. Using language to examine social phenomena

1
.Using language to examine social phenomena

• Dr Karen Douglas
• k.douglas_at_kent.ac.uk

2
Communication

• Many social processes are facilitated by language
• transferring knowledge and information
• including expressing how we are feeling, our
  attitudes
• how senders inform others what they are
  thinking, feeling etc.

3
Language use

• Implicated in most social psychological phenomena
• e.g., affiliation speech accommodation
• e.g., social perception accents, interrupting
• Dependent on social psychological factors
• e.g., power interrupting, politeness
• Given its importance, there is very little
  research on language and social phenomena

4
This talk

• Recent developments highlighting the importance
  of language in social processes
• linguistic abstraction interpersonal and
  intergroup attributions
• natural language use social, personality,
  cognitive and biological processes
• ethnonyms, ethnophaulisms prejudice and
  intergroup hostility

5
Linguistic abstraction

• Attributions about people and groups

6
Linguistic abstraction

• Linguistic category model
• (Semin Fiedler, 1988)
• descriptive action verb
• Beavis hit Butt-head
• interpretative action verb
• Beavis hurt Butt-head
• state verb
• Beavis hates Butt-head
• adjective
• Beavis is aggressive

7
Coding language abstraction

• Coding guidelines
• http//www.cratylus.org/
• go to resources
• Abstraction becomes a value between 1 and 4
• The person I know bought flowers for his
  girlfriend (1)
• My friend was violent in the nightclub (4)

8
How would you describe this behaviour?

• Mike Tyson is nibbling Evander Holyfields ear
• OR
• Mike Tyson is violent

9
How would you describe this behaviour?

• Mike Tyson is playing with a child
• OR
• Mike Tyson is sweet

10
Linguistic expectancy bias

• Expected behaviours ? abstract description
• Unexpected behaviours ? concrete description
• Wigboldus, Semin and Spears (2000)

11
Ingroups and outgroups

• The red teams player is aggressive
• OR
• The read teams player fouled the other player

12
Linguistic intergroup bias

• Stereotypical behaviours ? abstract description
• Counter-stereotypical behaviours ? concrete
  description
• Maass, Salvi, Arcuri and Semin (1989)

13
Language abstraction and biased communication

• Way of transmitting stereotypes? prejudice?
• Wigboldus, Semin Spears (2000)
• language abstraction influences beliefs of
  recipients
• passing on beliefs to others

14
Self talk and language abstraction
I bit his ear I am aggressive
I played with the child I am sweet

• Self-enhancing abstraction bias (Tanis, 1999)

15
Language abstraction and social influence

• Communicators are able to alter their language
  abstraction in response to communication goals
  (Douglas Sutton, 2003)
• e.g., to create a specific impression of a target
  for a recipient
• They can also inhibit biased use of language
  abstraction (Douglas, Sutton Wilkin, 2008)
• e.g., make your description as neutral as
  possible

16
Conscious or unconscious?

• Language abstraction is not necessarily under
  intentional control
• Franco Maass (1996, 1999)
• von Hippel, Sekaquaptewa Vargas (1997)
• Way of indirectly measuring peoples attitudes?
• also the effects of their explicit or implicit
  communication goals?

17
Does what we say about others say something about
us?

• Douglas and Suttons (2006) method
• participants view positive and negative
  behaviours
• description next to each scene
• one of the four levels of language abstraction
• e.g., Matthew is responsible
• What do you think is the describers attitude
  towards Matthew?

18
Results

• What is the describers attitudes towards
  Matthew?
• negative behaviours higher abstraction
  perceived bias against target
• positive behaviours higher abstraction
  perceived bias in favour of target

likelihood ratings
likelihood ratings
Negative behaviours
Positive behaviours
19
Other findings

• Recipients also make inferences about
• describers relationship with the target
• describers communicative intent
• describers likeability and politeness
• describers of positive behaviours are deemed more
  likeable when they word their descriptions
  abstractly
• opposite for negative behaviours

likelihood ratings
20
Functions of language abstraction Importance for
social phenomena

• Expectancies
• Stereotypes
• Influence (communication goals)
• Consequences of biased language use for
  describers
• what we say about others says something about us
• A strong way of experimentally examining language
  use and social phenomena
• largely implicit
• avoids socially desirable responding

21
Natural language use

• Social, personality, cognitive, biological and
  cultural processes

22
Natural language

• A lot of things can be achieved in a controlled
  laboratory setting a lot cannot
• Natural language is much less studied
• Why?
• costly, time-consuming
• But
• new software and methods makes text analysis
  easier
• e.g., Linguistic Inquiry and Word Count (LIWC,
  Pennebaker , Francis Booth, 2001)

23
What can we learn from natural language?

• How the words people use in their daily
  interactions reflect who they are and what they
  are doing
• biological activity
• mental health
• responses to stress
• deception
• status
• demographics
• culture

