (consciously or subconsciously) rather than women due to their own prejudices. This same study found that, in general, men are cited more frequently than women. In cases where men form the minority in terms of the numbers present in a particular domain, men are indeed cited less than women. All things equal, women do cite other women at a higher rate than men cite women. Be that as it may, men are still cited more than their percentage presence in any domain would lead one to expect. These trends are perpetrated and perpetuated by both men and women. This may imply that, perhaps when able to identify the author’s sex, women do make some concerted effort to quote women more frequently, whereas, men may be ignorant to the importance of trying to combat any kind of gender favoritism in the selection of their articles.
Constructive Criticism
Constructive criticism for this study might be that the researchers appear not to have enquired as to whether writers knew the gender of all the authors they were citing. This is relevant because with the current citing and referencing conventions being upheld, sex is not immediately apparent. As such, one may need to make a concerted effort to discover the gender of each author, perhaps by investigating to find out what the first names are, unless of course the authors themselves are already known to the writer. In other words, feedback on improving previous studies is that the initial question to ask is “Were the authors aware they were citing men or women when they cited them?”. In this way, it could be established whether women knew the sex of many of the authors they were citing and what bearing this knowledge may have had on women’s citation rates of other women.
Risks in Highlighting Women’s Presence
Another related issue mentioned in research by Atir and Ferguson (2018) and by Knobloch-Westerwick, Glynn and Huge (2013) is that making women unmistakably distinguishable from men could be counterproductive. This is because it could promote the tendency for men to appear more prestigious from the attitude taken towards male authors when their gender is known in contrast to how women are generally treated and referred to when their gender is known. Specifically, men tend to be mentioned by surname only, whereas women tend to be referred to by full name (i.e. first name and surname) or even sometimes by first name only.
To illustrate this, Hillary Clinton was often called merely ‘Hillary’ rather than ‘Clinton’. If one just says ‘Clinton’, most presume the person being mentioned is United States President Bill Clinton. As such, during the American presidential election Hillary Clinton was often cited simply with her first name ‘Hillary’. In contrast, her husband Bill, was rarely ever referred to simply as ‘Bill’. Taking talk show host Oprah Winfrey to serve as the figure of a woman who is independently famous (i.e. independently from any man). One may ask the question: How much more is she referred to as ‘Oprah’ rather than ‘Winfrey’? Now take the case of David Letterman, another famous talk show host. How much more is he referred to as Letterman than as David? When these trends are examined in conjunction with the fact that one of the main conclusions of the studies by Atir and Ferguson (2018) and by Knobloch-Westerwick, Glynn and Huge (2013) is: that the way we speak about others influences and is influenced by the way we think about them, we then realize how detrimental these everyday occurrences of sexism can be. As such, changing our approach and the way we speak about people so as to consciously decide how we will envisage them and how we will allow them to be perceived could make a substantial difference to the state of affairs. Some people may insist that it would be simplest to decide to call all people by their last name only – as is essentially done in the A.P.A. convention. Even so, it may be unrealistic to try to change centuries or perhaps millennia of tradition in men’s ownership of family names and this may lead to no adjustment in outcome. A more realistic approach might be to decide to refer to everyone by their full name and standardize that as the norm.
All the same, what is more damaging is that studies by Atir and Ferguson (2018) and Burbules (2013) have shown that being referred to by family name only (as men tend to be referred to) leads the audience to have a more prestigious impression of the person being discussed. Hence, there is a certain circularity in the prestige accorded to people who are cited by surname only. At very least, it could be drawn to the public’s attention that it is not as respectful to call women by first name only and that men are given unjustified prestige when referred to by only their last name. In this way, it may encourage the practice of talking about persons of all genders by only their full name. The latter study also touches on the topic of the Matthew Effect which is where work done receives greater acclaim when performed by a man compared to the level of recognition it would achieve had it (the same work) been conducted by a woman.
A critical caveat to this proposed study is that making females more visible may not actually do anything to dissociate surnames from concepts of maleness and may not lead to more women being cited or to greater recognition of women in scientific domains. A study by Knobloch-Westerwick, Glynn and Huge (2013) discusses the Matilda Effect where work done by and achievements made by women are attributed to men (either deliberately, or unintentionally and thoughtlessly). This, the Matilda Effect (or women being imagined to be men), is one of the main eventualities that this research paper wishes to avoid in making women’s presence more obvious in printed academic work.
