Discussion question

Press Start Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 URL: http://press-start.gla.ac.uk

Press Start is an open access student journal that publishes the best undergraduate and postgraduate research, essays and dissertations from across

the multidisciplinary subject of game studies. Press Start is published by HATII at the University of Glasgow.

Conjuring the Ideal Self: an Investigation of

Self-Presentation in Video Game Avatars

Holly Maxwell Pringle

University of Glasgow

Abstract

Self-presentation in online spaces has recently attracted a significant

amount of attention in psychological literature. Video games allow

players to create a detailed, unique character to represent themselves in

the online social world. Research has found that there is a relationship

between self-esteem and online self-presentation. However, little

research has examined gender differences within this topic. The study

aimed to address this gap in the literature by specifically examining

gender differences in avatar creation and how this extends to gameplay

choices, while confirming the previously noted effects of body self-

esteem on avatar creation. Forty participants created an avatar in The

Elder Scrolls Online and completed questionnaires on General Self-

Esteem and Body Self-Esteem plus an evaluation of their avatar. Results

found that body self-esteem predicted perceived avatar similarity and

gender affected class choice. Limitations and directions of future

research are discussed.

Keywords

self-presentation; avatars; psychology; social psychology; impression

management; online behaviour; self-esteem; MMORPG; The Elder

Scrolls Online

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 2 URL: http://press-start.gla.ac.uk

Introduction

Self-presentation

Self-presentation, or impression management, is a regulation of social

behaviour with the intention of controlling the perception of one’s image.

It was first described by Erving Goffman in his 1959 book “The

Presentation of Self in Everyday Life”, using analogies to theatre to

explain how individuals are like actors when they self-present. It was

then defined as a social behaviour by Edward Jones in the 1960s, and

has been studied extensively in social psychology since.

It is said that the concept of self is important as it can influence, and be

influenced by, social interaction (Schlenker, 1980; Leary & Kowalski,

1990). It is argued that a certain degree of self-presentation is essential

for successful social interaction, and it is not simply a sign of vanity,

shallowness or insecurity (Leary, 1995). It has been consistently found

that individuals will try to present the most positive version possible of

themselves (Alexander & Knight, 1971; Bromley, 1993). Ultimately, it is

thought that it is the desire for social approval that drives self-

presentation efforts (Arkin, 2013; Jellison & Gentry, 1978).

Self-presentation in online environments

With the explosion of social media use in recent years, literature has

shifted focus towards impression management online. Currently, 83% of

internet users use social networking sites, and computer mediated

communication (CMC) accounts for a quarter of the time spent online

(Brenner, 2013). Impression management is particularly relevant to

online environments, as publicness significantly increases motivation to

self-present positively (Leary & Kowalski, 1990). Websites and

applications such as Facebook, Instagram and Twitter provide a platform

for individuals to project themselves, and it can be their real self or an

idealised self they choose to project.

As early as 1996, Walther’s Hyperpersonal model of communication

predicted that the reason individuals positively self-present online is

because the technology allows unlimited time and editing, meaning that

they can be highly strategic about self-presentation. Online

communication allows the individual to consider their response more

carefully, as they do not need to respond instantaneously as they would

in a spoken conversation. Editability means they can revise their

message, or delete it (Walther, 2007). This means negative aspects can

be downplayed, and information presented selectively to portray a

positive image (Ellison et al., 2006; Bibby, 2008). It is for these reasons

that it is significantly easier to present the best possible self online than

it is in real life.

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 3 URL: http://press-start.gla.ac.uk

Early research on impression management online examined the use of

personal websites as a form of self-presentation (Schau & Gilly, 2003).

Chat rooms and bulletin boards were also examined and it was found

that users often pretend to be someone else or act out negative

impulses in the online world that they cannot in real life (Rheingold,

1995; Surratt, 1998; Turkle, 1995). Users of chat rooms were found to

embellish personal characteristics such as age, intelligence, income and

marital status (Caspi & Gorsky, 2006; Cornwell & Lundgren, 2001;

Whitty, 2002).

More recently, research attention has shifted to social media sites such

as Facebook. It is stated that self-presentation is a major motivation to

use social networking sites (Kramer & Winter, 2008, Nadkarni &

Hofmann, 2012). It has been found that some users will self-present by

only showing the most attractive photos (Ellison et al., 2006; Manago et

al., 2008; Strano, 2008). Facebook users rated their profiles as

portraying them as better than reality on qualities such as “funny”,

“outgoing” and “adventurous”, and believed that their profiles presented

them more positively than reality (Toma & Carlson, 2015).

