Big Five And Social Networking 2008

Big Five and Social Networking Running head: BIG FIVE AND SOCIAL NETWORKING

Big Five and Social Networking: Misconceptions of an Individual's Personality as Displayed via MySpace Kresenda Keith, Wilhemina Foreman, and Ray Rush University of Alaska Anchorage

1

Big Five and Social Networking Abstract On-line social networking websites can be utilized to predict personality; however, the accuracy of predicting personality can be misguided due to deception, self-knowledge, the veil of anonymity, and even the vary degrees of personality itself. Our previous research on personal websites stated that personality can be predicted accurately, given enough information. We, however, predicted that personality would be mispredicted by respondents due to the ability to customize the social networking site, MySpace. Eight MySpace profiles were viewed by 196 respondents and tested on a 6-point Likert scale for each dimensions of the Big Five. Out of the forty possible outcomes, only 5 were predicted properly implying that personality could be misleading when given a highly customizable website.

2

Big Five and Social Networking

3

Big Five and Social Networking: Misconceptions of an Individual's Personality as Displayed via MySpace Personality theories attempt to account for individual behavior and thus the scope of such theories is vast. They describe how genetic predisposition's and biological mechanisms combine with experience as children develop into young adults who will show behavioral consistencies over their life span. Personality researchers report heritability coefficients, relate MRI scans and EEG activity to intellectual performance and emotional reactions, and predict job outcomes and lifetime satisfaction. They examine the dimensions of self description and the many ways feelings, knowledge, and beliefs combine in behavior. Personality research ranges from tests of evolutionary theories of jealousy to analyses of the structure and content of one's personality traits or even one's different forms of self conception. One of the most comprehensive, empirical, and data-driven models is the Five Factor Model (FFM) or as it is more colloquially known, the Big Five. L. L Thurstone first mentioned the model in 1933 in his presidential address to the American Psychological Association. The Big Five is a lexical analysis of personality theories which lead to the summation of five overall traits of which each person can vary on: Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. The Big Five not only explains a person's personality into simple, categorical terms; it also allows for one to explore possible reasonings and formulate theories leading towards self knowledge and a possible self awareness. A working self-concept and the full self-knowledge means that different and even inconsistent beliefs about the self can coexist. In his self-discrepancy theory, Higgins (1987) asserts that the self-concept contains two more self-representations or self-schemata besides the

Big Five and Social Networking

4

actual self: the ideal self and the ought self. While the actual self refers to the currently activated and salient aspects of self; the ideal self represents the hopes, wishes, and goals of a person from the perspective of how this person would ideally see him- or herself. In contrast, the ought self represents all the responsibilities, tasks, and norms a person feels obligated to fulfill. The side most commonly presented in person, through conversation, and even via social networking sites would be the ideal self as people tend to put what they see as their ―best foot‖ forward. The evolution of psychological theories tends to fluctuate between optimistic advances and self critical analyses/reduction of efforts. Currently, in a new wave of information and personality opportunities, psychological theories have advanced into the age of technology and personality tests have become as simple as an Internet URL or on-line survey. Personality researchers have recently discovered that collecting and scoring personality responses via the Web is vastly more efficient than doing these things by hand, and that Web-based research allows one to tap into much larger and more diverse participant pools via self-report data (Buchanan, Johnson, & Goldberg, 2005; Gosling, Vazire, Srivastava, & John, 2004). When looking at people who use the Internet and why they are drawn to any site, personality gives an insight into motive. Research has only begun to tap into personality traits, utilizing the Big Five, and how they motivate Internet users: ―Agreeableness represents the tendency to be sympathetic, good natured, cooperative, and forgiving…Conscientiousness represents the tendency to be self disciplined, strong willed, deliberate, and reliable…Extraversion represents sociability, cheerfulness and optimism…Neuroticism represents a lack of psychological adjustment and emotional stability…Openness represents curiosity and willingness to explore new ideas‖ (McElroy, Hendrickson, Townsend, & DeMarie,

