Docstoc

Goal

Document Sample
Goal Powered By Docstoc
					New Advances in
   Measurement
    Ronald D. Rogge
TOPICS
 RELATIONSHIP QUALITY
   T1: IRT Optimization
      Study 1

   T2: Responsiveness to Change
      Studies 2-5

   T3: Bi-Dimensional View
      Studies 6-7

   T4: Implicit Measures
      Studies 8-11

 ATTENTION
   T5: Screening for Error Variance
      Studies 12-16
Acknowledgements
 Couples Satisfaction Index
      Janette Funk, Mike Maniaci, Maria Saavedra,
       Soonhee Lee

 Positive-Negative Relationship Quality
      Frank Fincham, Richard Mattson, Matt Johnson

 C.A.R.E. Program
      Tom Bradbury, Rebecca Cobb, Matt Johnson, Erika
       Lawrence, Lisa Story, Lexi Rothman

 Implicit Assessment
      Soonhee Lee, Harry Reis

 Attention / Effort
      Mike Maniaci, Janette Funk, Soonhee Lee, Maria
       Saavedra
Relationship Quality

Relationship satisfaction
    Self-report scales (DAS, MAT, QMI)
    30-50 years of research (over 4K studies)
    Excellent correlational validity

     Level of noise?
     Responsive to change over time?
     Are these the “best” items?
TOPIC 1: IRT Optimization
 Large sample method
       N at least 1,000 in smallest group

 Large item pool
       Unidimensional
       Non-redundant


 Used by ETS
       SAT, GRE, MCAT, LSAT

 Quality of each item
       Information
       Noise

 Advantages
       Over correlations
       Over small sample methods
IRT Approach
 Latent scores (q) for each subject
       Like GRE scores
       Assessing relationship satisfaction

 Parameters for each item
       Response curves
       Higher   q’s  higher responses?

 Item Responsiveness
       How informative?
       Where informative?

 Creates information profiles
       For individual items
       For sets of items
Study 1 - Measures
 141 satisfaction items:
      DAS, MAT, RAS, KMS, QMI, SMD
      71 additional items

 7 anchor scales:
      Neuroticism (EPQ-N)
      Conflict / Communication (MCI, CPQ, IAI)
      Stress (PSS)
      Sexual Chemistry (Eros)
      Instability (MSI)

 2 validity scales:
      Inconsistency (PAI)
      Infrequency (PAI)
Study 1 - Sample

 5,315 online respondents
     After removing:
       • Incomplete or invalid responses
       • Multivariate outliers


     26yo (10yrs)
     83% Female
     76% Caucasian
     26% High school ed. or lower
     $27K average income

     24% married, 16% engaged, 60% dating
Evaluating Previous Scales

IRT results

   Simultaneous analysis


   66 items of existing scales


   Some very informative items


    Many poor items
                               DAS-31
                               (Degree of happiness, all things considered, of your
                               relationship)

                                           1 - Extremely Unhappy
                                           2 - Fairly Unhappy
                                1                                                                         4.5
                                           3 - A little unhappy
                                           4 - Happy
                               0.9                                                                         4
                                           5 - Very Happy
                                           6 - Extremely Happy




                                                                                    .
.




                               0.8
                                           7 - Perfect                                                    3.5
Probability of each response




                               0.7
                                                                                                           3




                                                                                    Information (theta)
                               0.6
                                                                                                          2.5
                               0.5

                                                                                                           2
                               0.4

                                                                                                          1.5
                               0.3

                                                                                                           1
                               0.2


                               0.1                                                                        0.5


                                0                                                                          0
                                     -3   -2        -1          0       1   2   3                               -3   -2     -1        0         1   2   3
                                                         Satisfaction                                                            Satisfaction

                                               Response Curves                                                            Information Curve
                         DAS/MAT 5
                         Agreement on: FRIENDS


                                          1 - Always Disagree
                                1                                                                        4.5
                                          2 - Almost Always Disagree
.




                               0.9        3 - Frequently Disagree                                         4




                                                                                   .
                                          4 - Occasionally Disagree
Probability of each response




                               0.8
                                                                                                         3.5
                                          5 - Almost Always Agree

                               0.7        6 - Always Agree
                                                                                                          3




                                                                                   Information (theta)
                               0.6
                                                                                                         2.5
                               0.5

                                                                                                          2
                               0.4

                                                                                                         1.5
                               0.3


                                                                                                          1
                               0.2


                               0.1                                                                       0.5


                                0                                                                         0
                                     -3   -2       -1        0         1   2   3                               -3   -2      -1        0         1   2   3
                                                        Satisfaction                                                             Satisfaction

                                               Response Curves                                                           Information Curve
                               MAT 12
                               In leisure time, do you (and does your mate) prefer to be “on the
                               go” or to stay at home?

                                1          1 - Mismatch                                                4.5
.




                               0.9         2 - Both on-the-go                                           4




                                                                                 .
                                           3 - Both stay-at-home
Probability of each response




                               0.8
                                                                                                       3.5


                               0.7
                                                                                                        3




                                                                                 Information (theta)
                               0.6
                                                                                                       2.5

                               0.5
                                                                                                        2
                               0.4

                                                                                                       1.5
                               0.3

                                                                                                        1
                               0.2

                                                                                                       0.5
                               0.1


                                0                                                                       0

                                     -3   -2     -1        0         1   2   3                               -3   -2   -1        0         1   2   3

                                                                                                                            Satisfaction
                                                      Satisfaction
                                               Response Curves                                                     Information Curve
From Items to Scales

A scale‟s information

   = sum of information from each item




How informative

   Across different levels of happiness
Test Info for Current Measures
 .

                             70

                                                                        DAS (32)
 Relative Test Information



                             60
                                                                        QMI (6)
                             50                                         RAS (7)
                                                                        MAT (16)
                             40
                                                                        DAS-4 (4)
                             30



                             20



                             10



                              0

                                  -3   -2   -1        0         1   2        3

                                                 Satisfaction
Analysis of Existing Measures

Many uninformative items
   Particularly for DAS and MAT
      noise / error


Modest test information
   For all scales
   Notably poor for MAT and DAS


Room for improvement
Creating the CSI

141 item pool

   Screen for contaminating items

   Screen for redundant items

   IRT on remaining 66 items

   Select 32 most effective
Test Info for CSI Scales
               70                                         CSI-32
                                                          CSI-16
               60
                                                          DAS (32)
                                                          MAT (16)
               50
 .
 Information




               40



               30



               20



               10



                0
                    -3   -2   -1        0         1   2        3

                                   Satisfaction
Basic Psychometrics

                     Distress               Correlations
            Alpha
                    Cut Score   1     2      3       4     5     6