24
Function words

• Work of James Pennebaker and colleagues
• Function words, or particles
• I, The, And, To, A, Of, That
• approximately 0.04 of our vocabulary
• but over ½ the words we use in daily speech
• The way people use function words reflects their
  linguistic style
• different to using nouns and verbs
• I can believe that he gave that to her
• I cant believe that he gave her the ring
• I cant believe that he gave her a ring
• These are all different
• use of function words requires a sophisticated
  set of social skills to get meaning across

25
I / non-I words and social processes in
everyday language

• Biological activity
• Pennebaker, Groom, Loew Dabbs (2004)
• higher levels of testosterone associated with
  less non-I pronouns
• focus taken away from other people as social
  beings?
• Depression
• Mehl (2004) Stirman Pennebaker (2001)
• use of I is more frequent among people with
  high depression scores
• poets who committed suicide used I words more
  than non-suicidal poets
• disengaged from others? Too much focus on the
  self?

26
Function words and social processes

• Deception
• Newmann, Pennebaker, Berry Richards (2003)
• when people are telling the truth, they own it
  by using more I words and exclusive words
  (e.g., but, except) which predict honesty
• Status
• Pennebaker Davis (2006)
• in a dyad, the person who uses I less is the
  more dominant
• Other findings
• women use more I words than men
• younger people use more I words than older
  people

27
Function words and social processes

• Individual stress
• Pennebaker Lay (2002)
• Rudolph Guilianis use of I and similar
  increased in speeches from 2 to 7 in 2000
  (personal difficulties)
• decreased use of we
• Socially-shared stress
• Campbell Pennebaker (2003) Cohn, Mehl
  Pennebaker (2004)
• strong association between non-I pronoun use
  and health following stress
• collectively stressful events like Princess
  Dianas death and 9/11 followed by increase in
  non-I pronoun use

28
Function words and social processes

• Culture
• is we more common in collectivist cultures and
  I more common in individualist cultures?
• not necessarily the case self focus is
  important to achieve collectivist objectives
  (e.g., Markus Kitayama, 1991)
• but also note pronoun drop in some languages
  (Kashima Kashima, 1998) which indicates
  collectivist thinking

29
Descriptions of everyday objects and pictures

• Do such descriptions reveal information about the
  self?
• This example is from
• http//www.utpsyc.org/Bottle/index.html
• Others available from Pennebakers university
  website
• http//www.utpsyc.org

30
Instructions

• Look at the picture. How would you describe the
  picture to someone who couldn't see it? Try to
  describe the picture in a way that someone you
  didn't know could imagine exactly what it looks
  like just by your description.
• Once you are ready to describe the object, press
  the button below. You will have 5 minutes to
  write your description of the picture. Try to
  write continuously about what you see. Do not use
  abbreviations and try your best to spell
  correctly.
• Once you are ready to begin, press the button
  below.
• ltlt startgtgt

31
My description

• It is a bottle. The bottle is made of clear
  plastic and has ridges around its surface in the
  lower half of the bottle. The bottom of the
  bottle appears to be flat. The bottle is thicker
  at the bottom and tapered towards the opening at
  the top, as bottles typically are. It appears to
  be filled with a clear liquid, or is empty. The
  lid of the bottle is white, plastic and appears
  to screw to open. There is a label around the
  bottle in the middle to top part of the bottle.
  The label is predominantly red in colour with
  some writing on it. I can't read what it says
  but it begins "Oz...". I assume that this larger
  text depicts the product name and that the
  smaller writing indicates the ingredients in the
  bottle and some product information. There is a
  picture on the bottle, of a mountain. The
  product is probably therefore spring water. The
  volume of liquid is also indicated on the bottle
  but I can't read it. It is a nice, shiny bottle!

32
My results!
Visual dimension Your data Average
Words on the label verbal thinking 1.68 1.74
Colours and text visual sensitivity 2.23 3.74
Bottle contents functional thinking 6.15 1.67
The bottle itself tactile sensitivity 4.47 2.91
Light and shadow Contextual thinking 0.00 0.79
33
Example of written feedback

• The bottle itself Tactile sensitivity. Someone
  who is high in tactile sensitivity likes to touch
  things. They appreciate the surface, texture, and
  contours of objects and people. They have an
  appreciation of the dimensionality of objects. In
  describing the bottle, high scorers paid
  attention to the surface and contour of the
  bottle itself, perhaps thinking of it like a
  sculpture. Your writing suggests a true
  appreciation of form, depth, and touch. In
  looking at the bottle, you tend to see it as a
  living object. You can imagine it in your hand
  with an awareness of its surface and contour. You
  may be a sensual person who appreciates the
  complexity of objects and people. You may also
  have a fetish for bottles.
• Try it for yourself!