Big Five and Social Networking

5

2007). Landers and Lounsbury (2006) found that Conscientious people are more likely to spend time on-line engaged in academic pursuits and people high in Extraversion use the Internet as a tool to find information that they can share with others because they prefer face to face interaction. Personality traits are well described; however, how those traits are conveyed by an individual and perceived by others via the Internet. One may have ideas of what these individuals do on the Internet, yet we still do not have a clear idea of how the traits are perceived when information is presented on a social networking site. Using a web-based social networking sites is a simple way to meet new people without leaving the comfort of home. Being able to express one's personality by designing web pages or profiles that portray how the individual sees themselves is one way that an individual can establish new friendships via the Internet. Past research utilizing the social networking site Facebook suggests ―that the on-line networking websites are, in fact, a valid means of communicating personality‖ (Gosling, Gaddis, & Vazire). Facebook, however, is fairly constrictive as far as what can be put on the profile: a person must use their actual name, cannot change the background of the page, etc. On the social networking site MySpace, a user can conveyed his or her personality through backgrounds, pictures, icons, music, etc. A deeper look into his or her personality can either mean a deeper conveyance of the true self or can open the door to deception. It is unsurprising that deception is found on the Internet. Crawford (2006) states that ―people who are lying to another person in a chat room or in instant messaging use approximately one-third more words, probably in their attempt to construct a more cohesive and detailed story in order to seem believable‖ and often those being lied to may not be aware of a

Big Five and Social Networking

6

deception but will nonetheless change their behavior in reaction to lies. This means that people are implicitly aware that they are being lied to without being outwardly aware of the lies themselves. People seem to unconsciously know that they're being lied to on-line because deceptive websites will often ask more questions and have far more words than are usually required for an otherwise truthful statement. Wallace (as cited in Amichai-Hamburger, 2004) states that while sometimes people will express their ideal selves, and write about themselves as to what they want to be – for example, ―saying the right things‖ to gain social approval (and thus explicitly lie). This is called the ―ideal self.‖ Implicit (unconscious) lying may arise not because the person wishes to be willfully deceitful but merely because these individuals wish to express their ―ideal self,‖ or the person they would like to see themselves as. This phenomena is important because it can lead to incorrect results when a person receives the results of their Big-5 personality traits, and then has observers view their site and come up with drastically different interpretations and conclusions as to what the person being observed is in reality. This is important in regards to implicit and explicit lying because, according to Amichai-Hamburger (2004), there is an atmosphere on the Internet which induces a unique and protective environment allowing a fostering of freer expression than one would have in normal interactions in the offline world. It is important to note that with this freer expression and more open/anonymous atmosphere comes the territory to lie and have the feeling that the lying can be gotten away with. It is obvious that the Internet can be used as a vessel of understanding the expression of one's personality, however the accuracy of these interpretations can be argued. Machliek, Schutz, and Marcus (2006) researched visitor ratings and website content from the multiple websites'

Big Five and Social Networking

7

owners and found was that the reliability of websites as a personality indicator were only as useful as the information that they revealed. The validity of this information also relies heavily on the richness of the information provided. It can therefore be understood that where sometimes people are being implicitly or explicitly deceptive on their websites, other times they simply are not providing enough information to make accurate inferential judgments on one's personality. Deception, self-knowledge, the veil of anonymity, and even the vary degrees of personality can make the Internet a tricky avenue for predicting personality. We questioned if that communication of personality is even accurate when given a customizable social networking site,such as MySpace. When presented with the MySpace profile of an actual person, we believe that a respondent will have an inaccurate perception of a person's Five Factor personality. Method Participants Eight MySpace webpage owners were selected for this study. These MySpace participants were selected from acquaintances—four men, four women—based on the perceived variation in their personalities. There were 196 respondents. Minimal restrictions were placed: respondents needed to only be over the age of 18. Respondents were recruited through Internet requests and on-line advertisements via the UAA Psychology Department website (http://www.uaa.alaska.edu/psych/Research/participant.cfm), as well as through fliers posted around the University of Alaska Anchorage campus, incited with possible extra credit towards an undergraduate Psychology class (with instructor's permission). Acquaintances were also asked to participate.