Self-presentation in video games

Video games are another form of media which can encourage conscious

or unconscious impression management. Many current games allow

players the freedom to create a very detailed and unique avatar to play

through the game as. There are usually many options available for

customisation, varying from hair colour to hand size, from forehead

slope to eye width, meaning that players can easily create something

that resembles themselves. Therefore, it would be reasonable to expect

that people will do so.

It has been found that users will spend extra effort creating a unique

representation of themselves, despite there being ready-made options

available, and even if it requires significantly more time commitment

(Cheng et al., 2002; Taylor, 2002). In the social online game Second

Life, 60% of users were found to create an avatar based on their own

characteristics, and it was found that it was often an idealised version of

themselves (Ducheneaut et al., 2009). Players of World of Warcraft were

asked to rate their own attributes, and then rate their character’s

attributes. It was found that they rated their characters as having more

favourable attributes than their own attributes; therefore they had

created an idealised character (Bessiere et al., 2007). In a study of

undergraduate females, it was found that those who wished to be

thinner created avatars with a lower body mass (Thomas & Johansen,

2012).

Creating an idealised self avatar has been found to establish a stronger

avatar-self connection than creating a self-reflective avatar (Sa, 2010).

Furthermore, users who create an idealised avatar indicate greater

personal and emotional investment in their avatar. Introverted players

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 4 URL: http://press-start.gla.ac.uk

often create avatars which are idealised versions of themselves, while

extroverted players experiment with new identities (Yee, 2006).

Evidently, the process of avatar creation in video games can reveal

considerable information about self-presentation and social behaviour.

Gender and Gameplay Choices

Games such as The Elder Scrolls Online originate massive virtual worlds,

complete with their own history, politics and prejudices (or lack thereof).

Potentially, one of the reasons why these games are popular with both

genders is because in the game, male and female bodies are equally

strong. Both genders can be warriors or healers if they choose to be;

there are no gender stereotypes. There are many strong female leader

characters, the society is truly equal and not dominated by either

gender.

While MMORPG environments allow players to escape the gender

stereotyping of the real world, video games are still often marketed very

obviously to men or to women (Lien, 2013). For example, the Call of

Duty first person shooter series is usually marketed to men -

commercials for it are shown during sports games and action movies.

Games marketed to women are less action based and more light-

hearted, life simulations or puzzles such as Candy Crush and Farmville.

According to some research, men like action and adventure, or “serious”

games, while females prefer games which involve helping and nurturing

(Subrahmanyam & Greenfield, 2000; Hartmann & Klimmt, 2006).

However, in a more recent study of South African young adults, it was

found that females and males actually liked similar games, and that

males identified with female game characters. The paper argued that

people performed in accordance with gender stereotypes because of

social expectations (Amory & Molomo, 2012). This suggests that there

could potentially be many female gamers who do not openly discuss

their gaming habits because of the stereotypes associated with gaming.

Often, people do not identify as gamers because of the negative

connotations associated with gaming (Shaw, 2011). This research adds

to the argument that gender stereotyping within gaming is a widespread

issue.

Despite this, women now make up almost half (48%) of gamers and

contrary to stereotypes, women aged 18 or over make up more of the

gaming population (36%) than boys aged 18 or younger (17%). Adult

male gamers average 18 years’ experience of play while adult females

average 13 years (Entertainment Software Association, 2014). This

shows that women represent a significant amount of the gaming

population, and are almost equally experienced when it comes to years

of play. It highlights how false the stereotypes and marketing strategies

are.

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 5 URL: http://press-start.gla.ac.uk

Rationale of the current study

Research in this area highlights the potential of avatar-based media for

exploring fundamental psychological mechanisms underlying

perceptions, attitudes and behaviours relating to self-presentation.

(Seung-A, 2010). Avatar creation media provides an environment with

which to study the more unconscious aspects of self-presentation. In

online environments such as these, it becomes easy for people to

interact socially in a disembodied manner; they are granted anonymity

and can reveal as much or as little as they choose. This combination of

anonymity and disembodiment provides a computer mediated

environment wherein a new kind of identity production appears. (Bargh

et al., 2002; McKenna et al., 2002). The study of avatars and social

interaction online are important research areas in the human computer

interaction (HCI) and cyberpsychology fields, and the wider field of

social psychology.

Online massively multiplayer video games such as World of Warcraft

(WoW) have created idiosyncratic social environments with their own

culture, history, group behaviour and real economic markets. It is

unsurprising that they have become a focus of study for researchers in

psychology. WoW has been studied by researchers using methods such

as qualitative ethnography and quantitative analysis of census data

(Bainbridge, 2015). As we continue to spend more and more time

online, the study of social behaviour in online communities becomes

increasingly important.