Major Conclusions of Existing Literature
On the whole, existing literature on the topic of citation rates of men in contrast to women has brought forth the suggestion that it is crucial for women to contribute their thoughts and opinions to all scientific and academic structures, including citing conventions, organization of power and decisions on what is truly objective or neutral or not. (Research detailed in the following studies by Kaatz and Carnes (2014); Savigny (2014); Savigny (2017); and Oreskes (1996).) Some information sources insinuate that women may not have been contributing their fair share to the design of these structures and providing sufficient feedback on how established conventions may be affecting them. Apropos, it is essential that women be involved in the process of deciding on anything that affects them, and since everything affects them, women should be involved in all decision-making. In addition to this, every decision should be analyzed to determine whether or not it progresses, hinders or retrogresses ideals of gender equity in academia, as this is the only manner in which these entrenched stereotypes can be eradicated or transformed.
Summary of Literature Review
In summary, gender favoritism and prejudices are implicit in every single social interaction, regardless of the gender of those involved. In order to transform the existing order and inequalities, gender bias must be first identified and acknowledged. Leaving blame out of the equation could assist in the acknowledgement of such prejudices, especially since research shows they are perpetrated by males and females alike. It is key to realize that lack of blame does not equate to lack of responsibility. Vis-à-vis, we must come up with strategies and techniques for combatting gender inequalities because if nothing is done to try to change the situation gender bias will continue to unfairly affect how people are perceived in terms of prestige and how much exposure they receive. Lastly, these ideas must be tried out to see if they will work as temporary or long-term solutions to reducing gender partiality. Although it may be difficult to decide on which referencing format to choose in order to promote ideals of gender equity and fairness, and even if citing full names may initially be counterproductive for females, any effort is better than just leaving the existing status quo intact. To this end, women should be integrally involved in contributing to solutions at every level of decision-making and implementation of structures.
Chapter 3: Methodology
Purpose
Thesis Topic
This study will endeavor to demonstrate that typical referencing practices (viz. those that utilize only surnames and forename initials) do not change the public’s mental stereotypes about researchers and academics being predominantly male. In fact, these practices may further contribute to mental assumptions of maleness, possibly erroneously attributing (at the subconscious level) some publications written by females to males in the readership’s eyes.
Investigating How Bias Influences Assumptions
This research study is being conducted in order to learn to what extent preconceived ideas enter into people’s assumptions about the gender of academic authors. Which is to say, “How much more do people of either sex tend to imagine that researchers are men rather than women?”. Various themes will be investigated during the course of this study.
Research Method
Data Collection Strategies
Responses will be collected on the supposed sex of a list of fictitious researchers. Each survey will contain two types of lists: one list where it will be impossible to know each researcher’s sex based on the way their name is provided, which is surname and first initial or initials (of forename), and a second list where the gender will be more or less obvious (to most native English-speakers) as the person’s full name will be provided, and respondents or participants can deduce the sex from most English given names. The responses for these two lists will be compared.
Data collection will be quantitative. The main data collection strategy will be to utilize surveys with binary responses (‘male’ or ‘female’) to questions. Respondents will simply tick one option or the other, presented side-by-side in two separate columns. This response format and the tick option means that participants will spend an absolute minimal amount of time answering whether each name is ‘male’ or ‘female’. Responding with a tick mark means that respondents spend less time answering than it would normally take them to write the letter ‘M’ or ‘F’. Moreover, the time interval for participants to answer is probably equivalent or even less than the time taken to circle a letter ‘M’ or ‘F’ within a multiple choice answer format. It is hoped that this rapid binary response option and method for recording the responses will force responders to record their instinctive automatic first response, which it is anticipated will reveal their inner or subconscious biases. The logic in this is that if participants require less time transferring their responses from their minds to the paper, this should translate to less time for them to reconsider or change their (initial) responses. Besides which, to further discourage participants [or respondents] from editing their first choice responses, responders will be supplied only with non-erasable pens. In this way, it is hoped that this quantitative data collection method will prove to be fairly objective.
Consistency of Sample Set
Unfortunately, no previously existing data has been found which specifically targets the phenomenon of gender misassignment based on ambiguous name referencing in academia. Consequently, the methodological constraints of the study have been designed in accordance with this researcher’s own judgment as to what would be relevant. It has been decided that it would be best to collect data from young adults and even from minors as these persons may reveal future trends of society’s gender assumptions and prejudices based on ambiguous name referencing and may have less experience of or exposure to societal gender biases as a result of their younger age. In other words, young adult participants’ gender standpoint may be more fluid and less entrenched than those of older persons. However, it is worth emphasizing that this opinion is simply a conjecture and may be completely misguided. Nonetheless, whether misguided or not, selecting younger responders should not significantly affect the reliability of the data collected. Specifically, the data set will be limited to participants between the ages of fifteen and forty-five years of age.