1. DAS       .94      97.5      --    --     --      --    --    --

2. MAT       .84      95.5      .90   --     --      --    --    --

3. QMI       .96      24.5      .85   .87    --      --    --    --

4. RAS       .92      23.5      .86   .87   .91      --    --    --

5. CSI-32    .98      104.5     .91   .91   .94     .96    --    --

6. CSI-16    .98      51.5      .89   .90   .96     .95    .98   --

7. CSI-4     .94      13.5      .87   .88   .93     .94    .97   .97
Correlations with Anchors
                                                              Sexual
         Arguing   Instability   Comm.   Stress   Conflict
                                                             Chemistry
                                                                         Neuroticism

          IAI        MSI         CPQ     PSS       MCI        EROS        EPQN


DAS       -.79       -.74        .73     -.53     -.54         .42         -.40


MAT       -.76       -.74        .69     -.49     -.49         .41         -.38


CSI 32    -.79       -.78        .71     -.52     -.48         .45         -.38


CSI 16    -.80       -.78        .71     -.53     -.49         .43         -.38
What have we gained?
 Identical correlational results
     Strong convergent validity
     Strong discriminant validity
     Strong construct validity
        Measuring same thing

 Higher information…

  Should have
     Lower Noise
     Higher Precision
     Greater Power
Satisfaction Groups
 IRT satisfaction estimates
      For each subject
      Based on MAT, DAS, & CSI items

      (equivalent of GRE scores)



 Created satisfaction groups
      N = 265
      HIGHLY similar SAT within each group



 MAT, DAS & CSI scores also similar?
Precision: CSI-32 vs. DAS
Effect Size

 Ability to detect difference
     Between groups
     Pre – Post


 Effect Size =    M1 – M2 .
                  pooled SD

 Difference in SD units

 Power for detecting D‟s in SAT groups
Power: CSI-32 vs. DAS
                            2.5
 .


                                                                                   DAS
 Effect Sizes (Cohen's d)



                             2                                                     CSI(32)
                            1.5


                             1


                            0.5


                             0
                                  2   3   4   5   6   7   8   9 10 11 12 13 14 15 16 17 18 19 20

                                      Adjacent Satisfaction Group Contrasts
STUDY 1 - Conclusions

CROSS-SECTIONALLY
   CSI assess same construct
   Higher precision
   Higher power


NEXT STEP
   Longitudinal analysis
   Responsiveness to change over time
TOPIC 2: Responsiveness
 Detecting change
       Assumption
       Longitudinal

 External Criteria
         Treatment effect
         Clinician
         Interviewer
         Global report

 SERM (Sdiff)
       Noise over time
       Estimating

 Two main applications
       Individual change
       Clinically distinct groups
Studies 2 through 4
 Study 2
   267 online respondents
      1 & 2wk follow ups
      468 change scores

 Study 3
   156 online respondents
      1, 2, 3, 4, 6 & 12mo follow ups
      455 change scores

 Study 4
   545 online respondents
      1, 2, 3, 4, 6, 9 & 12mo follow ups
      1,552 change scores
Studies 2-4: Measures

Relationship satisfaction scales:
    DAS-32
    MAT-15
    CSI-32

    CSI-16*
    CSI-4*


3 global relationship change items
    Change since last assessment
Individual Change
 How many points of change needed?
  (to show significant change)

   SERM in “No Change”

     RCI (Jacobson & Truax, 1991)
     MDC95 (Stratford et al., 1996)

    MDC95 (SD units) = 1.96*SERM .
                          SD

   PRESENTING
     Meta-Analytic Summary
     Standardized Units
               Reliable Individual Change
                          .8                                        C
                                                   C
SD's to Show Sig Change




                          .7                               C

                          .6              B
                                  A
                          .5

                          .4

                          .3

                          .2

                          .1

                          .0
                               CSI-32   CSI-16   CSI-4   DAS-32   MAT-15
Detecting Change

Individual Change
   IRT optimization
   Longer scales


Distinct Groups
 Can scales distinguish?
   Mild deterioration
   No change
   Mild improvement
Perceived Change
 Much    Somewhat   A little   Stayed the    A little   Somewhat    Much
WORSE     WORSE     WORSE        SAME       BETTER       BETTER    BETTER

 -3         -2        -1           0          +1          +2        +3




 How much have these changed?
         Overall happiness in the relationship
         Feeling close and connected
         Stability of the relationship
Perceived Change
 Much    Somewhat   A little   Stayed the    A little   Somewhat    Much
WORSE     WORSE     WORSE        SAME       BETTER       BETTER    BETTER

 -3          -2       -1           0          +1          +2        +3
   Significant      Minimal       No         Minimal        Significant
  Deterioration      Deter.     Change      Improve.       Improvement




 Averaged responses
         Alpha = .92


 Created change groups
Distinct Change Groups
                         20
                                  DAS-32 Change Scores (with confidence intervals)
                         15
                                  CSI-32 Change Scores
 .




                         10

                          5
 Average Change Scores




                          0

                          -5

                         -10

                         -15

                         -20
                                 Minimal             No               Minimal         Significant
                               Deterioration       Change           Improvement      Improvement
                         -25
                                         Self-Reported Global Change
Quantifying Group-Level
Responsiveness
MCID
   (Guyatt, Walter & Norman, 1987)

   Noise over time (SERM)

   Effect Sizes:

     (Avg Change)IMPROVE – (Avg Change)NO CHANGE
                     SERM

     (Avg Change)DETERIORATE – (Avg Change)NO CHANGE
                     SERM
Analytic Strategy
 Improving method
     Multi-wave data
     Global change continuous
     Moderation


 HLM
     PV: Global change score
     Moderators:
        • Gender
        • T0 Satisfaction

     DV: Change scores on scales (n = 2475)
             Change scores ≈ 1pt global change
             MCID effect sizes

 Meta-Analytic Summary
            Responsiveness in Dissatisfied
            (1SD below M)
                                                           1.5
Effect Size to Detect 1 Point of




                                                                                       A   B
                                                           1.0
                                                                                               C
                                   Average Global Change




                                                                                                   C    C
                                                           0.5

                                                           0.0

                                                           -0.5                                        CSI-32
                                                                                   D                   CSI-16
                                                           -1.0                C
                                                                                                       CSI-4
                                                                           B
                                                           -1.5                                        DAS
                                                                  A   A
                                                                                                       MAT
                                                           -2.0
                                                                  Mild Deterioration   Mild Improvement
          Responsiveness in Satisfied
          (1SD above M)
                                                           0.5                          A   A   B   B    B
Effect Size to Detect 1 Point
                                of Average Global Change



                                                           0.0

                                                           -0.5
                                                           -1.0
                                                                              E
                                                           -1.5
                                                                                  D                     CSI-32
                                                           -2.0
                                                                                                        CSI-16
                                                           -2.5           C                             CSI-4
                                                                                                        DAS
                                                           -3.0   A   B
                                                                                                        MAT
                                                           -3.5
                                                                  Mild Deterioration   Mild Improvement
Responsiveness Conclusions
 Can be quantified
      Scale selection
      Power estimates