Big Five and Social Networking

8

Materials The participants were directed to an IPIP test at the Personality Project (http://personalityproject.org/), a nationally run survey by Professor William Revell at Northwestern University where the participants took a 74-item questionnaire. The IPIP was chosen because of the extensive literature showing that the scales in the IPIP implementation correlate highly with corresponding NEO-PI-R Domain Scores, with correlations that range from 0.85 to 0.92 when corrected for unreliability (International Personality Item Pool, 2001). The IPIP representation is also freely available in the public domain and relatively short ensuring that maturation is unlikely to occur. The respondents were asked to use SurveyMonkey to evaluate the participants. SurveyMonkey is a website for creating and executing on-line questionnaires. Results were analyzed utilizing a statistical program, SPSS, and the Microsoft program, Excel. Procedure The 8 MySpace participants were given mean scores for each of the five factors as well as a percentile score at the end of the Big Five survey. The percentile score reflecting their standing in the population of the participants. Once the self-reported scores were attained, respondents were recruited to visit 8 provided MySpace websites and rate the owner's personality via a 6-point Likert scale for each trait (Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism). Each dimension was defined on each SurveyMonkey page by an operational definition. The

Big Five and Social Networking

9

respondent had full access to a each individual's pictures, music, friends lists, comments, etc. Results To establish whether the true five-factor scores of each of the eight MySpace participants were accurately estimated by the respondents, confidence intervals for the mean predicted scores for each individual's trait were calculated. The 95% confidence intervals and descriptive statistics are presented in Table 1. Of the 40 items, 35 of the true means were not statistically predicted by the 95% confidence intervals. The robustness of the results was heightened by half of the MySpace pages being completely mispredicted on all traits. Of the 5 possible traits, Agreeableness and Neuroticism were both completely mispredicted for all participants. Figure 1 displays the inaccuracies within predicting the trait of Neuroticism as an example. Interestingly enough, each trait for each individual showed a range of 1 to 6, displaying a complete lack of internal reliability within each trait. For example, all of the five-factor traits for the participant Christie were mispredicted (Fig. 2). Most of the traits that she tested high in were predicted low and most of the traits she tested low were predicted high showing that respondents likely believed Christie to be the statistical opposite of who she reported herself as. A two-way analysis of variance (ANOVA) was conducted to evaluate the accuracy of predicting a trait (measured by the deviation scores) based upon each trait and the respondent’s gender. Of the

40 traits, 6 showed significant gender differences; five of the 6 were shown to be better predicted by females (Table 2). Interestingly, Neuroticism and Conscientiousness both displayed no gender differences.

Big Five and Social Networking

10

Discussion The results of this study support the hypothesis and imply that MySpace-based impressions of personality differ from self-reported Big Five personality assessments. Previous research by Gosling et al. had shown that Neuroticism was the least accurately predicted of the Big Five traits; however, a complete lack of consensus regarding impressions of Neuroticism and Agreeableness was displayed. Neuroticism possibly was completely mispredicted because people can easily hide their flaws on-line and display their ―ideal self.‖ The incorrect impression on Agreeableness is unknown and deems further research. Our study does not corroborate with past research which implied that personality was predictable via a social networking site (Gosling, Gaddis, & Vazire, 2007). We believe that these differential results are largely due to the fact that MySpace is far more customizable and allows people to customize their profiles, unlike other social networking sites such as Facebook. These findings could imply that a higher deception from owners/participants is prevalent on more customizable social networking sites or that more variance in interpretation by the respondents is due to the myriad of possible interpretations of vast information found on a MySpace profile. Either way, this data suggests that on-line networking sites are not as relevant or valid as previously thought at communicating personality. Limitations to this study include the lack of demographics on some respondents, although it is expected that the sample is education and age skewed. Demographics of all respondents are not known likely due to the format of the survey, having put the demographics section last; this is likely due to maturation. Another limitation would be that the Big Five is based on self report and can only be as accurate as the information given. The test used, however, has been proven to

