Psychster LLC online psychological research - PowerPoint PPT Presentation
1 / 17
Title:
Psychster LLC online psychological research
Description:
a link to funny video. 0.096. 0.000. my spirituality. 0.051. 0.044. what makes me ... moments, proud moments, heroes, and when they gave links to funny videos. ... – PowerPoint PPT presentation
Number of Views:81
Avg rating:3.0/5.0
Title: Psychster LLC online psychological research
1
What Elements of an Online Social Networking
Profile Predict Target-Rater Agreement in
Personality?
- David C. Evans Ph.D.
- Psychster LLC
- Samuel D. Gosling Ph.D
- University of Texas at Austin
- Anthony Carroll
- Psychster LLC
2
The Science of Interpersonal Perception
- We know a lot about what people think about
others personality, but far less about whether
they are right. - If you study what one live person thinks about a
fake person you get - Impressions. As thoughts about B.
- Stereotypes. Whether A thinks B is like Bs
group. - Consensus. Whether a bunch of As agree about B.
- Projection. Whether A sees B in As own image.
- Variables you can only get if you study two live
people. - Self-Other Agreement or Impression Accuracy
Whether A sees B as B sees him or herself. - Similarity. Whether A's and B's self-ratings are
alike. - Accuracy, Meta-accuracy, Reciprocity, Assumed
Projection
3
Impression Accuracy on the Web
- Social media (networking sites, blogs,
microblogs, photo sites, job sites) have become a
central environment for interpersonal perception
worldwide. - Are people getting to know each other via these
media? Are they at least seeing others as the
others see themselves? Under what conditions?
Flickr. Stabilo Boss.
4
Impression Accuracy on the Web
- The present study addressed the following
questions (see Funder, 1999) - How effectively are profile owners conveying
their personality to visitors? - Are some profile elements more informative than
others? - Are some people better at reading personality
than others? - Are some people more easily read?
5
In Situ Method
- We launched YouJustGetMe on the Facebook
platform. - Users rated their own personalities on 21 items
of the BFI-K (John, 2005) a Big-5 personality
inventory. - Users formed impressions by answering the same 21
items about others via questionnaires imbedded in
others profiles. - Upon adding the app, users also registered on
YouJustGetMe.com. - 4,484 impressions were collected on Facebook
among participants who were 18 and older.
6
In Vitro Method
- We created YouJustGetMe.com a fully functional
social networking website. - We only analyzed the impressions of randomly
assigned targets who had profile photos. - 819 impressions were collected on
YouJustGetMe.com among participants who were 18
and older. - Since then 9,469 impressions have been made by
6,500 registered users .
7
Cmon, give something back.
Docdave saw Rachel like this
Rachel saw Rachel like this
How accurate was Docdave?
- Big-5 domain scores of both the self-ratings and
the guesses were displayed so users learned which
traits they were right and wrong about.
- The central DV of the study impression accuracy
was calculated as the point-biserial Pearson r
product-moment correlation between the
self-ratings and the guesses. It was displayed to
users. Fisher-z transformed rs gave the same
results as the point-biserials.
- Similarity and projection were also calculated
and displayed.
8
Analysis
- Do environments matter (Facebook vs.
YouJustGetMe)? Does sex of rater and sex of
target matter? - Analysis of Covariance (ANCOVA). Unit of analysis
was the dyad. DV was the impression accuracy
score (Pearson r). IVs included environment
(Facebook vs. YouJustGetMe), rater sex, and
target sex. Covaried out age. - Which profile elements (or clues) were
associated with higher or lower impression
accuracy? (YouJustGetMe only.) - Multiple regression. Unit of analysis was the
dyad. Criterion (Y) was the impression accuracy
score (Pearson r). Predictors (x) were 33 binary
variables indicating simply whether a particular
profile element was or was not answered by the
target. Dummy codes tested facial photos (head
and shoulders), non-human photos (tractors, cats,
anime) and other photos (groups, distant shots).
Controlled for rater targets age, sex,
similarity projection.
9
Do environments differ? Does sex of rater and
sex of target matter?
guessing
- People get each other. Overall impression
accuracy was significantly above zero (r .41,
SD .21), t(6862) 147, p - People on Facebook get each other better (r
.42) than randomly assigned strangers on
YouJustGetMe (r .29), F(1,5293) 170.4, p .001. - Women are better guessers than men, but only when
guessing random strangers on YouJustGetMe.
F(1,5293) 12.5, p - Women are easier to get, but only when guessed by
random strangers on YouJustGetMe. F(1,5293)
12.5, p
.43
.42
.33
.25
being guessed
.44
.41
.34
.24
10
Which profile elements matter? Which dont?
11
Disclosure mediates gender effect.
- Women and men completed similar number of profile
elements, p .2. - Women were significantly more likely to complete
profile elements that predicted higher impression
accuracy, ps - Controlling for choice of profile element
eliminated target gender effect, p .3.