 Responsive scales
      Greater power
        • Individual
        • Group


 Cross-sectional  Longitudinal
      Precision & Power translate


 NEXT STEP  Treatment Effects
Topic 3: Bi-Dimensional View

Uni-Dimensional view
   Positive feelings opposite negative
    feelings

Bi-Dimensional view
   Pos/Neg independent

   Moderately “dissatisfied”
     • Ambivalent
     • Indifferent

   Uni-Dimensional obscuring?
Background
 Fincham & Linfield (1997)
  PN-QIMS
    Two 3-item scales
      • Qualities of spouse
      • Feelings toward spouse
      • Feelings about marriage

    Considering only (pos/neg)
    Separated in survey

  CFA in 123 couples

  Unique information
Study 5
 Mattson et al. (under review)

 New pos-neg scale
    7 SMD items of CSI
    Pos / neg separately

         Positive Items             Negative Items

   Interesting                   Boring
                    Sturdy                     Fragile
    Enjoyable                  Miserable
                     Good                       Bad
     Friendly                    Lonely
                      Full                     Empty
     Hopeful                  Discouraging

 Large online sample
    Ambivalent
    Indifferent
Study 5 - Sample
 1656 online respondents
  Demographics
     28yo (7yrs)
     94% Female
     87% Caucasian
     30k income
     5% ≤ high school

  Romantic relationships
     38% married (6.5yrs)
     19% engaged (3.6yrs)
     41% dating – exclusive (2.4yrs)
Ambivalence vs. Indifference

 Median Splits


                         Positives

                     High         Low

            Low     Satisfied   Indifferent
Negatives
            High   Ambivalent   Distressed
Ambivalence vs. Indifference

 Median Splits


                         Positives

                     High         Low

            Low     Satisfied   Indifferent
Negatives
            High   Ambivalent   Distressed
     Uni-Dimensional Satisfaction
              20


              18
CSI-4 Means




              16


              14


              12
                   Dissatisfied   Indifferent   Ambivalent   Satisfied
      Negative Conflict
               80


               60
MCI-Conflict




               40


               20


               0
                    Dissatisfied   Indifferent   Ambivalent   Satisfied
  Negative Affect
        13

        12

        11
EPQ-N




        10

        9

        8
             Dissatisfied   Indifferent   Ambivalent   Satisfied
Study 6
 IRT Optimized Positive & Negative Scales
   Item Pools
      20 positive items
      20 negative items

   Large sample
      UG respondents


   Analyses
      EFA
      Redundancy
      IRT


 Precision / Power / Validity
Study 6 - Sample
 1,814 undergrad respondents
    Demographics
         19yo (2yrs)
         77% Female
         72% Caucasian
         Together 2.6yrs
         26% dissatisfied

    Close relationships
         54% romantic partners
         38% friends
         5% family members
         3% roommates

    Romantic relationships
       76% dating – exclusive
       21% dating – non-exclusive
Positive-Negative
Relationship Qualities
 New PN-RQ scales
  Best 4 & 8 items by IRT


    Positive Items            Negative Items

 Enjoyable*      Fun          Bad*      Discouraging*
  Pleasant      Full*       Empty*       Unpleasant
   Strong      Exciting    Miserable*       Weak
    Alive     Energizing    Lifeless         Dull
PN-RQ Correlations

         CSI-4   Pos-RQ   Neg-RQ


 CSI-4    1


Pos-RQ    .68      1


Neg-RQ   -.55     -.47      1
Information Provided
                                                 35



                                                 30
Information Provided




                                                 25



                                                 20



                                                 15



                                                 10


                            POS-SMD
                                                  5
                            POS-RQ (IRT)
                                                  0


                       -3     -2           -1         0      1      2   3
                                      Satisfaction (standardized)
NEG: Information Provided

                                                 50

                            NEG-SMD              45

                            NEG-RQ (IRT)         40
Information Provided




                                                 35


                                                 30


                                                 25


                                                 20


                                                 15


                                                 10


                                                  5


                                                  0


                       -3     -2        -1            0      1      2   3
                                      Satisfaction (standardized)
Power: Positive-Quality Groups
               2.5                                         POS-RQ
                                                           CSI-4
                2
Effect Sizes




               1.5


                1


               0.5

                0
                     2    3    4    5    6   7    8    9     10
                         Adjacent Pos-Qual Group Contrasts
 Power: Negative-Quality Groups
               2.5
                                                NEG-RQ
                 2                              invCSI-4
Effect Sizes




               1.5

                 1

               0.5

                 0

               -0.5
                      2   3    4    5    6     7     8     9   10
                          Adjacent Neg-Qual Group Contrasts
     Uni-Dimensional Satisfaction
              20
              19
              18
CSI-4 Means




              17
              16
              15
              14
              13
              12
                   Dissatisfied   Indifferent   Ambivalent   Satisfied
        Negative Conflict
                    7
Negative Conflict




                    6


                    5


                    4


                    3
                        Dissatisfied   Indifferent   Ambivalent   Satisfied
     Forgivingness
              36

              34
Forgiveness




              32

              30

              28

              26

              24
                   Dissatisfied   Indifferent   Ambivalent   Satisfied
PN-RQ

Power from Optimization
   More precise


Unique Information
   Ambivalent vs. Indifferent


NEXT: Responsiveness
Study 7
 PREP
     Psycho-educational workshop
     Speaker-Listener Technique

 CARE
     Psycho-educational workshop
     Acceptance based techniques (IBCT)

 Awareness
     Self-guided
     Semi-strutured

 No Treatment
Study 7
 173 Newlywed couples
     Engaged or married <6mo
     Screened for severe discord (MAT below 85)

 Demographics
     AGE 29
     Caucasian            H:   58%   W:   54%
     Latino/a             H:   17%   W:   23%
     Asian Am             H:   9%    W:   11%
     African Am           H:   5%    W:   5%

 Assessed
     MAT, PN-QIMS
     6 points over 3yrs
Slope-Intercept HLM Results

MAT
    Drops over time for Men


Negative Qualities
    ns


Positive Qualities
    Drops only in No Treatment*
    TX: Sig better slopes in Men
Bi-Dimensional View

Distinct individuals

Distinct treatment effects

Enhance
    Theories
    Clinical work
Topic 4: Implicit Assessment
 Limitations of Self-Report
     Insight
     Biases

 Limitations of Observational Coding
     Costly
     Evaluation apprehension
     Not all constructs observable

 Implicit assessment
     Indirect
     Inexpensive
     Unique insights
Previous Work
 Me/Not-Me task
        Aron, Aron, Tudor, & Nelson (1991), Aron & Fraley (1999), Slotter
        & Gardner (2009)

