Gauging Eve: A Study on the Effect of Appearance Cues on the Attitude towards Women

The 1960s and 1970s saw the development of various theories on gender development. Money and Ehrhardt (1972) on questioning intersex parents, found social factors to be more important than biological factors in the formation of gender identity and gender roles and brought the nature-nurture debate to the forefront. At the core of gender differences in behaviour are the societal stereotypes about gender (Eagly & Wood, 2012). Stereotypes are considered to be based on kernels of truth (Prothro & Melikian, 1955) and participants are surprisingly accurate at judging gender differences that are found and supported in metaanalyses (Hall & Carter, 1999).


Introduction
The 1960s and 1970s saw the development of various theories on gender development. Money and Ehrhardt (1972) on questioning intersex parents, found social factors to be more important than biological factors in the formation of gender identity and gender roles and brought the nature-nurture debate to the forefront. At the core of gender differences in behaviour are the societal stereotypes about gender (Eagly & Wood, 2012). Stereotypes are considered to be based on kernels of truth (Prothro & Melikian, 1955) and participants are surprisingly accurate at judging gender differences that are found and supported in metaanalyses (Hall & Carter, 1999).
Biological arguments have long been made to justify gender inequality (Shields, 1975) and are often interpreted as deterministic. Current biological approaches instead emphasize the ways in which biological and social factors interact to produce behavior. Some of the most active research in this area has been on girls with Congenital Adrenal Hyperplasia (CAH), a genetic disease in which the foetus is exposed to elevated levels of androgens. Researchers have found that girls with CAH tend to be masculinized in some aspects of their preferences and behaviour (Berenbaum and Snyder, 1995). Studies of prenatal exposure to normal variations in hormones such as testosterone (Cohen-Bendahan, van de Beek & Berenbaum, 2005), and cross-species comparisons (Alexander and Hines 2002;Wallen, 1996) are gaining ground. As the originally bipolar view through the biological lens shifts towards a more inclusive theory of gender, it echoes sociological theories by Constantinople (1973) and Bem (1974) which propose that both men and women have masculine and feminine qualities. Bem (1974) further argues that being psychologically androgynous was optimal for psychological adjustment.
Such socialisation and cognitive approaches highlight the shaping of children's behaviour to match cultural gender role norms. Social role theory refers to the shared beliefs applied to people in certain roles. Based on this theory, gender roles beliefs are arguably formed based on the social roles we see men and women occupy (Wood & Eagly, 2010). Across cultures, men and women behave differently, and this creates the stereotypes we have about each gender.
These perspectives emphasize the role of the social environment, especially reinforcement and modeling of adults and peers, relative to cognitive developmental processes, such as the emergence of children's gender identity and knowledge of gender stereotypes. Bussey and Bandura (1999) describe some cognitive information-processing mechanisms, such as selective attention, forming cognitive representations, and forming plans of action, that mediate observational learning. Cognitive theorists describe the ways in which children interpret and respond to messages provided by socialization agents, such as peers (Ruble, Martin & Berenbaum, 2006). These interpretations, if not corrected or diversified, result in developing stereotypes.
A stereotype is the cognitive component of the attitudes of a person towards a social entity which includes their beliefs about what the entity is like. Stereotypes are manifestations of the attitudes held by people, which Eagly and Chaiken (1993) define as "a psychological tendency that is expressed by evaluating a particular entity with some degree of favour or disfavour." Prejudice is the affective aspect of the attitude of an individual (Baron & Branscombe, 2006). This present study focuses on the analysis of these gender stereotypes. According to gender stereotypes, men are more agenticassertive, competitive, and dominant. In comparison, women are more communalselfless, egalitarian, and emotionally expressive (Witt & Wood, 2010). Gender roles appear to be rooted in society's division of labour, whereby people observe men and women engaging in different roles (Eagly & Wood, 2012). Due to their strong concordance with biological sex, traits then become viewed as inevitable, leading to the assumption that there are differences in nature between the genders. However, this is a fundamental error; assuming that people are what they do (Wood & Eagly, 2010). The distribution of men in higher status roles and women in lower status roles also contributes to this bias (Eagly & Steffen, 1984). Media portrayal and folklore also perpetuate a gender stereotype that, through repeated observation, becomes effortlessly merged with gender (Wood & Eagly, 2010). Stereotypes often lead to negative rigid beliefs about women. These beliefs contribute to a larger feeling of prejudice against them, which in turn have a role in gender based violence and bias. Glick and Fiske (1996) conducted research on ambivalent sexism, which is a theoretical framework including two differing attitudes towards women, with hostile sexism and benevolent sexism, both being equally damaging. Hostile sexism is aggressive sexism directed towards women who do not conform to societal standards and expectations of femininity. It includes accusing them of emasculating men. Benevolent sexism, on the other hand, is the idea that women need to be protected and taken care of (Glick and Fiske, 2011). This kind of benevolent sexism is reflected in prison sentences. Women are judged more favourably than men in their criminal behaviour and more lenient sentences are meted to attractive women (Ahola, Christianson & Hellström, 2009). This is aimed towards maintaining the status quo in gender norms and upholding the stereotype of soft and fragile women.
As the world moves towards a more egalitarian future focusing on the fluidity of gender and gender roles in service and academic discourse, it is observed that much of these changes are gaining hold in the western world, and there is an observable lack of research and discourse on the fluidity of gender in the Indian subcontinent. Contrary to the developments in USA and Europe, gender based violence and discrimination seems to be on the rise (Kamdar, Kosambiya, Chawada, Verma, & Kadia, 2017), the question about rising beliefs in stereotypes and increasing gender-based bias appears.
The variables have been chosen for this study (Clothing, makeup and hair length) because presentation is one of the most important non verbal cues apart from body language, distance (proxemics) and physical environments/appearance, of voice (paralanguage) and of touch (haptics). They are used to make judgements about people in our everyday lives.