In addition to this, the recent “Gamergate” controversy (see Heron et

al., 2014) highlights the need for research in to gamer identity,

particularly with respect to gender and gaming. Despite the number of

female gamers, the games industry is still seen as a male dominated

industry (Stuart, 2011; Jayanth, 2014). Women make up just 12% of

the games industry workforce (Wallace & Robbins, 2006; Burgess et al.,

2007). Stereotypes prevail, which subsequently affect the design and

marketing of games. Research into gameplay choices will help to inform

game design, as well as revealing information about social behaviour,

individual differences and gender differences.

The main aim of the study is to investigate self-presentation in avatars

with relation to self-esteem. The secondary aim is to explore avatar-

directed gameplay choices with relation to gender. While a growing body

of research has been dedicated to self-presentation in avatars, none has

investigated the relationship between self-esteem and self-presentation

using Franzoi & Shields’s (1984) body self-esteem measure.

There has been some previous research in gender and avatar creation;

however, it has been limited. For example, Thomas & Johansen’s (2012)

study used a female-only sample and thus was not making a gendered

comparison. Currently, there is little research on gender in gameplay

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 6 URL: http://press-start.gla.ac.uk

choices. It has been found that females did show a slight preference for

healing/support based roles (DiGiuseppe & Nardi, 2007). However, this

study can be criticised for its methods and non-scientific approach. The

study was mostly quantitative and did not use any statistical methods to

analyse the strength of the relationship between gender and class

choice. Building on this research and using assumptions from gender

stereotypes to draw hypotheses, this study looks to investigate gender

differences in gameplay with regard to class choice.

The hypotheses of the study are as follows:

H1: self-esteem will predict scores of perceived similarity

H2: gender will affect class choice

Method

Design

The study used a mixed design to investigate two hypotheses. The

independent variables were self-esteem scores and gender. The

dependent variables were perceived similarity scores and class choice,

respectively.

Ethics

Participants were informed before the experiment commenced that all

their information would be kept fully confidential. They were reminded

that they could withdraw from the study at any time and could omit any

questions they did not wish to answer. Consent forms were used to

ensure that participants were aware of what they were taking part in.

Participants

There were 40 participants. 23 (57%) were female. 28 (70%)

participants were students. The remaining participants were either

graduates or non-students. Participants were recruited through

advertising on the University of Glasgow’s Psychology Facebook groups,

email and through word of mouth. Age of participants ranged between

20-35. The average age was 22.6 (SD = 3.05).

Twenty described themselves as regular gamers. The others had little to

no gaming experience. Of the regular gamers, the average time spent

playing games per sitting was 2.2 hours. The most common system

used was PC, followed by PS3. The most popular genres were

action/adventure, role playing (RPG) and first person shooter (FPS). The

genre of MMORPG games was played by 10% of participants (n = 4).

Materials

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 7 URL: http://press-start.gla.ac.uk

The video game used to create avatars was The Elder Scrolls Online, a

2014 massively multiplayer online role playing game (MMORPG). Players

are able to create a unique avatar upon starting the game, which will

represent them in the social online world as they play through the

game.

The avatar creation process is very detailed and allows players the

choice of nine races. There are three human races, the Redguards,

Nords and Bretons, differentiated by hair, eye and skin colouring,

culture and combat style. The three elven races are Dunmer, Bosmer

and Altmer. Lastly, there are three humanoid creature races, the Orcs,

Khajits and Argonians. For this study, race choice was left open to

participants. A brief outline of each race is provided in-game to inform

users of their traits and combat styles.

Figure 1: Available races in The Elder Scrolls Online. Top row L-R:

Argonian, Breton, High Elf; Middle row L-R: Khajit, Nord, Dark Elf;

Bottom row L-R: Orc, Redguard, Wood Elf

After choosing a race, users then choose a class, which determines their

style of combat and skills. The game provides a description of each

class, and if clarification was needed extra information was given to

participants to inform their decision. Choice of class is important as it

determines the player’s role in a social group. Dragonknights are

damage based warriors using heavy weapons, Nightblades focus on

damage using stealth, while Sorcerers use magic and Templars are

support based using magic to heal teammates.

Participants were told to choose the class as if they were going to play

the game, and reminded of the social and team work aspects involved.

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 8 URL: http://press-start.gla.ac.uk

Figure 2: Class choices in The Elder Scrolls Online

For the physical dimension of the avatar, the game allowed for the

manipulation of 48 attributes, using scale sliders and ternary diagrams.

These attributes varied from face structure such as cheekbone height,

eye width etc., to body shape such as waist size, hip size and gut size. It

was left to participants to manipulate as many or as little attributes as

they liked, to ensure that the process represented real life character

creation as closely as possible.

Figure 3: Example of avatar creation user interface in The Elder Scrolls

Online

Measures

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 9 URL: http://press-start.gla.ac.uk

The measures used were Rosenberg’s Self-Esteem scale, a Body Self-

Esteem questionnaire and a specially created avatar evaluation

questionnaire.