 Rxn Time on Evaluations
        Fincham, Garnier, Gano-Phillips, & Osborne (1995)

 Partner-focused IAT
        Zayas & Shoda (2005)
        Banse & Kowalick (2007)
        Scinta & Gable (2007)

 Self-focused IAT
        Dewitt, de Houwer, & Buysse (2008)

 Sequential priming task
        Scinta & Gable (2007)
Go/No-Go Association Task
Partner-GNAT
    Sort three types of words
      • Good
      • Bad
      • Partner


    Presented
      • One at a time
      • In random order


    Spacebar for targets
GNAT Stimuli
 Partner words
     First name
     Nick name
     Pet name / Distinguishing characteristic


               Good Words            Bad Words

 Studies            Peace                Death
                     Gift               Accident
 8 & 10            Vacation             Tragedy

                  Accepting            Attacking
 Study 9           Sharing              Nagging
                Understanding          Criticizing
GNAT Procedure

Procedure
    Obtain partner stimuli

    Sorting task
      • 16 practice trials: good as target
      • 16 practice trials: bad as target

      • 70 trials: partner + good as targets   Critical
      • 70 trials: partner + bad as targets    Trials

    Complete counterbalancing
GNAT

 Fast task (600msec)
     Accuracy
     D‟ index


 Proposed
     High performance on P-good
           Strong positive implicit attitude

     High performance on P-bad
           Strong negative implicit attitude
Studies 8 & 9: Samples
 Study 8
    122 online respondents
         39% married (for 3.3yrs)
         13% engaged (together for 2.7yrs)
         58% dating (for 2.4yrs)
      79% Caucasian
      87% Female
      43% provided follow-up data
      8 ended their relationships

 Study 9
    100 online respondents
         10% married (for 3.6yrs)
         12% engaged (together for 3.2yrs)
         77% dating (for 1.8yrs)
      77% Caucasian
      86% Female
      63% provided follow-up data
      11 ended their relationships
Method Variance
 P-good and P-bad
   r = .45
   Shared method variance
      Ability
      Effort
      Attention
      Comfort with computers

 Enter as pairs
      Simultaneous PVs
      Partial correlations
      Shared variance dropped


 Examine interaction
      Pos & Neg attitudes might interact
Study 8: Correlations

                  Initial Self Reports

           Relationship   Negative
                                     Neuroticism
           Satisfaction   Conflict


Partner-
Good          .14          -.05         .00

Partner-
Bad          -.21*         .12          .03
Studies 8 & 9: Analytic Strategy

 Discrete-time hazard modeling in HGLM
   LEVEL 1:
     Prob(Breakup) = P
     log[ P/(1-P) ] = p0 + p1(time) + e


   LEVEL 2:
     p0 = b00
     p1 = b10 + b11(relationship satisfaction)
            + b12 (hostile conflict)           Self-Report Controls
            + b13 (neuroticism)
            + b14 (partner-good)
            + b15 (partner-bad)                      Partner-GNAT
            + b16 (partner-good X partner-bad)       Performance
            + r1
 Study 8: Prediction of
 Relationship Breakup over 1 year
                                          B            SE          p           Odds
                                                                               Ratio
Intercept
  Intercept1                            -4.14*        0.36        .001         0.02
Predicting Breakups over 1 Year
  Intercept2                             2.21         0.51        .001         9.12
  Relationship Satisfaction             -0.05†        0.03        .065         0.95
  Neuroticism                            0.10         0.10        .342         1.10
  Hostile Conflict                       -0.07        0.07        .317         0.93
  Partner with Good (d´)                -1.75*        0.55        .002         0.17
  Partner with Bad (d´)                  0.38         0.93        .685         1.46
  P-Good X P-Bad                        -1.67*        0.83        .048         0.19


    NOTE: B = unstandardized beta; SE = standard error. † p < .10; * p < .05
Study 8: Probabilities of Breakup

  A)                    90          Low partner-good
                        80          High partner-good
                        70
   Percent of Breakup




                        60
                        50
                        40
                        30
                        20
                        10
                        0
                             Low partner-bad        High partner-bad
Study 9: Correlations

                  Initial Self Reports

           Relationship   Negative
                                     Neuroticism
           Satisfaction   Conflict


Partner-
Good          .06          -.13         .00

Partner-
Bad           -.03         .07          .03
 Study 9: Prediction of
 Relationship Breakup over 1 year

                                          B            SE           p          Odds
                                                                               Ratio
Intercept
  Intercept1                            -3.84*        0.34        .001         0.02
Predicting Breakups over 1 Year
  Intercept2                            2.69*         0.48        .001         14.75
  Relationship Satisfaction              -0.03        0.04        .363         0.97
  Neuroticism                            0.13         0.11        .232         1.14
  Hostile Conflict                       -0.03        0.04        .399         0.97
  Partner with Good (d‟)                -1.03*        0.37        .007         0.36
  Partner with Bad (d‟)                  0.86†        0.47        .074         2.35
  P-Good X P-Bad                         -0.22        0.61        .717         0.80

    NOTE: B = unstandardized beta; SE = standard error. † p < .10; * p < .05
Study 9: Probabilities of Breakup

  B)                   90          Low partner-good
                       80          High partner-good
                       70
  Percent of Breakup




                       60
                       50
                       40
                       30
                       20
                       10
                       0
                            Low partner-bad        High partner-bad
Studies 8 & 9: Summary
 Partner-GNAT
  Predicts Breakup over 1yr
     After controlling for SR scales
     Possible interaction


 Suggests
  Partner-GNAT provides unique information
     P‟s unable to report
     P‟s unwilling to report


 Next Step: Mechanism of action
Study 10
 Partner-GNAT
   Generic good or bad words
     Good stimuli: freedom, pleasure, gift
     Bad stimuli: death, accident, poverty


 Behavioral coding
   Two 10-minute Problem discussions
   Two 10-minute Social Support discussions

   Two teams of naïve coders

 Self-report data
Coding Process
 Two separate teams (5 and 7 coders)
       Weekly meetings

 Spouses coded in separate passes
       30sec intervals
       Global codes

 Counterbalancing
       Order of couples
       Order of spouses (within each interaction)
       Order of topics (H vs. W)

 Rated 15-18 dimensions

 All coders coded all tapes
       Codes averaged within coders & interactions
       Codes averaged across coders
       Created composite codes
Composite Codes

Support Behavior/Affect
    Emotional Support
    Negative Behavior


Conflict Behavior/Affect
    Empathic Listening
    Affection
    Negative Behavior
Study 10: Sample
 57 couples
      48% engaged to be married (in 4.8mo)
      52% married (for 3.7mo)