Big Five and Social Networking

11

be reliable and utilizes countermeasures and adjusts for response sets; however, self report data can also reflect a person's ideal vision of the self. More research is needed to investigate personality and how it is perceived on the Internet as well as the repercussions of companies using web based networking sites as a profiling tool to determine employment, coverage, benefits, etc. Further research should also explore the lack of respondent agreement on traits such as Neuroticism. Findings from this study deem for future research furthering the use of MySpace and any information found on the Internet as an accurate perception of a person's personality.

Big Five and Social Networking References Amichai-Hamburger, Y. (2004). Personality and the Internet. Retrieved November 10, 2008, from www.geocities.com/loversrock_s/Hamb-Ch02.pdf. Buchanan, T., Johnson, J. A., & Goldberg, L. R. (2005). Implementing a Wve-factor personality inventory for use on the Internet. European Journal of Psychological Assessment, 21, 116–128. Crawford, F. (2006, November 15). Getting at the many tangled webs of digital deception we seem hardwired to weave. Cornell University Chronicle Online. Retrieved December 04, 2008, from http://www.news.cornell.edu/stories/Nov06/SS.Hancock.html. Gosling, S. D., Gaddis, S., & Vazire, S. (2007). Personality impressions based on Facebook profiles. Symposium conducted at the meeting of ICWSM, Boulder, CO. Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we trust webbased studies? A comparative analysis of six preconceptions about Internet questionnaires. American Psychologist, 59, 93–104. Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319-340. International Personality Item Pool. (2001). A scientific collaboratory for the development of advanced measures of personality traits and other individual differences. Retrieved October 17, 2008, from http://ipip.ori.org/.

12

Big Five and Social Networking Landers, R. N., & Lounsbury, J. W. (2006). An investigation of Big Five and narrow personality traits in relation to Internet usage. Computers in Human Behavior, 22(2), 283-293. Machliek, F.., Schutz, A., & Marcus, M. (2006) Personality in cyberspace: Personal websites as a media for personality expressions and impressions. Journal of Personality and Social Psychology, 50(6), 1014-1031. McElroy, J. C., Hendrickson, A. R., Townsend, A. M., & DeMarie, S. M. (2007). Dispositional Factors in Internet use: Personality versus cognitive style. MIS Quarterly, 31(4), 809-820.

13

Big Five and Social Networking

14

Table 1 Ranges, Standard Deviations, Means, & 95% Confidence Intervals for Each Individual's Trait