Self-esteem

Firstly, to measure general self-esteem, the Rosenberg scale (1965) was

used. This self-report test measures ten items on a Likert-type scale

with responses ranging from ‘strongly agree’ to ‘strongly disagree’.

Scores range from 0-30 with scores under 15 signifying low self-esteem.

Secondly, to measure body self-esteem, Franzoi & Shields (1984) Body

Self-Esteem scale was used. This self-report questionnaire uses a Likert

scale to measure how participants feel about 35 body parts and

functions, such as “appearance of stomach”, “thighs”, “nose”, “lips” etc.

Participants respond with a number between 1 and 5 in the space

provided. The minimum score is 35 and the maximum is 175.

Evaluation

The avatar evaluation asked participants to rate their avatar in terms of

facial features and body, and then asked them how similar to

themselves they perceived the avatar to be. Lastly, participants were

asked to write a short sentence explaining why they chose the race and

class they did.

Additional Measures

A questionnaire on gaming experience and habits was given to

determine if participants were regular gamers or non-gamers. This

asked if they considered themselves regular gamers or not, and if so

how many hours they would typically spend playing. It also asked which

genres of games they played and on what gaming system. Lastly, it

asked if participants played online games and if so, if they played with

friends or with strangers.

Procedure

Firstly, participants were given their questionnaire package and

informed of what they would be doing. They were asked to read and

sign a consent form and read the instructions sheet. Secondly, they

completed the gaming habits questionnaire.

After this, they began creating their avatar. They were asked to

complete a trial to acclimatise to the user interface of the game, by way

of replicating a practice avatar from a printed image. To ensure that this

example avatar did not influence participants’ own avatar choices, it was

given extreme and unusual features that they would be unlikely to

choose for their own avatar.

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 10 URL: http://press-start.gla.ac.uk

The trial ensured that the user interface was not a confounding variable,

as difficulty or confusion using it could affect a participant’s avatar

design process and lengthen their creation time, which was being

measured. If participants were regular gamers or felt confident about

using the interface they were allowed to skip this stage.

After the trial, the avatar settings were reset and the participants began

creating their own avatar. While doing this, a screen recording was

taken in order to measure the time spent on creating.

Figure 4: The practice avatar

Upon finishing the avatar, participants then began completing their

questionnaires. Firstly, they completed the Rosenberg Self-Esteem

scale, followed by the Body Self-Esteem scale, and lastly the avatar

evaluation questionnaire.

Participants were then debriefed on the study. They were told what the

questionnaires had measured, how the data was going to be used and

what the study was investigating. They were reminded that their data

would be anonymised, and that they could email to ask any questions or

enquire about the results of the study. The duration of the experiment

was on average 30 minutes for each participant.

Results

Self-esteem and similarity

H1: self-esteem will predict scores of perceived similarity

A multiple regression was run to test if self-esteem would predict

perceived similarity ratings. Self-esteem was a composite of two

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 11 URL: http://press-start.gla.ac.uk

independent variables: the Rosenberg general self-esteem measure and

the body-self-esteem measure. A multiple regression was used as it

allows analysis of the effects of multiple independent variables on a

dependent variable; in this case if one or both self-esteem measures

predicted similarity.

The results were found to be highly significant: F(2, 37) = 5.73, p =

0.007. This supports the hypothesis that self-esteem will predict scores

of perceived similarity. However, analysis of coefficients revealed that

only one of the variables added statistically significantly to the

prediction, the body self-esteem measure (p = 0.002). This signifies

that while body self-esteem was a highly significant predictor of

perceived similarity, there was no relationship found between general

self-esteem and perceived similarity rating.

Table 1: Descriptive Means

The unstandardised coefficient value for body-self-esteem was 0.081.

This means that as body self-esteem rating increased, perceived

similarity rating increased by 0.081. This supports the hypothesis that

self-esteem will predict scores of perceived similarity.

Gender and class choice

H2: gender will affect class choice

For this hypothesis a Chi-Square goodness-of-fit test was used. The

results were moderately significant: x(3) = 9.264, p <0.026. Thus the

hypothesis that gender would affect class choice was supported.

It was found that the most popular class choice for women was the

healing role (Templar) as 34.8% (n = 8) of female participants chose

this, while only 1 male participant chose it. For males the most popular

choice was the damage focused role (Dragonknight) with 47.7% (n = 8)

making this choice. The Sorcerer class was more popular with females

(30.4%, n = 7) than males (11.8%, n = 2) while the Nightblade class

was more popular with males (65.3%, n = 6) than females (17.4%, n =

4). Examination of Cramer’s V reveals that the strength of the

relationship between gender and class choice is very high, as the

Gamma value is high (0.481) while the significance is low (0.026). This

supports the hypothesis that gender affects class choice.