 Relationships
      Together 3.3yrs
      Highly satisfied (avg. CSI = 141)
      81% premarital cohabitation
      93% living together at T0
      14% had children at T0

 Demographics
      Age 28yo
      91% Caucasian
      53k joint income
      9% ≤ HS education
Study 10: Analytic Strategy
 Actor-Partner Interdependence Modeling
  in HLM

 Modeling trajectories over time
    Two level model
       • Level 1 – Individual differences
           GNAT indices
           Coded behavior
           Initial self-report

       • Level 2 – Dyadic variables
           Relationship length
           Number of children
Male Emotional Support
(during support interaction)


   A)                        0.5           Female Low Part-Bad
                             0.4           Female High Part-Bad
                             0.3
    Male Emotional Support




                             0.2
                             0.1
                               0
                             -0.1
                             -0.2
                             -0.3
                             -0.4
                             -0.5
                                    Female Low Part-Good Female High Part-Good
Male Negative Behavior
(during support interaction)


   B)                       0.5           Female Low Part-Bad
                            0.4           Female High Part-Bad
                            0.3
    Male Negative Support




                            0.2
                            0.1
                              0
                            -0.1
                            -0.2
                            -0.3
                            -0.4
                            -0.5
                                   Female Low Part-Good Female High Part-Good
Female Empathic Listening
(during conflict interaction)


  C)                0.5           Female Low Part-Bad
                    0.4           Female High Part-Bad
                    0.3
                    0.2
   Female Empahty




                    0.1
                      0
                    -0.1
                    -0.2
                    -0.3
                    -0.4
                    -0.5
                           Female Low Part-Good Female High Part-Good
Female Affection
(during conflict interaction)


   E)                     0.5           Female Low Part-Bad
                          0.4           Female High Part-Bad
                          0.3
    Female Pleasantness




                          0.2
      Female Affection




                          0.1
                            0
                          -0.1
                          -0.2
                          -0.3
                          -0.4
                          -0.5
                                 Female Low Part-Good Female High Part-Good
Male Negative Behavior
(during conflict interaction)


    D)                        0.5            Male Low Part-Bad
                              0.4            Male High Part-Bad
                              0.3
     Male Negative Behavior




                              0.2
        Male Hostility




                              0.1
                                0
                              -0.1
                              -0.2
                              -0.3
                              -0.4
                              -0.5
                                     Male Low Part-Good    Male High Part-Good
Study 10: Summary

Partner-GNAT
   Linked to own behavior
   Linked to partner‟s behavior
   Across domains

      Might shape each other


Tailoring GNAT
   Implicit assessment of attachment?
Self Report Attachment Scales

ECR-R
   Attachment Anxiety
     “I worry a lot about my relationship”
   Attachment Avoidance
     “I find it difficult to allow myself to depend on
       romantic partners”

   Difficult to disentangle
     • Attachment
     • Preoccupied / Dismissive Behaviors

   Requires insight / honesty
Study 11
 Standard Battery of SR scales

 Implicit Attachment
     Partner-GNAT*
     Self-GNAT*
     *New Valence Categories
      • Relationally Worthy
      • Relationally Worthless


     Hypotheses
      • Partner-GNAT*  internal working model of others
      • Self-GNAT*  internal working model of self
GNAT Stimuli
 Partner words                 Self words
     First name                    First name
     Nick name                     Last name
     Pet name /                    Nick name /
      characteristic                 characteristic

                Relationally          Relationally
                  Worthy               Worthless
                         Loved             Inferior
                         Liked            Rejected
 Valence               Agreeable        Disagreeable
                       Accepted           Pathetic
 Stimuli               Cherished         Unwanted
                        Valued           Abandoned
                        Adored          Insignificant
Study 11: Sample
 Recruitment underway

 First 48 couples
       79% committed dating relationships (1.6yrs)
       4% engaged (2.9yrs)
       17% married (4.5yrs)

 Relationships
       Quite satisfied (CSI-16 = 70)
       Dissatisfied
           • 6% of married
           • 9% of dating


 Demographics
         Age 24yo
         76% Caucasian
         37k joint income
         9% ≤ HS education
Study 11: Correlations

                  Initial Self Reports

           Relationship   Attachment   Attachment
           Satisfaction     Anxiety    Avoidance


Partner-
Good         .27*           .01          -.23†

Partner-
Bad           -.14          .06          .17
Study 11: Correlations

                  Initial Self Reports

           Relationship   Attachment   Attachment
           Satisfaction     Anxiety    Avoidance


Self-
              .04           .07          .03
Good

Self-Bad      .01           -.08         -.03
Future Directions
 Unique information
       Beyond SR
       Clinically useful?

 Shapes behavior
       Longitudinal mediation?
       Change over time?

 Can be Tailored
       Attachment?

 Alternate Targets
       Family
       Friends
       Behaviors

 Moderators
       Mindfulness
       Assimilation of Partner into Self-Concept
          • IOS
          • RISC
Topic 5: Attention / Effort
 Inattention
      Adds error / noise
      Reduces power


 Quantifying
      Large Clinical Inventories (e.g., PAI)
        • Infrequency
        • Inconsistency

      Experimental Research
        • Instruction reading (IMC; Oppenheimer, 2009)

      Survey Research
        • Unknown
Study 12

 Quantifying Inattention

     Behavioral Measures
       • 7 directed responses
       • 20 pronoun task
       • 2min video


     Self-Report
       •   Inattentive
       •   Patterned
       •   Rushed
       •   Instruction skipping
Study 12 - Sample

 575 online respondents
     54% Mturk.com
     13% online forums
     33% UG psychology students


 Demographics
     29yo (12yrs)
     70% Female
     77% Caucasian
     21% ≤ High School
     30% ≤ $30k / year
Behavioral Inattention
                        60
                                Any Errors
% Showing Inattention




                        50      >50% Errors


                        40

                        30

                        20

                        10

                         0
                             Mistakes on      Not Watching   Mistakes on
                             Directed Q's      Full Video     Pronouns
 Self-Reported Inattention
                        90
                                           Ever
                        80                 >50% of the time
% Showing Inattention




                        70
                        60
                        50
                        40
                        30
                        20
                        10
                         0
                             Inattentive        Patterned     Rushed     Skipping
                                                                       Instructions
Distinct from Desirability
                             Self       Impression
Inattention Measure        Deception   Management
Mistakes on directed Q‟s     -.16         -.16
Time watching video          .11          .12
Mistakes on pronouns         -.06         -.07
SR: Inattentive              -.30         -.34
SR: Patterned                -.15         -.20
SR: Rushed                   -.28         -.24
SR: Skip Instructions        -.15         -.18
Screening for Inattention

Developing ARS
   Item pool
     • Infrequent items
     • Inconsistent pairs


   3 large online samples

   Ability to discriminate
     • P responses
     • Random data
     • Random responders
Studies 13 through 15
 Study 13
   1195 online respondents
      85% female
      77% Caucasian
      26yo (SD = 8.4)