For ease of access to a quantity of willing participants from which to choose, and for the previously stated reason about using younger persons, the intention is to gather the data on a local university campus. In terms of the gender distribution among responders, the objective is to collect fifty complete [viable] surveys from males and fifty from females, totaling one hundred viable surveys. Unfortunately, should a potential participant declare themselves as transgender, they will be asked to choose the sex with which they feel they most closely identify. If they object to choosing a gender, they will be informed that this survey was not designed for and cannot accommodate participants of an uncertain gender: the purpose of this research is to discover whether existing preconceptions about gender in a certain context is dependent upon the sex of the responder. Obviously, if the responder has no gender, this would nullify the point of the research.
Additionally, the objective is to limit the useable sample to monolingual native English-speakers. The purpose of this objective is to control for the influence or interference of linguistic knowledge from other languages in participants’ answer choices. For instance, in Spanish, almost all words and names ending in the letter ‘o’ are masculine and those ending in ‘a’ are feminine. It may be surmised that a bilingual English and Spanish-speaker who is ignorant of the Greek myth about Odysseus and Calypso, might guess incorrectly that ‘Calypso’ is a boy’s name based on their Spanish linguistic knowledge of names ending in ‘o’ usually being masculine. Conversely, a monolingual English-speaker who is also ignorant of this Greek myth may have no such linguistic theory to interfere with their assumption that the name ‘Calypso’ is male or female, and similarly with girls’ names such as ‘Coco’, ‘Indigo’ or ‘Juno’. This hypothesis would also apply in an opposite exemplification for boy names ending in ‘a’. A monolingual English-speaker may see nothing usual about the boy’s names ‘Luca’, ‘Mica’, ‘Ezra’, ‘Aria’ and ‘Elijah’, whereas a Spanish-speaker might believe these were girl’s names because of the final ‘a’ or final ‘a’ sound. It is hoped that trying to reduce the interference of additional linguistic knowledge in this way will help maintain the validity of the data collected from the sample set. Still, since the university campus where the data will be collected is located in Hatfield, Hertfordshire, England, the United Kingdom, it is expected that most of the eligible responders will be of the following ethnicities: white English, black English and East Indian English [called Asian or South Asian in the U.K.]. Since only monolingual English-speaking respondents are eligible to participate, it is expected that the data set may consist in few or no first-generation or second-generation British citizens. But, a person’s citizenship is completely irrelevant to this survey; it is merely necessary that they be monolingual native English-speakers.
Summary of Sample Set
In summary, eligible participants for the sample set will be subject to the following constraints: They are constrained by age to persons between fifteen and forty-five years of age; by gender to persons who are able to choose a specific gender for themselves as male or female; and by language ability, to persons who speak English as their native language (whether British-English, Scottish-, Irish-, American-, Australian-, Caribbean- or any other type of English) and importantly, to persons who do not speak or know any additional languages.
Data Collection Tools
Design and Composition of Surveys
The surveys consist in lists of the names of an equal number of supposed* male and female researchers. (* ‘Supposed’: because the researchers’ names are in fact fictitious). Most human populations consist in an equal proportion of males to females, or roughly fifty percent females and fifty percent males. Although it might have been a bit more accurate to employ sex-specific and gender-neutral first names with the same frequency as they are shown to occur in the general population in the United Kingdom, viz. 48% female, 45% male, 7% unisex (as at 2011; current figures unknown); it is unjustifiable to possibly skew the results by having unequal numbers of males and females in the names lists on the survey, or worse, to possibly confuse respondents with too many unisex names. As such, the figures have been adjusted so that the same number of male and female forenames is utilized, both at 48%, and less unisex names are used, reduced to 4% with 2% female unisex and 2% male unisex first names. Which, in any case is consistent with the ratio of an equal number of males to females in the population. In this way, the reliability of the data collected should be safeguarded.
Please go to the Appendix to see the full progression towards the creation of the final lists used. The contents of the lists are names from the one hundred and fifty most common family names and first names in the United Kingdom as obtained through public data sources. Family names were chosen based on the criteria of having English linguistic origin. i.e. Family names of foreign language origins such as ‘Singh’, ‘Ahmed’, and ‘Begum’ would be excluded from the survey list even though they may be found within the leading one hundred plus surnames in Britain. The one hundred last names chosen were selected with little concern for including all letters of the English alphabet; they were simply the top family names. As such, several letters of the alphabet are missing from the list and several are repeated. Precisely, to increase the variety of last names drawn on, certain last names with several similar variations were also excluded based on the policy of not wanting to use too many similar last names for the list. Despite this latter policy, it was decided to leave some similar family names on the lists in order to preserve the validity of the sample of the hundred or so most frequent English surnames. Although some similar pairs made it onto the list, it was decided to pick only one surname in each of certain pairs or groups of similar family name, thereby reducing the number of similar pairs of last names on the survey and increasing the alphabetical range of the first letters of the surnames. (Specifically both Baker and Barker, Davies and Davis, Harris and Harrison, Thomas and Thompson were sourced as surnames in the lists even though the surnames in each pair are very similar. Whereas, only one name was employed from the groups of similar common surnames: Allan or Allen, Adam or Adams, Brown or Browne, Clark or Clarke, Cook or Cooke, Dixon or Dickinson, Green or Greene, Morris or Morrison, Richards or Richardson, Roberts or Robertson, William, Williams or Williamson.)
