The methods of the Frame studies 1

1
The methods of the Frame studies 1 2
Kyungil, Andrew, Brian, and Patrick
Project Boss Arthur B. Markman
Project Manager Leora Orent
2
Common methods (Frame manipulation)

• Procedure
• Materials 8 activities
• 4 exciting activities video game, solitaire,
  blackjack, and dart game
• 4 unexciting activities alphabetizing,
  proofreading, transcribing video/audio tapes, and
  surveying by phone,
• Choice tasks
• Task 1 Which is the one you would most like to
  do?
• Task 2 Which activity is the one you are most
  interested in not doing?
• Frame manipulation (see instruction used below)
• A subtraction task (5 minutes)
• Instructions for frame manipulation
• If you do well on the subtraction task,
• Promotion you will be allowed to do the task
  you chose as you most would like to do for 15
  minutes, otherwise you will not.
• Prevention you will not have to do the task you
  chose as you most would not like to do, otherwise
  you will for 15 minutes.

3
Common methods

• Manipulation of Need to smoke (NTS) imagery task
  (CepedaBenito Tiffany, 1996)
• Materials 6 different stories
• For High need to smoke (HNS) group each story
  described smoking cravings (urge).
• For Low need to smoke (LNS) group each story
  described smoking unrelated things (no-urge).
• Except for the above difference, stories are the
  same between conditions. (see the two examples in
  the last page)
• Procedure (after each story)
• Imagery task after reading each story (8 sec)
• Please imagine the situation described in the
  sentence.
• Self-description task
• Please write down the sentence that we asked you
  to imagine
• Others vividness rating / positive experience
  rating / negative experience rating

4
Common methods

• Coffee Manipulation
• Ask subjects to drink a cup of coffee
• High Need to Smoke (HNS) group just coffee
• Low Need to Smoke (LNS) group coffee smoked a
  cigarette
• We controlled even cigarette brand ?

5
Common methods (Scales)

• Manipulation check
• Scale (Cox, Tiffany, Christen, 2001)
• The brief questionnaire of smoking urges
  (QSU-brief)
• 10 questions ex) I have a desire for a
  cigarette right now.
• 7-point scale
• Cigarette length
• 14 different lengths (real length between 7th
  and 8th)
• UWIST mood adjective checklist
• 24 adjectives / 4 subscales
• We measured three subscales
• Hedonic Tone e.g., happy and sad
• General Arousal e.g., vigorous and
  unenterprising
• Tense arousal e.g., tense and calm
• Citation Matthews, G., Jones, D. M.,
  Chamberlain, A. G. (1990). Refining the
  Measurement of Mood - the Uwist Mood Adjective
  Checklist. British Journal of Psychology, 81,
  17-42.

6
Procedures

• Experiment 1
• Frame manipulation ? Coffee manipulation ? NTS
  manipulation ? manipulation check ? Item rating
• Experiment 2
• Frame manipulation ? Coffee manipulation ? NTS
  manipulation ? manipulation check ? Concern rating

7
Rating materials

• Experiment 1 5 item categories
• Cigarette brands
• Instrumental items / affiliated items / related
  beverages
• Smoking unrelated items
• Experiment 2 4 concern categories
• Smoking related health concerns e.g., lung
  cancer
• Unrelated health concerns e.g., skin rash
• Unrelated non-health concerns e.g., tax reporting

8
Legend for SPSS outputs

• Conditions
• Frame condition CD1_FRM (1 Promotion / 2
  Prevention)
• Need condition CD2_NEED (1 High / 2 Low)
• Categories in Experiment 1
• Cxxxxxxxxxxxxxxxxxxxxxxx
• Cxxxxxxxxxxxxxxxxxxxxxxx
• Categories in Experiment 2 (z-score)
• Smoking related health concern Z-SRHC
• Smoking unrelated health concern Z-SURHC
• Smoking unrelated non-health concern Z-SURNHC

9
Experiment 1
10
Descriptive data Scales others
UWIST
QSU Length
11
Experiment 2
12
Descriptive data scales others
13
Descriptive data age gender
14
Manipulation check - QSU

• Significant effect of Need manipulation
  (CD2_NEED)
• No effect of frame (CD1_FRM) manipulation

15
Manipulation checks QSU (cont.)
Promotion
Prevention

16
UWIST

• No systematic difference in subscales
• Hedonic Tone
• Tense Arousal
• General Arousal

17
(No Transcript)
18
3-way interaction

• Significant 3-way interaction frame, need, and
  concern category
• No systematic effect of Age or Gender
• Age F(1,149)0.46, p .831 / Gender
  F(1,149).927, p.337

19
Promotion condition

• No significant difference in ratings for each
  concern category between the two Need conditions

20
Promotion condition (cont.)
21
Prevention condition

• Significant differences in smoking related health
  concerns (valuation) and unrelated health
  concerns (devaluation) between the two Need
  conditions ? center-surround?

22
Prevention condition (cont.)
23
Others individuals total concern ratings

• No effect of the two conditions on overall
  concern ratings (e.g., mean of total ratings)

24
Others correlations (Total data)

• Overall, smoking need is weakly related to
  ratings for concern categories

25
Others correlation (Promotion)

• No significant correlations between Need and
  concern ratings

26
Others correlation (Prevention)

• Smoking need has a significantly positive
  relationship with smoking unrelated health concern

27
?????
28
Length Data (collapsed data)

• Surprisingly, there is a significant difference
  in participants length rating between conditions.
  
• But this result comes from a data set of Exp1
  Exp2 Exp3. Only with Exp 1 and 2, the t-value
  is a little smaller than needed.
• Further we do have a significant correlation
  between QSU scale value and this length rating.

29
Length data only with data from Exp 12
30
Example of stories in NTS manipulation

• I am with my friends and we are laughing and
  having a good time. As I watch some of them
  lighting up I think about joining them, knowing
  that smoking would make things even better. (for
  HNS group)
• I am with my friends and we are laughing and
  having a good time. As I watch some of them
  playing cards I think about joining them, knowing
  that it would make things even better. (for LNS
  group)

31
Subject Analysis With a collapsed data of Exp 1
2
32
Test of interaction

• Number of Participants

• No interaction between condition (high vs. low)
  and Experiment (1 and 2)

33
Test of Exp (2)X Condition (2) X Item Type (5)
34
Collapsed data of Exp 1 2 (Subject Analysis)

• Manipulation check
• High gt Low

35
Collapsed data of Exp 1 2 (Subject Analysis)

• Comparison of preference scores (z-score) between
  conditions
• For detail, see the next page.

36
(No Transcript)
37
Collapsed data of Exp 1 2 (Subject Analysis)

• Individual ANOVAs
• For detail, see the next page.
• The last item is of collapsed data (Cigarette
  Brands Instrumental items)

38
Individual t-tests
39
Item Analysis With a collapsed data of Exp 1 2
40
Individual ANOVAs

• Clearer pattern (z-score data)
• For post-hoc analysis, see the next page

41
(No Transcript)