 Study 14
   1878 online respondents
      91% female
      85% Caucasian
      28yo (SD = 7.1)

 Study 15
   547 online respondents
      74% female
      72% Caucasian
      20yo (SD = 1.3)
Final ARS scales

Two scales
   11 infrequency items
     • I enjoy the music of Marlene Sandersfield
     • I look forward to my time off

   11 inconsistency item pairs
     • I am an active person
     • I have an active lifestyle


Agreement with PAI
   Study 14
     • Continuous:    r‟s = .64 and .83
     • Categorical:   kappa = .72
     Ability to Detect Inattention
                                        ARS Hit Rate
                                        ARS False Positives
               1
                                        PAI Hit Rate
              0.9
                                        PAI False Positives
              0.8
Proportions




              0.7
              0.6
              0.5
              0.4
              0.3
              0.2
              0.1
               0
                    Random Data   Random Responders
Convergent Validity

Study 12 indices

ARS inattentive respondents
 Higher on inattention indices?
   Behavioral Markers
   Self-Report


 Comparable regression results?
ARS Inattentive P‟s

                                                   Cohen‟s
Index                           Effect               D
Directed Q‟s               3 more mistakes / 7      -1.43**

Video Watching             40 seconds less / 120     .95**

Pronouns                   7 more mistakes / 20     -1.01**

SR Inattentive               1.0 pts higher / 6     -.91**

SR Patterned                 1.3 pts higher / 7     -1.11**

SR Rushed                    0.7 pts higher / 7     -.60**

SR Skipping Instructions     0.8 pts higher / 7     -.53**
 ARS Inattentive P‟s
 Robins et al. (2001)
       Big 5  Self Esteem
       R2 = .34**
       3 sig coeffs
                                                   .45
 Attentive P‟s
                        Variance Predicted in DV
                                                   .40
       N = 621
                                                   .35
       R2 = .41**
       3 sig coeffs                               .30
                                                   .25
 Inattentive P‟s
                                                   .20
       N = 55
       R2 = .08 ns                                .15
       no sig coeffs                              .10
                                                                                   ns
                                                   .05
                                                   .00
                                                         Ronbins   Attentive   Inattentive
ARS Convergent Validity

ARS inattentive respondents

 Higher inattention
   Behavioral Markers
   Self-Report


 Adding noise
   Lowering power
Study 16
 Reading instructions?

 ARS vs. IMC
   Oppenheimer (2009)
   IMC: Instructional Manipulation Check

     Single paragraph / item
     Eliminates 20-40% of P‟s

     Enhances power
       • Paragraph manipulations
           Sports ticket
           Can of pop
                                       Sports Participation


Most modern theories of decision making recognize the fact that decisions do not take place
in a vacuum. Individual preferences and knowledge, along with situational variables can
greatly impact the decision process. In order to facilitate our research on decision making we
are interested in knowing certain factors about you, the decision maker. Specifically, we are
interested in whether you actually take the time to read the directions; if not, then some of our
manipulations that rely on changes in the instructions will be ineffective. So, in order to
demonstrate that you have read the instructions, please ignore the sports items below, as
well as the continue button. Instead, simply click on the title at the top of this screen (i.e.,
“Sports Participation”) to proceed to the next screen. Thank you very much.


                  Which of these activities do you engage in regularly?
                                      (click all that apply)


        skiing            soccer          snowboarding            running            hockey

       football         swimming            basketball             tennis            cycling


                                             Continue
                                       Sports Participation


Most modern theories of decision making recognize the fact that decisions do not take place
in a vacuum. Individual preferences and knowledge, along with situational variables can
greatly impact the decision process. In order to facilitate our research on decision making we
are interested in knowing certain factors about you, the decision maker. Specifically, we are
interested in whether you actually take the time to read the directions; if not, then some of our
manipulations that rely on changes in the instructions will be ineffective. So, in order to
demonstrate that you have read the instructions, please ignore the
sports items below, as well as the continue button. Instead,
simply click on the title at the top of this screen (i.e., “Sports
Participation”) to proceed to the next screen. Thank you very much.

                  Which of these activities do you engage in regularly?
                                      (click all that apply)

        skiing            soccer          snowboarding            running            hockey

       football         swimming            basketball             tennis            cycling


                                             Continue
Study 16 - Sample

 652 online respondents
     60% Mturk.com
     40% UG psychology students


 Demographics
     28yo (11.5yrs)
     70% Female
     74% Caucasian
     27% ≤ High School
     30% ≤ $30k / year
ARS – IMC Agreement

                       IMC
 Kappa = .16                       Total
                 OK      Exclude

            OK   435         138   91%
 ARS
       Exclude   24          30    9%

    Total        73%         27%   627
      Inattention Indices
                           1.2                                                                                           ARS Effect Sizes
Cohen's D between Groups




                                                                                                                         IMC Effect Sizes
                             1

                           0.8

                           0.6

                           0.4

                           0.2

                             0
                                        's                  ns                   e   d            ve         ed            ing               ns
                                     dQ                  ou                  ern              enti       u sh          tch               tio
                                  cte                 ron              Pa
                                                                          tt               tt
                                                                                                       RR           Wa                 uc
                              e                   P                                     Ina                                         str
                           Dir                n                  S   R               SR
                                                                                                     S
                                                                                                               ide
                                                                                                                  o              In
                                       e   so                                                                V              k ip
                               s    tak                                                                              S  RS
                             Mi
               Sunk Cost Task (Thaler, 1985)
            Imagine that your favorite football team is playing an important
            game that you
                       [have paid handsomely for.]
                       [have received from a friend.]
            However, on the day of the game, it happens to be freezing cold.
            What do you do?
                                                               Full Sample
                      0.5
                                                                Screened - Good
Manipulation Effect




                      0.4                                       Screened - Bad


                      0.3
                                             ns
                      0.2

                      0.1

                       0
                                    ARS                 IMC
        Soda Pricing Task (Thaler, 1985)
     You are on the beach on a hot day. For the last hour you have
     been thinking about how much you would enjoy an ice cold can of
     soda. Your companion needs to make a phone call and offers to
     bring back a soda from the only nearby place where drinks are
     sold, which happens to be a
                            [run-down grocery store.]   [fancy resort.]
     Your companion asks how much you are willing to pay for the soda
     and will only buy it if it is below the price you state. How much
     are you willing to pay?
                                                             Full Sample
       0.7
                                                             Screened - Good
Manipulation Effect




       0.6
                                                             Screened - Bad
       0.5                    ns
                      0.4
                      0.3                                      ns
                      0.2
                      0.1
                       0
                                ARS                     IMC
Inattention Summary