Since first names are rarely gender neutral, it was necessary to assign a sex prior to selecting each forename. Correspondingly, once the last names were chosen and arranged in decreasing frequency, a gender was assigned alternately down the list, starting with the male sex. In the part of the survey where only the first initial of the forename is used, it was still important to assign a gender to each researcher, even though the names are fabricated because it was felt that for the sake of greater authenticity each name should be as realistic as possible. It should also be stressed that the final order of the genders of the researchers invented for this study will be completely random, such that even if the responder were to go down the list assigning the alternate gender to each name, it is unlikely they will achieve a score with a high level of accuracy. It is precisely the assignment of gender at this early stage in the design of the survey that permits such a level of randomness to emerge on the ultimate list. (Please see the penultimate paragraph in this section of the Methodology for an explanation of the presentation of the finalized list after alphabetical sorting of surnames.)
Conversely as compared to surnames, forenames were selected with due consideration to including all the letters of the alphabet. This strategy of using the entire alphabet for the first names was decided on to ensure that the first names and first name initial letters would be quite diverse, but would still be sufficiently familiar to participants. The technique utilized to combine surnames and forenames was to list the top hundred-odd family names by decreasing frequency, then assign letters of the alphabet in sequential order to each family name, using the letters Q and X only once for each gender. The English alphabet has twenty-six letters. Thus, for the list of one hundred names it worked out that the alphabet was used four times (except for the letters Q and X which were availed of only twice). Once a letter of the alphabet was assigned to each surname, the most common first names for each respective letter of the alphabet were found for the relevant sex. From this pool of the leading given names, usually the most popular first name was chosen for each letter, unless the forename was clearly not of British origin or if it was not actually a proper name, in which case it was converted to a proper name. For instance, the leading boy’s first name was ‘Alfie’, but this is actually a nickname for ‘Alfred’, thus, ‘Alfred’ was utilized instead. By the same token ‘Oliver’ replaced ‘Ollie’; ‘Maxwell’, ‘Max’; ‘Charles’, ‘Charlie’; and ‘William’, ‘Will’. It was decided to leave ‘Nathan’ in place of ‘Nathaniel’ because, although, strictly speaking, ‘Nathan’ is a nickname, it has become so popular that it could now be judged a proper name. A similar judgement was made about the girl’s names ‘Bella’, the diminutive of ‘Isabella’, and ‘Katie’, the short form of ‘Katherine’, which have both also become proper names in their full right. Nevertheless, it was decided to convert ‘Bella’ to ‘Isabella’, but to leave ‘Katie’ as is. Other than these items, there were no top girl’s nicknames which needed to be converted into proper name format.
The last condition for choosing a name from the list of the most popular forenames for each letter and gender was that of having only a limited number of unisex forenames. All the same, rarely was the top first name for any letter of the alphabet gender neutral. Thus, it was necessary to specifically search out unisex first names with the aim of meeting the required percentage quota of forenames of ambiguous gender.
Another reason for employing this approach of using the entire alphabet to create the master list was in case certain letters may have connotations of a particular sex with certain participants. As a case in point, take ‘J’ connoting ‘Jay’, a boy’s name; or ‘B’ as short for ‘Beatrice’, a girl’s name, etc. This method was not utilized in choosing the last names because the surnames form the focus of this study – whose premise is that family names alone contain metadata alluding to maleness. It was therefore vital to draw from the one hundred or so most common English surnames with which all participants should be familiar.