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 12 URL: http://press-start.gla.ac.uk

The bar chart below illustrates the differences in class choice between

genders. It is clear that Templar was a significantly more popular choice

for females than males, while Dragonknight was more popular with

males.

Figure 5: Class choice for males and females

However, a second cross-tabulation revealed that within gamers of both

genders, Dragonknight was most popular (22.5%, n = 9) and Templar

least popular (5%, n = 2) Being a gamer did not significantly predict

class choice (p = 0.066) as class was varied within the regular gamer

group.

Discussion

Summary

Primarily, the study sought to investigate self-presentation in video

games. Previous research found that self-presentation is a key feature

of online social interaction, and that self-esteem can predict the nature

of the self portrayed. This study proposed that self-esteem would predict

the perceived similarity rating of participants’ avatars. Gender and class

choice was examined to determine if gender plays a role in gameplay

decisions, to build on previous but limited research in this area.

Interpretation of the results

Self-Esteem & Similarity

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 13 URL: http://press-start.gla.ac.uk

While there were no significant results found for general self-esteem,

there were highly significant results found for the relationship between

body self-esteem and similarity rating. The strength of the relationship

was examined and found to be very strong. As body self-esteem score

increased, perceived similarity rating increased. This supported the

hypothesis that self-esteem would predict avatar similarity.

As expected, those with a high body self-esteem score created an avatar

similar to themselves, potentially because they were happy with

themselves and did not feel the need to change things dramatically.

Conversely, those with low body self-esteem score created something

they perceived to be slightly similar, but downplayed or exaggerated the

features they didn’t like in themselves - these features were determined

through the results of the body self-esteem questionnaire. This supports

Jensen et al.’s (2003) statement “Whether this takes the form of a kind

of literal ‘mapping’ of the self onto one’s avatar, or a concealment of

one’s “real self” within it, avatars are viewed as a form of self-

representation.”

These results are consistent with findings from previous research (e.g.

Axelsson, 2002; Ducheneaut et al., 2009; Bessiere et al., 2007; Thomas

& Johansen, 2012; Companion & Sambook, 2008; Trepte & Reinecke,

2010) and suggest that body self-esteem plays a fundamental role in

online self-presentation. Despite the study using a small sample, the

effects of body self-esteem on online self-presentation were

pronounced.

However, an alternative interpretation of the results could be that those

who have high body self-esteem are more likely to describe the avatars

as similar to themselves, because they believe themselves to be closer

to the idealised version they have created, even if this is not the case.

Similarly, those with low body self-esteem may state that their idealised

avatar does not look like them, but an objective observer may think it

does, when the person cannot see it for themselves because their

judgment is clouded by low self-esteem. Therefore, to fully investigate

the relationship between body self-esteem and avatar similarity, a

future replication could use an additional measure of similarity, such as

the judgment of a neutral party or even a comparison of actual body

measurements and facial structure vs. the avatar’s features.

Gender

It was hypothesised that gender would affect class choice. This

hypothesis was supported by the results of a Chi-Square test. Males

were more likely to choose the Dragonknight class, while females were

more likely to choose the Templar class. This suggests that there are

gameplay differences between males and females with regard to the role

they play in a social online game. These results posit that women prefer

healing, supportive roles while men prefer damage-based, more

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 14 URL: http://press-start.gla.ac.uk

aggressive roles. This is consistent with typical gender stereotypes and

marketing in video games. However, it is possible that females are

simply acting in according with the stereotypes expected of them, as

argued by Amory & Molomo (2012). It was found that female gamers

showed more variation in class choice, suggesting that class choice

could potentially be more to do with gaming experience. This supports

DiGiuseppe & Nardi’s (2006) findings that as gaming experience

increased, females showed more variation in class choice. If females do

not have gaming experience, perhaps they are more likely to fall into

stereotypical behaviour. Of course the same could apply to males too, as

they may choose the more aggressive role to fulfil the stereotype of

being physically strong. However, as afore-mentioned the sample was

small, thus a larger study could determine if these effects are

widespread.

Limitations

The study was not without its limitations. Predominantly, the sample

size was very small, meaning it is difficult to generalise the results to

the population. This was due to time constraints; in future the study

could be replicated with a larger sample to increase statistical power.

Secondly, age in the sample only varied between 20 and 35. This was

because it was mainly undergraduates used, and it was difficult to gain

access to individuals spanning each age group. It was decided that the

sample would be limited to this age range as it would have been difficult

to counterbalance across all age groups. Future research could improve

on this by testing individuals in all age groups to determine if age

produces any effect. In terms of gender, there was a disproportionate

amount of females in the sample. Ideally there would be an equal

amount of both genders to ensure the results are generalisable to the

population.