Inattention
 Skipping instructions
   As high as 20-40%


 Skimming items
   5-10%


 ARS effectively screens
   Enhances power
Summary
 RELATIONSHIP QUALITY
   T1: IRT Optimization
      Study 1

   T2: Responsiveness to Change
      Studies 2-5

   T3: Bi-Dimensional View
      Studies 6-7

   T4: Implicit Measures
      Studies 8-10

 ATTENTION
   T5: Screening for Error Variance
      Studies 11-15
Limitations

Online samples

Largely female

Largely Caucasian

Lacking behavioral criteria
Thank You.
Existing Measures

 Measure    Citations   Cit./Year

 DAS (32)    2,237        77.1

 MAT (16)    1,476        32.1

 QMI (6)      218         9.9

 KMS (3)      179         9.4

 RAS (7)      150         8.8
Criterion Validity

 DAS Distress groups
   Current gold-standard

   DAS score < 97.5
   1027 DAS distressed P‟s

   ROC‟s to identify CSI cut scores
   Identified CSI distressed P‟s

   91% agreement w/ DAS
Precision: CSI-16 vs. MAT
Power: CSI-16 vs. MAT
                            2.5
 .


                                                                                   MAT
                             2
 Effect Sizes (Cohen's d)




                                                                                   CSI(16)

                            1.5


                             1


                            0.5


                             0
                                  2   3   4   5   6   7   8   9 10 11 12 13 14 15 16 17 18 19 20

                                          Adjacent Satisfaction Group Contrasts
Studies 2-4: Demographics
 SAMPLE
   N = 2,056 initial respondents
   N = 968 (47%) respondents with longitudinal data

 AGE
   M = 27.7yo (9.3yrs)

 GENDER
   71% Female
   29% Male

 RACE
     83% Caucasian
     5% Asian
     4% African American
     4% Latino

 SES
   10% High school diploma or less
   25K avg yearly income
Studies 2-4: Relationships
 Relationship Types
     37% Married: 7.9 yrs (7.9 yrs)
     13% Engaged: 3.2 yrs (2.4 yrs)
     50% Dating: 1.8 yrs (1.9 yrs)

 Relationship Satisfaction (MAT)
     Married: 108 (32)
     Engaged: 122 (24)
     Dating: 116 (24)


 Dissatisfied Respondents
     24% (n = 487)
Study 2 - Sample
 N=596 initial respondents
         27yo (SD = 10yrs)
         77% Female
         84% Caucasian
         8% ≤ High school
         22K income
         30% married, 14% engaged, 55% dating
         16% dissatisfied

 372 provided email (62%)
 267 completed follow ups (71%)

 NS differences on
       Length of relationship      Relationship satisfaction
       Age                         Education
       Gender
Study 3 - Sample
 N=398 initial respondents
         26yo (SD = 8yrs)
         86% Female
         80% Caucasian
         9% ≤ High school
         20K income
         30% married, 12% engaged, 58% dating
         24% dissatisfied

 252 provided email (63%)
 156 completed follow ups (62%)

 NS differences on
       Length of relationship      Relationship satisfaction
       Gender                      Ethnicity
Study 4 - Sample
 N=1062 initial respondents
         29yo (SD = 9yrs)
         79% Female
         83% Caucasian
         11% ≤ High school
         29K income
         44% married, 13% engaged, 43% dating
         28% dissatisfied

 746 provided email (70%)
 545 completed follow ups (73%)

 NS differences on
       Length of relationship
       Age                         Ethnicity
RCI and MDC95 Equations
 RCI
     SERM = √2*SD2(1 – rxx)

     SERM = √2*MSE

     RCI = (x2 – x1) / SERM

     If RCI > 1.96
         Sig individual change


 MDC95
     Solve RCI eq for (x2 – x1)

     MDC95 = 1.96*SERM
Estimating Noise &
Reliable Individual Change
Guyatt, Walter & Norman (1987); Jacobson & Truax (1991)



NOISE = SEM =                           2*MSE
                 (Standard Error        (MSE = Mean Squared Error from a Repeated
                   of Repeated          Measures ANOVA on the T0, F1, & F2 scores of
                  Measurement)          „No Change‟ individuals)




RELIABLE   Signal                                Dx
         =                                =                  = 1.98
CHANGE     Noise                               SEM

Dx = Minimal Detectable Change (MDC95)
 (smallest change in scores needed in an individual to suggest reliable change)
Reliable Individual Change
         Range   SERM   MCD95   SD units

CSI-32   0-161   8.0    15.7    .49 SD‟s

DAS      0-151   5.8    11.4    .65 SD‟s

CSI-16   0-81    4.8     9.4    .55 SD‟s
MAT      0-158   10.7   21.0    .76 SD‟s

CSI-4    0-21    1.6     2.8    .70 SD‟s
Estimating Power
for Detecting Perceived Change
Guyatt, Walter & Norman (1987)




                                   Sensitivity to
                                 Perceived Change
                Signal       (difference in avg change scores between
                                 adjacent perceived change groups)
POWER =                  =
(Effect size)
                Noise                       SEM
                                       (Standard Error of
                                     Repeated Measurement)
Sensitivity to Perceived Change
CSI-16 vs. MAT
                          15
                                   MAT Change Scores

                                   CSI-16 Change Scores
  .




                          10
  Average Change Scores




                           5



                           0



                           -5



                          -10
                                  Minimal           No         Minimal       Significant
                                Deterioration     Change     Improvement    Improvement
                          -15

                                      Self-Reported Global Change Over 1-Week
Responsiveness Model

Level 1 – repeated assessments
  X2 – X1 = p0 + p1*(global change) +
            p2*(deterioration) + e

Level 2 – individuals
  p0 = b00
  p1 = b10 + b11*(T0 rel sat) + b12*(male) + r1
  p2 = b20 + b21*(T0 rel sat)
Differences by Gender
 Scales showed slightly smaller effect
  sizes in men
                              0.1
  Reduction in Effect Sizes




                                0

                              -0.1
                                                        *
                              -0.2                                    *
                                       *        *
                              -0.3

                              -0.4

                              -0.5
                                     CSI-32   CSI-16   CSI-4   DAS   MAT
Study 5
 Responsiveness to Mild Intervention
   Reissman, Aron, & Bergen (1993)
      Pos. activities over 10wks
      Fun/Exciting  Enhanced satisfaction

   158 randomly assigned to:
      Control
      Fun / Exciting Activities Feedback

   2wk follow up
      25 Fun / Exciting behaviors
      Satisfaction

 Scales
      CSI-32       DAS-32           MAT-15   QMI
      SMD          RAS              KMS      PN-RQ
  Fun/Exciting Activities Feedback
 Background
    There is a large body of research supporting the importance of
    fun in relationships.

    Unfortunately, many couples slowly forget to make time to do
    fun things together the longer they are together.