Table 4. Extract from the Most Common English Surnames
| Most Common Surnames by Frequency | First Initial | Most Common First Names | Gender | |
| 1 | Smith | A. | Alfred | M |
| 2 | Jones | B. | Bernadette | F |
| 3 | Williams | C. | Charles | M |
| 4 | Brown | D. | Daisy | F |
| 5 | Taylor | E. | Edward | M |
| 6 | Davies | F. | Florence | F |
| 7 | Evans | G. | George | M |
| 8 | Wilson | H. | Harriet | F |
| 9 | Thomas | I. | Isaac | M |
| 10 | Roberts | J. | Jessica | F |
| 11 | Johnson | K. | Kevin | M |
| 12 | Lewis | L. | Lily | F |
| 13 | Walker | M. | Maxwell | M |
| 14 | Robinson | N. | Nancy | F |
| 15 | Wood | O. | Oliver | M |
| 16 | Thompson | P. | Poppy | F |
| 17 | White | Q. | Quin | M/U |
| 18 | Watson | R. | Ruby | F |
| 19 | Jackson | S. | Stuart | M |
| 20 | Wright | T. | Terry | F/U |
| 21 | Green | U. | Ulysses | M |
| 22 | Harris | V. | Violet | F |
| 23 | Cooper | W. | William | M |
| 24 | King | Y. | Yvette | F |
| 25 | Martin | Z. | Zachary | M |
Key: M means male; F means female; F/U means unisex name but the person it represents is female in this case; M/U means unisex name but the person it represents is male in this case.
Source: https://www.SoFeminine.co.uk
Notes: (1) Meaning of ‘Common English Surnames’: Names such as Patel, Lee, Khan, etc. were deleted from the list for name origin reasons because they are not of English linguistic origin. (2) Name deletions: Names such as William and Evan were deleted from the list for their similarity to Williams and Evans.
The final step in readying the lists for responders was to sort the surnames in ascending alphabetical order. In this way, the alphabetically assigned first names would now seem completely random, though varied and comprehensive, and the last names are simply in alphabetical order as one would expect for any bibliographical list.
Each participant will receive a double-sided sheet of paper containing a list of names flowing from one side to the other. The paper will contain names numbered from 1 to 50 on one side and 51 to 100 on the other. In this way, respondents know where to start. In addition to answering the implied question about the gender of the researchers named on the paper, responders will need to state at the top of the paper their sex, age, the languages they speak and their ethnicity. All participants will respond with their guess on the sex of the same one hundred named researchers. The only difference will be that different participants will do so based on seeing the first fifty names on the list in the format ‘Surname, Initial Letter of Forename’ and the last fifty names on the list in the format ‘Surname, Forename’ and another participant [or responder] will answer based on seeing the first fifty names on the list in the format ‘Surname, Forename’ and the last fifty names on the list in the format ‘Surname, Initial Letter of Forename’. The final survey sheets employed in this research study may be found in the Appendix.
Data Collection
Procedure for Data Collection
Data collection will be a one time occurrence. The data will be collected personally during one week at the university near the researcher’s residence, which is the University of Hertfordshire in Hatfield, Hertfordshire, England, in the United Kingdom. Permission to conduct the research may need to be obtained from the university’s dean or administrative staff. Clipboards and pens will be handed out to participants. The survey sheets were printed in a mixed fashion so that the person who takes the next survey sheet from the pile will do get a survey with the first set of names set out in a different format from the survey sheet of the previous person.
Sensitive Information Gathered
Certain sensitive information will be gathered from participants. This includes their gender, age, civil status, level of education, ethnicity, and finally, whether they speak any foreign languages and if so, which ones. In spite of this data being gathered, no additional personally identifying information such as name, date of birth, address, etc. will be collected from respondents, accordingly, there will be never be a means of relating any of the sensitive data to a specific person. A legend will be used to record this information.
Table 5. Key for Inputting Data into Excel Spreadsheet
| A | B | C | D | E | F | G | V |
| Age | Batch Number | Civil Status | Degree | Ethnicity | Foreign Languages Spoken | Gender | Validity of Particular Survey Sheet |
| [any value between 15 to 45] | Identifying number on the survey sheet | V = single, never married | 1 to 5 years of completed university | B = black | N = no | F = female | N = no |
| Date | C = common law marriage | H = high school diploma | W = white | Y = yes | M = male | Y = yes | |
| Versions 1 to 4 of the survey sheet | E = engaged to be married | C = college diploma [a 2 to 3 year institution] | Y = South East Asian (e.g. Chinese, Korean, Thai, Japanese, Vietnamese, etc.) | [If Y, please specify which] | |||
| M = married | B = bachelor’s degree | I = East Indian or South Asian | |||||
| S = separated | M = master’s degree | R = Aboriginal, Native American, Native Indian, First Nations or Pacific Islander | |||||
| D = divorced | D = doctorate [PhD] | X = Mixed races or mixed ethnicities |
Information Provided to Participants
Participants will be told only that the survey contains names of a list of researchers. In truth, all names listed are fictitious. Real researchers’ names will not be sourced, nor will reference titles be quoted in order to avoid influencing responders’ opinions based on such extraneous data, for instance, in the case where a participant may have heard of the researcher and is familiar with them. Be that at is may, it is absolutely critical to communicate clearly to all participants that all names on the list represent researchers. A study which evaluates responders’ tendency to assume any name is male or female would still be a beneficial study, but it would not elucidate the full extent to which persons of both sexes are predisposed to believe that an academic or a scientist is a man – which is the goal of this study. In the same vein, so far as possible, name combinations that would allude to some famous person or company will be avoided. As a case in point, some of the initial random surname and forename assignment combinations resulted in the names of famous people or clothing chains, e.g. the name ‘Samuel Jackson’ (American actor) was changed to ‘Stuart Jackson’; ‘Bella Jones’ (a French clothing company) was changed to ‘Bernadette Jones’; and ‘Henry Lloyd’ which is akin to ‘Henri Lloyd’ (a men’s clothing company), was changed to ‘Harrison Lloyd’.