Another issue with the study was software problems. The Elder Scrolls

was chosen as it had a very detailed character creation interface;

however some participants still commented that the attributes changed

very subtly and because of this they could not make extreme “cartoony”

looking characters. An additional problem of using this software was the

user interface, as it is perhaps not the most intuitive to use. There are

sliders for each attribute which works well when the attribute is on a

scale (e.g. height); however for attributes like hair style, the user is

required to click through each option as there is no thumbnail to display

each style. This frustrated some participants and could have lessened

the effort they expended in creating their avatar.

Implications

The study presents some interesting implications with regard to self-

presentation in video games. It has been found that participants who

played a game with their own personalised avatar displayed higher

levels of aggressive behaviour than those who played with a non-

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 15 URL: http://press-start.gla.ac.uk

personalised avatar. In violent video games, participants were found to

experience more arousal and self-activation if they played with a

personalised character, which subsequently increased aggressive

responses (Fischer et al., 2010). It would be interesting to investigate

the effect of personalised avatars on behaviour in other types of games

such as prosocial games, to determine if this is a widespread effect.

Additionally, avatars in video games could allow for an increase in self-

esteem. These environments also allow people to express themselves

fully, in ways that they may not feel able to in real life - they are able to

reveal their “hidden selves” (Sular, 2002). It is for these reasons that

online environments play a key role in identity empowerment. (Zhao et

al., 2008).

Future research

A fundamental direction for future research would be to isolate the

specific features of the body self-esteem questionnaire, and examine

each one in relation to the corresponding feature on the avatar. For

example, comparing ratings of own biceps in males to size of biceps

they choose for their avatar. This would serve to explore in more depth

the relationship between self-esteem and avatar creation.

Secondly, the difference between the effects of general self-esteem and

body self-esteem should be investigated further. The study found that

there was no effect of general self-esteem on the avatar’s perceived

similarity. This suggests that either avatar presentation solely reflects

physical attributes, or that the sample size was simply not large enough

to find any effects. Perhaps general self-esteem could be reflected in

other non-physical attributes commonly found in avatars such as

intelligence, honesty, wisdom, etc. It could be possible that people

separate their physical and mental qualities when creating an avatar,

and as there were only physical qualities to customise, general self-

esteem was not a significant factor in influencing this.

As previously mentioned, a limitation of this study was the age range in

the sample. Future research in the area could aim to test participants

from every age group to determine if the effects are age-specific. Would

older adults display the same amount of self-presentation efforts as

younger adults? Does exposure to video games and social media affect

this, or is self-presentation a timeless concept found across many

mediums? A future study could compare self-presentation in video game

avatars, social media websites and traditional mediums to determine if

self-presentation efforts have increased with the use of these

technologies.

Lastly, future research could comprise cross cultural study. Research has

found that there are differences in the way Western and Asian users

self-present in games. Western games more commonly have

customisation options while Asian games will typically only have pre-

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 16 URL: http://press-start.gla.ac.uk

defined characters to play as. It was found that Asian gamers disliked

detailed character creation because they did not like that other, more

skilled players could create avatars more attractive than theirs (Yee,

2006). This could possibly represent a difference in cultures, as Western

cultures focus on individualism and Eastern on collectivism. Future

research could investigate how self-presentation differs across cultures,

to establish if the research by Yee is generalisable to the population.

Conclusion

This study contributed to the literature by expanding on current

research while investigating new hypotheses. The effect of self-esteem

on avatar self-presentation found in previous research was confirmed

and found to have a strong relationship. The study built on previous

research by using quantitative rather than qualitative methods to find

that gender affected gameplay decisions.

The results of this study could be applied to many areas, such as games

design and marketing, cyberpsychology and the study of online social

behaviour. It is important to determine if real life social behaviours

translate to online environments, or if different rules apply in these

environments.

As previously discussed, there is an argument that the games industry

seems to be discordant with the reality of their audience: marketing for

mainstream titles is often clearly directed at men or women. As such,

gender is a contentious topic within the industry, and therefore

something that needs to be explored. Research in this area could lead to

better game design and marketing as designers will learn more in depth

information about their audience, while informing the field of social and

cyber psychology on trends in online behaviour.

References

Alexander, C., Knight, G. (1971). Situated Identities and Social

Psychological Experimentation, Sociometry, 34(1), p65-82

Amory, A., & Molomo, B. (2012). Gendered Play and Evaluation of

Computer Video Games by Young South Africans, Gender, Technology

And Development, 16(2), p177-196

Arkin, R. (2013). Self presentation styles. In J. Tedeschi, Impression

Management Theory and Social Psychology Research (2nd ed.).