34
The methods behind this analysis

• Mathematically determines the dimensions that
  most people see in an image
• meaning extraction
• Task first performed by a several hundred people
• Text analysed for common words
• Words are factor analysed for words the hang
  together
• yellow, banner, writing, letters, paragraph
• ka, right, left, oza, side, read
• psychological value is attached to these meaning
  clusters

35
Conclusions about natural language

• The words we use reflect our thoughts, feelings
  and characteristics in often unpredictable ways
• Useful for social science research because
  self-reports have shortcomings
• are peoples responses to scales really
  objective?
• analysing everyday language can be naturalistic
  and unobtrusive
• rich source of information which is easy to
  obtain

36
Problems with using natural language

• Without aid of computers, this is painstaking
  work
• Even with computer programmes, some words can be
  miscoded
• I am mad about him
• Hes as mad as a hatter
• Recognising the importance of natural language
  use for measuring social phenomena has led to
  vast improvements in software

37
Ethnonyms and ethnophaulisms

• Prejudice, conflict and intergroup hostility

38
Language and group hostility

• Earliest stages of intergroup conflict are often
  evidenced by the use of derogatory names
  (Allport, 1954)
• Ethnophaulisms
• Greek origin a national group to disparage
• words used as ethnic slurs to refer to groups
• commonly reported in media reports of interethnic
  conflict

• http//www.everythreeweekly.com/archive/show_story
  /348
• Lets standardize our insults!
• Everyones just a dirty Euroff!

39
Ethnophaulisms

• Methodology examining ethnophaulisms focuses on
  archival analyses of variables such as valence
  and linguistic complexity
• Complexity
• when an array of terms can be categorized into
  fewer categories and most can be clustered in one
  category, this represents low complexity in
  cognitive representation
• e.g., slut, tart, slag
• Scotts H statistic

40
Ethnophaulisms

• Mullen (2001) ethnic groups that are smaller,
  less familiar and more foreign are referred to
  with terms that are
• less complex
• more negatively valenced
• Groups targeted with simple, negative terms are
  more likely to be the target of intergroup
  hostility
• Mullen, Leader Rice (2005)

41
Ethnophaulisms

• Mullen (2004)
• analysed childrens books about ethnic minorities
  (immigrants to USA)
• fewer books and shorter books were written about
  ethnic immigrant groups that are represented by
  less complex ethnophaulisms
• similar results for descriptions of ethnic
  minority children and portrayal of ethnic
  minority children in books

42
Referring to the ingroup

• Research has focused on names for the outgroup
• BUT people also distinguish themselves from the
  outgroup by using terms to describe themselves
  (ingroup)
• Ethnonyms
• Greek a national group name
• what is their impact on intergroup behaviour?

43
Ethnonyms

• Why are they important?
• anchorage point for personal identity
• people with more socially desirable names tend to
  be more popular, live longer and are better
  adjusted
• verbal realism
• group members may respond to their own names as
  strongly as they would respond to the things
  represented by the names
• Do ethnonyms influence the degree of intergroup
  hostility displayed by groups?

44
Archival studies

• Mullen, Calogero Leader (2007)
• examined the relationship between ethnonyms and
  ethnographic data on intergroup conflict
• used a database of over 4500 names representing
  over 2000 African peoples (nations, empires,
  chiefdoms, tribes, villages)
• used ethnographic data relevant to intergroup
  hostility for 125 African cultures (warfare
  prevalence, military glory, killing enemies)
• Found that intergroup hostility varies as a
  function of an ingroups ethnonyms
• mainly due to ethnonym complexity
• the simpler the ethnonym, the more conflict with
  outgroups
• greater complexity ? greater tolerance of
  outgroups?

45
Archival studies

• Mullen et al. (2007)
• ethnonyms in North America
• native cultures
• African Americans
• results largely replicated
• ethnonyms are collective representations that
  stand as symbols of the ingroup
• these words are powerful
• the more simple the ethnonyms, the higher the
  conflict with other groups

46
Archival research

• The words we use to describe ourselves and others
  can predict relationships between groups
• Useful for social science research
• real groups, real language about real
  social interactions
• in terms of what to analyse, the skys the limit

47
Conclusions

• Todays talk
• words and language are important
• a snapshot of some of the social phenomena that
  can be examined through language
• some of the methods we can use to examine them
• experimental, naturalistic and archival methods
• Not in todays talk
• gesture
• nonverbal communication tells us a lot about
  personality, status etc.
• features of language such as speech rate and
  accent
• allow people to make inferences about speakers
  status, personality etc.
• gender differences in language use
• womens language, gender roles and status
• Language is a rich source of social information
  which is largely ignored by social scientists

48
Some key readings

• Chung, C., Pennebaker, J. (2007). The
  psychological functions of function words. In K.
  Fiedler (Ed.), Social communication (pp.
  343-359). New York Psychology Press.
• Mullen, B. (2004). Sticks and stones can break my
  bones, but ethnophaulisms can alter the portrayal
  of immigrants to children. Personality and
  Social Psychology Bulletin, 30, 250-260.
• Wigboldus, D.H.J., Douglas, K.M. (2007).
  Language, expectancies and intergroup relations.
  In K. Fiedler (Ed.), Social communication (pp.
  79-106).  New York Psychology Press.