Respondents will be provided two lists of fifty names in continuous ascending alphabetical order. Participants will be asked to decide on the gender of each researcher (or name) in the list. A sample size of at least one hundred useable surveys would be judged an acceptable initial sample size for the collection of data on this topic. Each survey will have one hundred responses: fifty responses where it will be impossible to know the researcher’s sex and fifty responses where either a gender-specific or gender-neutral first name is used. Responders will tick the relevant column, either ‘male’ or ‘female’ according to what they believe is the researcher’s sex. There will be no option for ‘unknown’. Respondents will be obliged to decide on a sex or not respond by leaving the response cell blank. Nonetheless, participants will be discouraged from leaving any cells blank. What is more, in order to offset any preference of choice that may arise from having the ‘male’ column first, there will be an equal number of survey sheets where the ‘female’ column comes first. Moreover, there will also be an equal number of survey sheets where the list of fifty names in standard reference format comes first and where the fifty names in full name format comes first. All participants will be warned repeatedly that the column headings for ‘male’ and ‘female’ are switched over leaf. Only two directives will be announced to participants: (1) to answer as rapidly as possible; (2) to watch for the switched gender columns overleaf.
A column has also been included for respondents to justify their response (if they so choose) by providing the gender and or given name of someone they know with the particular first name or with the particular initial, and which makes them think of that sex when they see that specific letter, that specific forename or that specific name combination. As an illustration of this, although ‘Charlie’ is usually thought of as a boy’s name, someone may know a girl named ‘Charlie Smith’ which causes them to think of a female when they see the name ‘Smith, Charlie’, ‘Smith, C.’ or just the first initial ‘C.’. Nonetheless, respondents will only be asked to look over their sheet and add these notes after they have completed the entire survey to avoid tuning in their mental radar to the fact that the study is all about gender discrimination before they have filled their answer choices.
Collation of Data
The data collected will be quantitative. If it is feasible to obtain several electronic tablets or Kindle Fire tablets, an electronic survey could be designed for respondents to answer directly on these tablets. This would enable instant storing and collation of the data. In the alternate situation, then once all completed hard copy survey sheets have be gathered, the information therein will need to be input into by hand into an Excel spreadsheet prepared with complex formulas that will count and tally all responses. On the one hand, at the most basic level, these calculations will provide a percentage response to the question of the number of participants who decided each name was ‘male’ or ‘female’ and the respective sex of those respondents. On the other hand, a more complex revelation might be that of which group of people seems the have the strongest preconceived notions about gender roles pertaining to scientific researchers.
Chapter 4: Results and Analysis
Proposed Analysis of Findings
Analysis of the data will be quantitative. Results will be tallied into table format, collated and then graphed. Beyond this, ‘if, then, and’ formulas will be used to calculate the amount of skew generally by gender, age, ethnicity, civil status and level of education. Likewise, calculations will be made in order to unearth the relationships that may exist between several variables at once, as in the skew found in presumptions made by ‘young men’; ‘older women’; ‘less educated men’; ‘black women’; etc. On a more sophisticated level, an optimal outcome would be the clarification of the interplay between from three up to five variables at once. To elucidate this last statement, combined categories would be those such as: ‘young white women’; ‘older divorced East Indian men’; or ‘middle-aged married Arab women with a high school education’ – the latter of which category takes into account all five of the variables being collected in this study.
Anticipated Bias in Results
It is anticipated that a certain amount of bias will be inevitable in the administration of the study itself. Specifically, those participants who respond on survey sheets where the names in full name format are to be answered first may become clued into the fact that the survey is interested in discovering responders’ gender biases. This is because participants may notice that there are an equal number of male and female names on the first list to which they responded. Those respondents who realize this may be conscious of this detail when answering the second side of their survey sheet. As a result of this awareness, they may find themselves trying to adjust their instinctive response to compensate for what they imagine to be their own ingrained prejudices. It is then likely that such respondents may overcompensating the other way.