Axelsson, A. (2002). “The Digital Divide: Status Differences in Virtual

Environments.” In R. Schroeder (Ed.) Social life of Avatars: Presence

and interaction in Shared Virtual Environments (pp.188-204). New York:

Springer.

Bainbridge, W. (2015) World of Warcraft, The International Encyclopedia

of Digital Communication and Society, p1–4.

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 17 URL: http://press-start.gla.ac.uk

Bargh, J., McKenna, K., & Fitzsimons, G. (2002). Can You See the Real

Me? Activation and Expression of the “True Self” on the Internet, Journal

Of Social Issues, 58(1), p33-48

Bessière, K., Seay, A. F., & Kiesler, S. (2007). The ideal elf: Identity

exploration in World of Warcraft, CyberPsychology & Behavior, 10,

p530-535.

Bibby, P. (2008). Dispositional factors in the use of social networking

sites: Findings and implications for social computing research, Lecture

Notes in Computing Science, 5075, p392-400

Brenner, J. (2013). Pew Internet: Social Networking. Retrieved 5 June,

2015 from http://www.pewinternet.org/fact-sheets/social-networking-

fact-sheet/

Bromley, D. (1993). Reputation, image, and impression management.

Chichester, England: Wiley.

Burgess, M. C. R., Stermer, S. P. & Burgess, S. R. (2007). Sex, lies and

video games: The portrayal of male and female characters on video

game covers. Sex Roles, 57, p419-433.

Caspi, A., & Gorsky, P. (2006). Online deception: Prevalence, motivation

and emotion, CyberPsychology & Behavior, 9, p54-59.

Cheng, L., Farnham, S., & Stone, L. (2002). Lessons learned: Building

and deploying shared virtual environments. In R. Schroeder (Ed.), Social

life of avatars: Presence and interaction in shared virtual environments.

New York: Springer.

Companion, M., & Sambrook, R. (2008). The Influence of Sex on

Character Attribute Preferences. Cyberpsychology & Behavior, 11(6),

p673-674

Cornwell, B., & Lundgren, D. C. (2001). Love on the Internet:

Involvement and misrepresentation in romantic relationships in

cyberspace vs. realspace, Computers in Human Behavior, 17, p197-211.

DiGiuseppe, N., Nardi, B. (2007) Real Genders Choose Fantasy

Characters: Class Choice in World of Warcraft, First Monday, 12(5)

Ducheneaut, N., Wen, M., Yee, N., & Wadley, G. (2009). Body and

mind: A study of avatar personalization in three virtual worlds.

Proceedings of CHI 2009.

Ellison, N., Steinfield, C., Lampe, C. (2006) The Benefits of Facebook

“Friends”: Social Capital and College Students, Journal of Computer-

Mediated Communication, 12(4), p1143-1168

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 18 URL: http://press-start.gla.ac.uk

Entertainment Software Association. (2014). Essential Facts About the

Computer and Video Games Industry. Retrieved 5 June 2015, from

http://www.theesa.com/wp-content/uploads/2014/10/ESA_EF_2014.pdf

Fischer, P., Kastenmuller, A., Greitemeyer, T. (2010). Media violence

and the self: The impact of personalised gaming characters in

aggressive video games on aggressive behaviour, Journal of

Experimental Social Psychology, 46(1), p192-195

Franzoi, S.L. & Shields, S.A. (1984). The Body-Esteem Scale:

Multidimensional structure and sex differences in a college population,

Journal of Personality Assessment, 48, p173-178.

Goffman, Erving (1959). The Presentation of Self in Everyday Life. New

York: Doubleday.

Hartmann, T., KIimmt, C. (2006) Gender and computer games:

Exploring females’ dislikes, Journal of Computer-Mediated

Communication, 11, p910-931

Heron, M., Belford, P., & Goker, A. (2014). Sexism in the circuitry.

SIGCAS Comput. Soc., 44(4) p18-29

Jayanth, M. (2014). 52% of gamers are women – but the industry

doesn’t know it | Meg Jayanth. the Guardian. Retrieved 9 March 2015,

from http://www.theguardian.com/commentisfree/2014/sep/18/52-

percent-people-playing-games-women-industry-doesnt-know

Jellison, J., Gentry, K. (1978). A Self-Presentation Interpretation of the

Seeking of Social Approval, Personality and Social Psychology Bulletin,

4(2), p227-230

Jones, E.E. (1964). Ingratiation: A social psychological analysis. New

York, New York: Appleton-Century-Croft.

Kramer, N., Winter, S. (2008) Impression Management 2.0, Journal of

Media Psychology, 20(3), p106-116

Leary, M. (1995). Self-presentation: Impression management and

interpersonal behavior. Social psychology series. Madison, WI, US:

Brown & Benchmark Publishers

Leary, M., & Kowalski, R. (1990). Impression management: A literature

review and two-component model. Psychological Bulletin, 107(1), p34-

47

Lien, T. (2013). No girls allowed. Polygon. Retrieved 19 May 2015, from

http://www.polygon.com/features/2013/12/2/5143856/no-girls-allowed

Manago, A. M., Graham, M. B., Greenfield, P. M., & Salimkhan, G.