Clothing: Formal and Casual.
Clothes are cultural artefacts, embedded in current and historical sets of meanings, shaped by social and economic forces, reflecting current social and cultural concerns (Twigg, 2009). Clothes are means of visual communication that have been governed and policed throughout history against a backdrop of appropriateness and conformity to social norms. One's clothes display status, power, personality and other aspects about one's individuality. People have expectations regarding clothing in different contexts and violation of these norms elicit negative reactions from onlookers (Dunbar & Segrin, 2012). Clothing influences impressions of others (Albright, Kenny, & Malloy, 1988;Forsythe, 1990;Reid, Lancuba, & Morrow, 1997) and how others are treated (Darley & Cooper, 1972;Suedfeld, Bochner, & Metas, 1971).
In research on clothing styles, the appearance cue of clothing style has often been a variable of analysis, and researchers have studied attitudes by manipulating styles of clothing (Angerosa, 2014). Thourlby (1978) states that educational background and moral character is solely judged upon clothing. In addition, judgments about one's credibility, likability, interpersonal attractiveness, and dominance are made by clothing (Molloy, 1988;Raiscot, 1986). The studies are based on simple interactions or just judging by appearances and thus, first impressions are studied (Buckley, 1983;Lennon & Miller, 1984). Cavallaro (1998) claims that if clothing can successfully "fashion" the individual by reinforcing the gender roles required of him or her, the individual may also be able to challenge those prescribed gender roles through the conscious manipulation of gender deviant clothing. As demonstrated by the above mentioned research, an individual's clothing can significantly manipulate the perceptions and subsequent behaviour of the onlookers. Due to the absence of considerable research of clothing on attitudes held by young adults in the Indian context, the variable of 'clothing' is varied and the effects of the attire are observed.