Another way in which error is expected to creep into this study is through the cultural background and knowledge participants introduce into the study. It is impossible to separate a person from their cultural upbringing, so, it goes without saying that the cultural education each participant has will be carried with them wherever they go and will be brought into effect whenever they answer any question whose underpinnings are essentially cultural and societal in nature, as are all the questions in this study. Be that as it may, since little or nothing may be done to minimize the consequences of this cultural influence, it may be best to ignore it, while at the same time noting it as a possible factor contributing to any inaccuracies in the results.
Expected Results
Though it would be scientifically satisfying to be able to collate some of the data on all five factors simultaneously, the researcher imagines this will actually prove impossible. The reason for this assumption is that the language restriction imposed on this study will likely lead to a very small number of ethnically diverse respondents being eligible to participate in this research study. This restriction was deemed necessary in order to control for possible interference from additional linguistic knowledge affecting the study. For example, if, ‘Una’, one of the girl’s names used in the survey, sounds like or turns out to be a boy’s name in the Arabic language, this would be an interference that this research study would not be equipped to evaluate or quantify. An Arabic speaker might know this, but this researcher would not. Prior knowledge and linguistic interference are both contributing elements that would be very challenging to measure under any circumstances. In attempting to keep the sample group as homogeneous as possible, it is hoped that more valid results will be obtained that could then be compared to results from future studies with a wider scope and less restrictions applied.
It is expected that the category of person who will be the most biased towards designating researchers as males is ‘older married white men with the least amount of education’, and the category with the most bias towards designating researchers as females will be that of ‘younger never married white women with the most education’. At this point in the study, it is almost impossible to imagine which category of person may be the least biased towards either gender: one may hesitatingly offer the guess of ‘younger never married females with little education’. The foundation of this guess is the idea that any female with significant exposure to the structures present in the world, including to academic and research structures, would be aware that there are likely to be more male researchers than females because of the obstacles stacked against women and the challenges they still face today regarding the progression of their careers.
Chapter 5: Conclusion
Conclusion
In conclusion, it is believed that the present research study is unique in that, to date, it is the only study that has as its stated aim to specifically uncover the presumptions of maleness generated within the human mind through surname-only referencing. Although, it may well be the case that this particular study is not up to the task of clarifying the full extent of things in that area – an outcome that would not be surprising given that it is well-known that the human mind is a complex and confusing arena to try to decipher, – even so, it is an momentous achievement to have attempted to do so.
In continuation of the theme from the point above, even if this study proves unable to quantify or to draw out responses to the two main queries of this study, which is to say: (1) whether surname-alone references engender skewed judgements about the gender of the individual whose surname is cited; and (2) which is the specific group of persons who displays the most or least gender bias; still, at the very least this research will have sensitized the public about the issue of people’s internal presumptions potentially reinforcing generalizations, and it will also have informed the public that, at any rate, the current standard of surname-only citations does nothing to transform the status quo.
Considerations of Next Steps
Presentation of Results
The finalized results will be presented to Teach-Now. It is hoped that, subsequently, the American Psychological Association (A.P.A.) will one day also honor this researcher with the privilege of perusing and contemplating the results of this pilot study.
Impact of this Research
The main objective of this study is to influence existing academic bodies into analyzing and considering the impacts of some of the structures they endorse, even something so unassuming as the way they recommend authors’ names be referenced for printed resources. A slight change in the existing directives of such influential organizations could clearly reveal the partiality in choice of citing materials which is currently being concealed to the detriment of fair opportunities for female academicians. Such an exposure could cause a revolution from which female academics might benefit, and therefrom all scientists will benefit because for the first time they will access a truly diverse range of published works. This humble researcher aspires only to be the trigger that initiates this movement of change for the betterment of all.