12
Discussion Which sources reveal which traits?
Source
Web site
Face book
Bed room
Office
Top-10 Music
Social beh.
Short FTF
Long term
Trait
Extraversion Agreeableness Conscientiousness Neuro
ticism Openness Mean
.29
.22
.37
.24
.42
.29
.14
.45
- Gosling, S.D., Gaddis, S., and Vazire, S. 2007.
- Gosling, S.D., Ko, S.J., Mannarelli, T., and
Morris, M.E. 2002. - Rentfrow, P.J. Gosling, S.D. 2006.
- Watson, D. 1989.
13
Discussion - Specific
- How effectively are profile owners conveying
their personality to visitors? - Online impression accuracy (whether visitors to
Facebook and YouJustGetMe profiles saw the
profile owners as they saw themselves) was
significantly above chance. - People are posting information on their profiles
that is consistent with their self-view of their
personality, and that information is in turn
consistent with the personality impression formed
by visitors. This is also true of the
impressions formed of strangers, which are
overall less accurate that impressions of people
in your network, but still above chance levels. - Are some judges better than others?
- Women formed more accurate impressions of others
than men, but only when they rated random
strangers. On Facebook, were people rated others
in their network, this gender difference
disappeared. - Are some targets more easily judged?
- Impressions of women were more accurate than
impressions of men, but again, only on when they
were rated by random strangers. On Facebook, were
people were rated by others in their network,
impressions of women and men were not
significantly different. - Do certain profile elements hinder or assist the
impression formation process? - Impression agreement was significantly higher
when the profile owners made statements about
their spirituality, beliefs, joys, embarrassing
moments, proud moments, heroes, and when they
gave links to funny videos. - Impression agreement was significantly lower when
the profile owners posted pictures of
non-persons, named an awful person or website, or
surprisingly, named a great book.
14
Discussion - General
- Why are certain profile elements better at
conveying your personality? - Items that go deeper than traits and preferences
to tap issues of values and identity seemed most
revealing. - I understand why women are better at guessing.
Why are they easier to guess? - Female targets were significantly more likely
than male targets to disclose answers on the
profile elements that were better predictors of
impression agreement such as relationship saga
and proudest thing I ever did. - However, male targets were significantly more
likely to disclose their political views, great
art, awful person, awful movie, and awful
website. - Why were effects of rater and target gender found
for the YouJustGetMe profiles and not the
Facebook profiles? - This study cannot answer that conclusively, but
we can speculate. The most likely reason is we
analyzed Facebook impressions of people you chose
but we only analyzed the YouJustGetMe impressions
of people who you were randomly assigned to which
would greatly diminish the likelihood that you
knew them. - But additionally, Facebook profiles have a
greater amount of information. Perhaps impression
agreement had reached a ceiling for the Facebook
profiles.
15
Discussion - General
- Dont some people lie about themselves
purposefully? - If profile owners are lying, they are lying to
themselves on the self-ratings consistently with
how they are lying when constructing their
profiles. - Therefore, the majority are not purposefully
creating a qualitatively different personality
profile. If they are lying, it is a minority, or
on other individual differences, or by
incremental degrees of positive
self-presentation. - We suspect that more often, profile owners are
trying to manage the views they have of
themselves with the impressions they make on a
diverse array of peers, such that they converge
in a beneficial way.
16
Select References
- Funder, D.C. 1999. Personality judgment A
realistic approach to person perception.
Academic. - Green, R.K., Evans, D.C., Gosling, S.D. 2008.
Researching first impressions in the age of
online profiles. Psi Chi, The National Honor
Society in Psychology, 12, 3. - Gosling, S.D., Gaddis, S., and Vazire, S. 2007.
Personality impressions based on Facebook
profiles. In Proceedings of the International
Conference on Weblogs and Social Media (Boulder,
Colorado, USA, March 26 - 28, 2007). - Gosling, S.D., Ko, S.J., Mannarelli, T., and
Morris, M.E. 2002. A room with a cue Personality
judgments based on offices and bedrooms. Journal
of Personality and Social Psychology 82, 3
(2002), 379-398. - John, O.P. 2005. BFI-K (Form S). Berkeley
University of California, Berkeley, Institute of
Personality and Social Research. - Kenny, D.A. 1994. Interpersonal perception A
social relations analysis. Guilford Press. - Rentfrow, P. J., Gosling, S. D. 2006. Message
in a Ballad The role of music preferences in
interpersonal perception. Psychological Science,
17 (2006), 236-242. - Vazire, S. and Gosling, S.D. 2004. e-Perceptions
Personality impressions based on personal
websites. Journal of Personality and Social
Psychology, 87, 1(2004), 123-132 - Watson, D. 1989. Strangers' ratings of the five
robust personality factors Evidence for a
surprising convergence with self-report. Journal
of Personality and Social Psychology, 57, 1
(July, 1989), 120-128.
17
Auxiliary Slides
Recommended
CrystalGraphics Presentations