 Request
    As part of this study, we would like you and your partner to
    make an effort to have more fun with each other over the next 2
    weeks.

    Specifically, we would like you to try to do some fun activities
    that get you out of the house and/or out of your normal routines .

      These activities should be fun and exciting for both of you.

      These activities should also involve things that you can do together
       (like going to dinner) rather than more solitary activities (like
       reading).
Fun/Exciting Activities Feedback
Based on your responses, here is a list of
  activities you rated as most fun and exciting:
                                               How fun/exciting   How often you
Fun activity                                     you rated it     currently do it
Engaging in intimate sexual activity             Extremely        7x in 2 weeks
Going to a movie                                 Extremely        1x in 2 weeks
Playing sports                                   Extremely        2x in 2 weeks
Going camping                                       Very          0x in 2 weeks
Spending time with friends                          Very          6x in 2 weeks
Going on a hike                                     Very          3x in 2 weeks
Going on a picnic                                   Very          0x in 2 weeks
Attending community events (e.g., festivals)        Very          0x in 2 weeks
Going to the beach or lake                          Very          1x in 2 weeks
Going to a restaurant                           Somewhat          4x in 2 weeks
Study 5 - Sample
 158 initial respondents (first 3 ½ days of recruitment)

         30yo (SD = 11yrs)
         74% Female
         83% Caucasian
         18% ≤ High school
         53K income
         39% married, 10% engaged, 51% dating
         18% dissatisfied


 xxx completed follow ups (73%)

 NS differences on
       Length of relationship
       Age                         Ethnicity
PREP
Prevention and Relationship Enhancement Program

 14 hour workshop over 4 sessions
      One weekend day
      Three weeknights

 Communication Skill Focus
      Speaker-Listener Technique
      Problem-Solving skills
      Time Outs
      Building Positive Behaviors

 Goals
      prevent conflict escalation (improve resolution)
      enhance/protect positive aspects of relationship
CARE
Compassionate and Accepting Relationships through
Empathy
 14 hour workshop over 4 sessions
      One weekend day
      Three weeknights

 3 Acceptance based skill modules
      Support skills
      Conflict skills
      Forgiveness skills

 Goal: increase understanding/acceptance
      To buffer „rough spots‟
      To smooth out conflict discussions
      To protect positives
Awareness Condition

“Movie” Treatment
   List of relationship-focused movies
   Watched 5 movies together
   40 min guided discussion after each
   First movie in a group setting (at UCLA)

Yoked Control Group
   Equivalent time together
   Equivalent time discussing relationship
   No active psycho-educational component
Treatment Conditions

               CARE   PREP Aware No Tx

N assigned      53     45   45    52

Initial
satisfaction
               121    117   125   117

Tx Dropout      9      10   21     --
Hypotheses
 All treatment conditions would show
  better marital quality than no tx


 CARE and PREP would show better
  marital quality than the minimal tx


 CARE would demonstrate comparable
  tx effects to PREP
Longitudinal Assessments
 T0 – 1-2 months prior to workshop

 T1 – start of workshop

 T2 – 6 months after workshop

 T3 – 1 year

 T4 – 2 years

 T5 – 3 years
 Previous Work
 Me/Not-Me task
        Implicit Closeness  3mo shift in SR closeness
         Aron, Aron, Tudor, & Nelson (1991), Aron & Fraley (1999), Slotter & Gardner (2009)

 Rxn Time on Evaluations
        High Accessibility  Stronger effects among SR scales
                Fincham, Garnier, Gano-Phillips, & Osborne (1995)

 Partner-focused IAT
        Pos Implicit Atttitude  Secure attachment (& lower attch avoidance)
                Zayas & Shoda (2005)
        Pos Implicit Attitude  Criterion validity (separating groups)
                Banse & Kowalick (2007)
        Pos Implicit Attitude  Current relationship satisfaction
                Scinta & Gable (2007)

 Self-focused IAT
        Implicit Relational Worthiness  lower attachment anxiety & preoccupation
        Implicit Relational Anxiety  preoccupied attachment
                 Dewitt, de Houwer, & Buysse (2008)

 Sequential priming task
        Neg Implicit Attitude  3mo shift in SR satisfaction
               Scinta & Gable (2007)
Composite Codes
 Support Behavior/Affect
      Emotional Support
        • Understanding             Reassuring
        • Responsive                Relieving blame

      Negative Behavior
        • Frustration               Hostility
        • Disagreeing               Blaming
        • Tension

 Conflict Behavior/Affect
      Empathic Listening
        • Tuned into P‟s feelings   Supportive
        • Validating                Interested / Curious

      Affection
        • Warm / Affectionate       Humorous / Playful

      Negative Behavior
        • Hostile                   Frustrated
        • Angry                     Blaming
Study 10: Analytic Strategy
Actor-Partner modeling in HLM
LEVEL 1:
Relationship Behavior =
 + p1(male X own satisfaction) + p2(male X spouse’s satisfaction)
 + p3(female X own satisfaction) + p4(female X spouse’s satisfaction)   Self-Report Controls
 + similar sets of APIM terms for hostile conflict & neuroticism

 + p13(male X own partner-good) + p14(male X spouse’s partner-good)
 + p15(female X own partner-good) + p16(female X spouse’s partner-good)    Partner-GNAT
 + p17(male X own partner-bad) + p18(male X spouse’s partner-bad)          Performance
 + p19(female X own partner-bad) + p20(female X spouse’s partner-bad)

 + APIM terms for interactions between partner-good and partner bad
 +e

LEVEL 2:
 p1 = b 10
 p2 = b20
 p3 = b 30 + b 31(rel length) + b 32(# of kids) + r3
 p4 = b 40
  (similar equations for remaining lvl2 effects)
           POS-RQ Distinct Change Groups
                         4

                         2
.




                         0
Average Change Scores




                         -2

                         -4

                         -6

                         -8

                        -10

                        -12
                               Substantial        Mild          No         Mild       Substantial
                              Deterioration   Deterioration   Change   Improvement   Improvement
         NEG-RQ Distinct Change Groups
                        16

                        14
.




                        12
Average Change Scores




                        10

                        8

                        6

                        4

                        2

                        0

                        -2
                              Substantial        Mild          No         Mild       Substantial
                             Deterioration   Deterioration   Change   Improvement   Improvement
                    POS-RQ Distinct Change Groups
                         4
                              CSI-4 Change Scores (with confidence intervals)
                         2
.




                              POS-RQ Change

                         0
Average Change Scores




                         -2

                         -4

                         -6

                         -8

                        -10
                                    Minimal            No             Minimal      Significant
                        -12       Deterioratio       Change         Improvement   Improvement
                                       n
                                         Self-Reported Global Change

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:17
posted:4/9/2011
language:English
pages:156