Cosmetics: Makeup and No Makeup.
As women grow up, transformations keep occurring within their lifestyle. Due to the prevalence of traditional gender roles, women are expected to be concerned about their own appearances (Drakuli, 1993(Drakuli, , 1996. Wearing cosmetics can be considered as one of the many different ways for enhancing one's appearance. Faces are often very important when interacting with others because an impression is often formed and based on facial features (Tuk, Verlegh, Smidts & Wigboldus, 2008). Moreover, facial characteristics tend to influence human judgments owing to whether they are attractive or not (Fink & Neave, 2005). Mulhern, Fieldman, Hussey, Leveque and Pineau (2003) suggested that cosmetics can be used in different ways, implying that different levels of makeup can be done to manipulate the facial characteristics, thereby becoming a useful tool to enhance one's attractiveness.
The makeup used in the current study include foundation, eyeliner, mascara, and a dark shade of eyeshadow. For the lips, a dark coloured lipstick and lipliner was used. According to Mulhern, Fieldman, Hussey, Leveque and Pineau (2003) and Winter (2005), using foundation helps to cover blemishes, protects the skin from drying out, giving it a soft, smooth and healthy look. For eye makeup, eyeliner, eyeshadow and mascara are often paired together. Eyeliner gives shape to the eyes and accentuates them. Eye shadow is used to highlight the lid and give it some colour. Mascara is used to lengthen, darken and thicken the eyelashes. The combination of the three can change the appearance of the size of the eyes. As for lip make-up, lipsticks, lip pencils and gloss help in shaping and colouring the lips.
It may be apprehended that there would be different types of beauty standards across different cultures, due to different and varied conventions. However, Fink and Neave (2005) showed that judging and rating the attractiveness of females among different racial groups did not appear to be too different, indicating that raters were able to agree as to who, or who is not attractive between and within cultures.
According to Graham and Jouhar (1981), both females and males rated the female targets with facial make-up as people who were cleaner, tidier, more feminine, more physically attractive, more secure, sociable, interesting, confident, organized and popular. Similarly, Cox and Glick (1986) discovered that increased usage of makeup positively correlates with the perceptions of attractiveness, femininity, and sexiness, but does not or negatively correlates with likeability, morality, emotionality, and decisiveness. Moreover, increased cosmetic usage negatively correlates with women's ability in women-dominated jobs and either negatively or does not correlate with women's ability in non-gendered jobs. Makeup is often used to increase attractiveness and Patzer (1985) proposed that attractive individuals are perceived as more successful in both work and school settings and are considered to be happier, and as ones having higher salaries than the less attractive individuals.
However, a distinction can be made on the effects of people's attitudes on use of makeup and attractiveness, as increased use of cosmetics can have detrimental effects on the perception of a woman's prowess in the workplace. Kyle and Mahler (1996) conducted research to determine if cosmetic usage and women's hair colour affected people's perception of their credentials and abilities in a professional setting. According to their findings, female applicants who wore more glamorous makeup were considered to be less capable than female applicants wearing little or no makeup. Further, females wearing no makeup were assigned a higher starting salary than those wearing light to moderate amounts of makeup, indicating that the key here lies in the intensity of the look, rather than the mere usage of the cosmetic products or attractiveness.
This area is in the want of recent research that focus primarily upon the use of cosmetics and the stereotypes associated with its use in more diverse populations. Hence, the present research aims at focusing upon this dimension in context to stereotypes and prejudices held by individuals in the Indian context.