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 19 URL: http://press-start.gla.ac.uk

(2008). Self-presentation and gender on MySpace, Journal of Applied

Developmental Psychology, 29, p446–458.

McKenna, K., Green, A., & Gleason, M. (2002). Relationship Formation

on the Internet: What's the Big Attraction?. Journal Of Social Issues,

58(1), p9-31

Nadkarni, A., & Hofmann, S. (2012). Why do people use Facebook?

Personality And Individual Differences, 52(3), p243-249

Rosenberg, M. (1965). Society and the adolescent self-image. Princeton,

NJ: Princeton University Press.

Sa, J. (2010) “I feel more connected to the physically ideal mini me than

the mirror-image mini me”: theoretical implications of the “malleable

self” for speculations on the effects of avatar creation on avatar-self

connection in Wii, Cyberpsychology, Behaviour and Social Networking,

13(5), p567-70

Schau, H., & Gilly, M. (2003). We Are What We Post? Self‐Presentation

in Personal Web Space. Journal Of Consumer Research, 30(3), p385-404

Schlenker, B. (1980). Impression Management: The Self-Concept, Social

Identity, and Interpersonal Relations. Monterey, California: Brooks/Cole.

Shaw, A. (2011). Do you identify as a gamer? Gender, race, sexuality,

and gamer identity. New Media & Society, 14(1), p28-44

Strano, M. M. (2008). User Descriptions and Interpretations of Self-

Presentation through Facebook Profile Images. Cyberpsychology:

Journal of Psychosocial Research on Cyberspace, 2(2), article 1.

Retrieved 5 June 2015, from

http://cyberpsychology.eu/view.php?cisloclanku=2008110402

Stuart, K. (2011). Game changers: the women who make video games.

The Guardian. Retrieved 9 March 2015, from

http://www.theguardian.com/technology/2011/dec/08/women-

videogames-designing-writing

Subrahmanyam, K., Greenfield, P. (2000). Computer games for girls:

What makes them play? J. Cassell, H. Jenkins (Eds.), From Barbie to

Mortal Kombat, The MIT Press, Cambridge, Mass, pp. 46-71

Surratt, C. (1998) Netlife: Internet citizens and their communities, Nova

Science, New York

Taylor, T. L. (2002). Living digitally: Embodiment in virtual worlds. In R.

Schroeder (Ed.), The social life of avatars: Presence and interaction in

shared virtual environments (pp. 40–62). London: Springer-Verlag

Maxwell Pringle Conjuring the Ideal Self

Press Start 2015 | Volume 2 | Issue 1 ISSN: 2055-8198 20 URL: http://press-start.gla.ac.uk

Thomas, A. G., & Johansen, M. K. (2012). Inside out: Avatars as an

indirect measure of ideal body self-presentation in females.

Cyberpsychology: Journal of Psychosocial Research on Cyberspace, 6(3)

Toma, C., & Carlson, C. (2015). How Do Facebook Users Believe They

Come Across in Their Profiles?: A Meta-Perception Approach to

Investigating Facebook Self-Presentation. Communication Research

Reports, 32(1), p93-101.

Trepte, S., & Reinecke, L. (2010). Avatar Creation and Video Game

Enjoyment. Journal Of Media Psychology: Theories, Methods, And

Applications, 22(4), p171-184

Turkle, S. (1995) Life on the screen: Identity in the age of the Internet,

Simon & Schuster, New York

Wallace, M., & Robbins, B. (Eds.) (2006). IGDA 2006 casual games

white paper [online] Retrieved 5 June 2015, from

https://c.ymcdn.com/sites/www.igda.org/resource/collection/BCB11E9B

-13E6-40D0-B390-

952B5E11D35A/IGDA_CasualGames_Whitepaper_2006.pdf

Walther, J. B. (1996). Computer-mediated communication: Impersonal,

interpersonal, and hyperpersonal interaction. Communication Research,

23, p3–43.

Walther, J. B. (2007). Selective self-presentation in computer-mediated

communication: Hyperpersonal dimensions of technology, language, and

cognition. Computers in Human Behavior, 23, pp.2538–2557

Whitty, M. T. (2002). Liar, liar! An examination of how open, supportive

and honest people are in chat rooms. Computers in Human Behaviour,

18, p343-352.

Yee, N. (2006). Avatar and Identity in MMORPGs. Nickyee.com.

Retrieved 24 February 2015, from

http://www.nickyee.com/daedalus/gateway_identity.html