References
Atir, Stav & Ferguson, Melissa J. (2018). How Gender Determines the Way We Speak About Professionals. Proceedings of the National Academy of Sciences, vol. 115, pp. 7278-7283. Retrieved from https://roberta345.files.wordpress.com/2019/03/34280-atirferguson_pnas_2018.pdf
Baby Name Wizard. Browse Names Beginning with a Letter of the Alphabet. Wild Sky Media, Inc. Retrieved from http://www.babynamewizard.com/baby-name
Baby Names. Baby Names Starting with the Letter. Retrieved from https://www.babynames.co.uk/letter/
Burbules, Nicholas C. (2013-11-12). The Paradigmatic Differences Between Name/Date and Footnote Styles of Citation. Educational Research: Material Culture and Its Representation, vol. 8, pp. 191-199. Publisher: Springer, Cham. Retrieved from https://link.springer.com/chapter/10.1007/978-3-319-03083-8_13
Coates, Richard & Hanks, Patrick. (2016-11-17). The Oxford Dictionary of Family Names in Britain and Ireland. Arts and Humanities Research Council (AHRC). University of West England (UWE), Bristol. Oxford University Press. Oxford, England, United Kingdom. Retrieved from https://www1.uwe.ac.uk/cahe/research/bristolcentreforlinguistics/researchatbcl/fanuk.aspx
Dion, Michelle L.; Lawrence Sumner, Jane & McLaughlin, Sara. (2018). Gendered Citation Patterns across Political Science and Social Science Methodology Fields. Political Analysis, vol. Doi:10.7910/DVN/R7AQT1. Retrieved from http://michelledion.com/files/DSM2018PAfinal.pdf
Express. (2016-09-01). Revealed: Most popular first names and most common surnames of last 500 years. United Kingdom. Retrieved from https://www.express.co.uk/news/uk/706354/Most-popular-first-names-most-common-surnames-last-500-years
Garnham, Alan; Doehren, Sam & Gygax, Pascal. (2015). True Gender Ratios and Stereotype Rating Norms. Frontal Psychology, vol. 6, (issue 1023). Published online 2015-07-22. Doi: [10.3389/fpsyg.2015.01023] Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4510832/ and https://www.frontiersin.org/articles/10.3389/fpsyg.2015.01023/full
Hanks, Patrick; Richard Coates & McClure, Peter. (2016-11-17). Family Names of the United Kingdom (FaNUK). The Oxford Dictionary of Family Names in Britain and Ireland (Oxford University Press). University of West England (UWE) Bristol Press. Retrieved from https://www1.uwe.ac.uk/cahe/research/bristolcentreforlinguistics/researchatbcl/fanuk.aspx
Kaatz, Anna & Carnes, Molly. (2014-06). Stuck in the Out-Group: Jennifer Can’t Grow Up, Jane’s Invisible, and Janet’s Over the Hill. Journal of Women’s Health, vol. 23, (issue 6), pp. 481-484. Doi: 10.1089/jwh.2014.4766. Published online 2014-05-20. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/24844292 and http://europepmc.org/articles/PMC4046346
Knobloch-Westerwick, Silvia; Glynn, Carroll J. & Huge, Michael. (2013). The Matilda Effect in Science Communication: An Experiment on Gender Bias in Publication Quality Perceptions and Collaboration Interest. Science Communication, vol. 35, (issue 603). First published 2013-02-06. https://doi.org/10.1177/1075547012472684 Retrieved from http://journals.sagepub.com/doi/abs/10.1177/1075547012472684 and https://www.kcl.ac.uk/ioppn/depts/addictions/Diversity-Inclusion/Matilda-effect-Science-Communication-2013-Knobloch-Westerwick-603-25.pdf
Office for National Statistics (ONS). https://www.ons.gov.uk/
Oreskes, Naomi. (1996). Objectivity or Heroism? On the Invisibility of Women in Science. Osiris, vol. 11, pp. 87-113. Won the Margaret W. Rossiter History of Women in Science Prize in 2000. Retrieved from https://www.jstor.org/stable/301928?seq=1#page_scan_tab_contents
Savigny, Heather. (2014). Women, Know Your Limits: Cultural Sexism in Academia. Gender and Education, vol. 26, no. 7, (issue 794). Bournemouth University, Bournemouth, U.K. Retrieved from http://eprints.bournemouth.ac.uk/21459/1/G%26E%20final%20version.pdf
Savigny, Heather. (2017-11). Cultural Sexism is Ordinary: Writing and Re‐Writing Women in Academia. Gender, Work & Organization, vol. 24, (issue 6), pp. 643-655. First published 2017-07-05 https://doi.org/10.1111/gwao.12190 Retrieved from https://onlinelibrary.wiley.com/doi/10.1111/gwao.12190
So Feminine. List of the Most Common Surnames Nationwide. Au Feminin, United Kingdom. Retrieved from https://surname.sofeminine.co.uk/w/surnames/most-common-surnames-in-great-britain.html
Thompson, Audrey. (2004). Gentlemanly Orthodoxy: Critical Race Feminism, Whiteness Theory, and the APA Manual. Educational Theory, vol. 54, (issue 1), p. 2. First published 28 June 2008 https://doi.org/10.1111/j.1741-5446.2004.t01-5-00abs.x Retrieved from https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1741-5446.2004.t01-5-00abs.x
van den Brink, Marieke & Lineke Stobbe, Lineke. (2009-06-22). Doing Gender in Academic Education: The Paradox of Visibility. Gender, Work & Organization, vol. 16, (issue 4), pp. 451-470. Retrieved from https://doi.org/10.1111/j.1468-0432.2008.00428.x
Appendix
roberta345 