Hair length: Short and Long.
Rose Weitz describes hair as "part of a broader language of appearance, which, whether or not we intend it, tells others about ourselves" (Weitz, 2004). Head hair is considered similar to dressing style as it can be modified and styled with ease, and like dressing style, hair provides important information about a person's self and group identity. Women spend much more time and money than men for styling and maintaining their hair (Manning, 2010). Pergament (1999) found that judgments and inferences about a person's sexual orientation, morality, religious sentiments, political persuasion and in some cultures, even socioeconomic status can often be surmised merely by seeing a particular hairstyle.
As a result, there are many stereotypes that exist based on hair, especially hair length. For instance, women with shorter hair are more likely to be perceived as "masculine" whereas women with longer hair are more likely to be perceived as "feminine" (Manning, 2010). This stereotype seems to be a cross-cultural stereotype or an archetype (Bereczkei & Meskó, 2006). Professionally, short hair is preferred for women in the workplace; as women with short hair are seen as intelligent, knowledgeable, mature, and confident (Jorgensen, 2014).
In a study conducted by Eyssel and Hegel (2012) in Germany it was found that gender stereotypes affect the social perception of not only humans but also of robots. The shorthaired robot was perceived as more masculine and agentic than the long-haired robot, who was perceived as more feminine and communal. Traits like authoritative, assertive, determined, aggressive, cold, organized, confident, hard-hearted, dominant, tough were included in the dimension of agency, and traits like affectionate, empathetic, delicate, friendly, sincere, family-oriented, sensitive, cooperative, affable, polite, and sentimental were in the dimension of communion. The male robot was also considered more suitable for typically male tasks while the female robot was considered more suitable for typically female tasks. Typical male tasks included transporting goods, repairing technical equipment, guarding a house, steering machines, handcrafting, servicing equipment while typical female tasks included child care, household maintenance, after-school tutoring, patient care, preparing meals, elderly care.
It has been found that longer hair length improve the attractiveness ratings as well as the perceived femininity of women. This may be because long hair frames the face in such a way that it reduces the size of the jaw and hides the low, less pronounced cheeks which are considered unfavourable in ratings of feminine attractiveness (Meskó & Bereczkei, 2004). It can be predicted that women with long hair, are thus rated lower on traits like aggressiveness, which was confirmed by Terry and Krantz's study (1993) where women with longer hair are judged to have lower social forcefulness or power, measured by aggression and courage, and social presence, measured by dominance, extraversion and sociability. The researchers suggested that because women's long hair softens the contours of the face and makes it appear more round and "baby-ish", and young girls are more likely to have long hair than adult women, it contributes to lower perceived social forcefulness.
Although research on hair length have been done in western countries, little work has been done on perception of women's appearances in the South Asian population. Thus, the current study seeks to analyse the effect of women's clothing styles, use of makeup and hair length on the judgements young adults make on women. Using the Attitudes towards Women Scale (Gamst, Liang & Der-Karabetian, 2011), and questionnaires constructed by the authors, the attitude towards women and the changing appearance of women are studied. Because young adulthood is accompanied by an increase in people looking for employment and marriage opportunities, leading to diversification of social circles, and women's appearance may have a large influence on the perception of them by their peers. However since hair serves as a cue for gender stereotyping, the present research has focused on the relationship between hair length, and the attitudes towards women.
As is evident from the above discussion, various American and European studies have discovered similar attitudes of both men and women towards the use of clothes, cosmetics and hair length. The authors of the present study have tried to identify differences in the attitudes of women and men in the Indian subcontinent, if any. The variables chosen are the most prominent appearance cues and have been used to make judgements about people often, hence the study aims to uncover differences between men and women's attitudes towards women with varying use of clothes, cosmetics and hair length.

Hypotheses of the Present Research:
1. There will be a significant difference between men and women with respect to the perception of clothes.
2. There will be a significant difference between men and women with respect to the perception of cosmetics.
3. There will be a significant difference between men and women with respect to the perception of hair length.
4. There will be a significant relationship between attitudes towards women with hair length, cosmetics and clothes.
A quasi experimental one-group posttest only design is used with three independent variables (appearance cues) and two levels in each variable. The dependent variable (attitudes) are measured through questionnaires immediately after exposure to the independent variables.