Individual Trait

Range

SD

Mean

95% Confidence Interval Lower Bound

Upper Bound

Self Reported Value *

Christie Openness

5

1.639 3.42

3.04

3.80

2.8

Christie Conscientiousness

5

1.497 3.24

2.90

3.59

4.4

Christie Extraversion

5

1.476 4.01

3.67

4.36

2.4

Christie Agreeableness

5

1.213 3.30

3.02

3.58

4.6

Christie Neuroticism

5

1.184 3.45

3.17

3.72

1.7

ArmyJeep Openness

5

1.618 3.72

3.34

4.00

4.6

ArmyJeep Conscientiousness

5

1.313 3.95

3.64

4.25

5.6

ArmyJeep Extraversion

5

1.218 4.46

4.18

4.74

5.5

ArmyJeep Agreeableness

5

1.159 4.00

3.73

4.27

5.1

ArmyJeep Neuroticism

5

1.208 2.66

2.38

2,94

1.1

Vila Wolf Openness

5

1.438 3.04

2.71

3.37

5.6

Vila Wolf Conscientiousness

5

1.342 2.92

2.61

3.23

4.3

Vila Wolf Extraversion

5

1.452 2.59

2.26

2.93

2.7*

Vila Wolf Agreeableness

5

1.102 2.27

2.02

2.53

4.5

Vila Wolf Neuroticism

5

1.664 3.26

2.87

3.64

0.1

Bobby Openness

5

1.406 3.68

3.35

4.00

5.8

Bobby Conscientiousness

5

1.385 3.58

3.26

3.90

3.6*

Bobby Extraversion

5

1.432 3.39

3.06

3.72

5.4

Bobby Agreeableness

5

1.341 3.85

3.54

4.16

5.0

Bobby Neuroticism

5

1.346 3.45

3.13

3.76

2.1

Rachel Openness

5

1.223

5.11

4.82

5.39

4.9*

Rachel Conscientiousness

5

1.505 3.62

3.27

3.97

4.2

Rachel Extraversion

5

1.264 5.07

4.77

5.36

4.8*

Rachel Agreeableness

5

1.395 3.74

3.42

4.07

4.2

3.72

1.9

Rachel Neuroticism 5 1.422 3.39 3.06 *Self-reported value captured within the 95% confidence interval

Big Five and Social Networking

15

Table 1, continued Ranges, Standard Deviations, Means, & 95% Confidence Intervals for Each Individual's Trait

Individual Trait

Range

SD

Mean

95% Confidence Interval Lower Bound

Upper Bound

Self Reported Value *

Karl Openness

5

1.397 3.53

3.20

3.85

5.9

Karl Conscientiousness

5

1.452 4.03

3.69

4.36

3.9*

Karl Extraversion

5

1.495 3.77

3.42

4.12

6.0

Karl Agreeableness

5

1.072 4.59

4.35

4.84

6.0

Karl Neuroticism

5

1.344 2.97

2.66

3.28

0.6

Llauren Openness

5

1.414 4.20

3.88

4.58

5.4

Llauren Conscientiousness

5

1.321 2.70

2.40

3.01

5.2

Llauren Extraversion

5

1.453 4.74

4.41

5.08

5.3

Llauren Agreeableness

5

1.567 3.36

3.00

3.73

4.7

Llauren Neuroticism

5

1.458 3.77

3.43

4.11

2.1

Mark Openness

5

1.330 4.36

4.06

4.67

4.8

Mark Conscientiousness

5

1.382 3.38

3.06

3.70

4.3

Mark Extraversion

5

1.294 4.32

4.02

4.62

5.2

Mark Agreeableness

5

1.364 3.69

3.37

4.01

5.1

3.23

0.5

Mark Neuroticism 5 1.275 2.93 2.64 *Self-reported value captured within the 95% confidence interval

Big Five and Social Networking

16

Table 2 Significant ANOVAs measuring the accuracy of predicting a trait (measured by the deviation scores) based upon each trait and the respondent’s gender

Individual Trait Christie Extraversion Bobby Agreeableness Rachel Openness Rachel Extraversion Llauren Openness Llauren Extraversion

Absolute deviation scores Male 1.2545 2.0000 1.4636 1.7455 2.2364 1.9727

Female 1.9948 1.2727 0.9052 1.0649 1.2079 1.0395

ANOVA F(1,86) = 4.289, p= 0.041 F(1,86) = 4.081, p= 0.046 F(1, 86) = 5.059, p= 0.027 F(1,86) = 8.283, p= 0.005 F(1,85) = 7.271, p= 0.008 F(1,85) = 7.251, p= 0.009

Big Five and Social Networking Figure Captions Figure 1. Inaccuracy of predictions for neuroticism Figure 2. Inaccuracy of prediction on all personality traits for participant Christie

17