Participants
For the purpose of the study, the sample consisted of Indian unmarried participants from Kolkata, fluent in English, of the age range: 18-25 years, which was divided into two subgroups of 30 participants (30 males and 30 females). Homogenous purposive sampling was used for sample selection to maximise the generalizability of the results to Indian young adult population.

Materials
A written consent form, with complete and relevant information about the research was signed. The information schedule used for the present research comprised of items to record the preliminary demographic details of the subjects and whether they met the selection criteria mentioned above. (Spence, Helmreich & Stapp, 1973).

Attitude towards Women Scale
The instrument used, Attitude Towards Women Scale (AWS), consists of 25 items, with has four response alternatives, ranging from agree strongly to disagree strongly. Each of the items is given a score from 0 to 3, with 0 representing the most traditional and 3 the most contemporary, pro-feminist response (Spence, Helmreich & Stapp, 1973). In scoring the items, option A was given a score of 0, option B a score of 1, option C a score of 2 and option D a score of 3; except for certain marked items where the scoring was reversed. A high score indicates a pro-feminist, egalitarian attitude while a low score indicates a traditional and conservative attitude. Correlation coefficients between the original 55-item AWS and the 25-item AWS were found to be no lower than 0.956 for either of the samples. (Gamst, Liang & Der-Karabetian, 2011).

Scale developed by research investigators.
Another scale was developed by the investigators, where for each level of Formal-Informal, Makeup-No Makeup and Short Hair-Long Hair, five photographs were taken and were then presented to raters who chose one photograph from the five displayed. Two photographs were chosen for each variable, one corresponding to each level. For each of the domains of Social, Familial and Employability, three positive and three negative statements were constructed and one out of the three positive and one of the three negative statements were chosen by experts. The domains were selected to represent the different roles of women in social situations, inside the home, and in professional settings. Positive statements included judgements like "she would be a good homemaker" in the familial domain, and negative statements included statements like "she is aggressive" in the social domain. Hence, finally two statements (one positive, and one negative) were chosen for each domain by the experts.

Procedure
Following preparation of protocol, permissions were obtained by the ethical committee of the funding body. Female volunteers within the age group of 18 to 25 years were contacted, were informed about the purpose and course of the study and about the use of their photographs and given consent forms. For each level of the variables (Formal, Informal, With Makeup, Without Makeup, Short Hair, Long Hair) five photographs were taken. These photographs were then presented to three raters, who are psychologists, who chose two photographs for each level. Finally, six out of the thirty photographs were selected by them.
Written consent was taken from 30 male and 30 female subjects that fit the inclusion criteria. After completing the information schedule, the 25 item AWS questionnaire was administered. After this, the six photographs of the women for all the six variables of informal and formal dressing, with no make-up and with make-up, and with short hair and long hair, were shown to the subjects for four seconds and they were asked to rate the statements provided on a five point Likert scale on the basis of their perception of the women in the photographs simultaneously.

Scoring, Tabulation and Statistical Analyses.
Data for the Attitudes toward Women Scale were scored according to the scoring schedule for the questionnaire. The statements prepared by the investigators were also scored using a 5 point Likert Scale with direct scoring of the positive statements and reverse scoring of the negative statements. The scores assigned to the different categories on the Likert Scale were 0, 1, 2, 3 and 4. All the scores were tabulated and statistical analyses were conducted.

Results
The objective was to ascertain whether significant gender differences existed in the perception of the appearance cues (clothes, cosmetics and hair length), the domains of the scale developed by the research investigators (Familial, Social and Employability) and the Attitudes towards women. Descriptive and inferential statistics were calculated in the form of Mean, Standard Deviation, and t test. Correlation was used to determine whether any significant relationship existed between the subjects' Attitude towards Women and the appearance cues. The findings and their interpretations are as follows:t-test was conducted to determine if any significant differences existed between men and women on their Attitude towards Women Scale (AWS) scores and their perception of the appearance cues: Informal Clothes, Formal Clothes, With no makeup, With makeup, Short hair and Long hair. No statistically significant gender differences were found on the Attitude towards Women and the perception of the appearance cues. t-test was also conducted to determine if a significant gender difference existed on the familial, social and employability domains of the scale developed by the research investigators. The results are illustrated in Table 2. No significant gender difference was found in social, familial and employability domains.

Discussion
There were no statistically significant gender differences on the Attitude towards Women Scale, the perception of appearance cues (Informal Clothes, Formal Clothes, With no makeup, With makeup, Short Hair and Long Hair) and in the social, familial and employability domains of the scale developed by the research investigators. Moreover, there was no significant correlation between Attitude towards Women and the perception of the appearance cues. There was a significant positive correlation between informal and formal clothes, indicating that egalitarian attitudes towards women in informal clothes is accompanied by egalitarian attitudes towards women in formal clothes and traditional attitudes towards women in informal clothes were accompanied by traditional attitudes towards women in formal clothes. The attitudes towards women in formal clothes also had significant positive correlation with attitudes towards women with both long and short hair.
Although the differences between the genders' attitudes towards women were not statistically significant, in regards to the social, familial or employability domain and the scores on the Attitude towards Women, there are statistically significant correlations among the attitudes towards the specific independent variables. There was a significant positive correlation between informal and formal clothes, indicating that egalitarian attitudes towards women in informal clothes is accompanied by egalitarian attitudes towards women in formal clothes. Formal clothes have also been observed to be positively correlated with short and long hair, but not informal clothes.
Attitudes towards make-up and no make-up are also positively correlated. It can be inferred that one's egalitarian or traditional attitudes towards women are not swayed by the presence or absence of make-up. The scores in the condition with the presence of make-up were further positively correlated with the scores in the conditions with short hair, as well as long hair, leading one to believe that one's attitudes towards women who wear make-up remain constant regardless of their hair length, however, the same cannot be said in certainty for a woman without make-up. Although the sample in Kyle and Mahler's study (1996) offered lower salaries to women with makeup, this decision remained constant through varying hair lengths and colours.
Since the scores in the conditions in which women wore formal clothes and makeup were significantly correlated with both scores in both conditions of long and short hair lengths, it can be inferred that hair length did not play a major role in influencing the perceptions of the respondents, when the woman was in formal clothes, but not when she was in informal clothes. The variables, cosmetics and clothes may have been the major determinants of perception or the major cues. This is in accordance with a study which found that facial attractiveness is more influential in generating associations than hair length as highly attractive faces elicited much higher scores on desirable personality traits than less attractive women with similar hair length. In other words, the effect of hair length on physical attractiveness and personality judgement is much weaker than the effect of facial traits (Bereczkei & Meskó, 2006).
The selected sample's attitudes towards women with long hair positively correlate with their attitudes towards women with short hair, which indicate minimal contribution of hair length to one's egalitarian or traditional attitudes towards women. The present study thus concurs with available literature that hair length is not instrumental in making judgements about a woman's familial, social and occupational roles and abilities.
The current study reports the changing attitudes towards women by young adults with changes modes of presentation. It was observed that both women and men held similar attitudes towards women and hair length did not play a role in changing the attitudes held by them. People biased against women with long hair, no makeup and formal clothes were similarly biased against women with short hair and makeup.
Since the present sample comprised of college students belonging to middle and upper socioeconomic classes, they had a favourable attitude towards women in the familial and the employability domain. This study is solely carried out on an Indian sample and adapted to the dressing, hair length and style of cosmetic use of the Indian population. The variables chosen and the method used for the study, have not been the focus of previous Indian studies, and is the first to include three appearance cues. This study was limited to photographs of female cisgendered models. Inclusion of non-binary and transgender models and cisgendered men with long hair and makeup, may elicit widely different attitudes from a wider and inclusive sample than financially secure college educated young adults. Further research could include people from both urban and rural areas as attitudes between them are likely to differ.