Approaches to Well-being
How people evaluate their lives
Day Reconstruction Method
Norbert Schwarz, Michigan Daniel Kahneman, Princeton Alan Krueger, Princeton David Schkade, UC San Diego Arthur Stone, SUNY Stony Brook
Funded by Hewlett Foundation and NIA
?C ??How satisfied are you with your life as a whole??? ?C Basis of most of what we know about well- being ?C Answers ? highly context dependent ? poorly related to objective conditions
Approaches to Well-being
How people live and experience their lives
?C Time use & experience ? How much time spent doing what? ? How does it feel? ?C Pioneered by Tom Juster in 70??s ? ??An important ingredient in the production and distribution of well-being is the set of satisfactions generated by activities themselves" (Juster, 1985, p. 333)
Day Reconstruction Method (DRM)
Builds on Juster??s approach ? Takes advantage of last 20 years of cognitive research into reports of emotions and experience
?C Juster: global enjoyment ratings for classes of activities ("dislike very much" to "enjoy a great deal"). ?C DRM: tailored affect profiles for specific episodes of yesterday, which are
??reinstantiated?? in memory in some detail
Step 1 ? R??s reconstruct ??yesterday?? on a private form
?C List major meaningful episodes from getting up to going to bed, noting some details ?C Can insert new episodes as later ones prompt memories of earlier ones ?C The interlinked structure facilitates rich ?? reinstantiation?? of the last day.
Step 2 ? R??s complete a form for each episode, reporting
?C Where ?C Who with ?C What ?C How they felt during the episode
R??s indicate approximate times (to estimate duration) ? R??s keep this form, which serves as their own memory prompt for later questions.
Specifics tailored to the purpose of the study
DRM Episode Questions
Where were you? (check one) __ At home __ At work __ Somewhere else
DRM Episode Questions
What were you doing? (check all that apply)
__ commuting __ shopping __ doing housework __ eating
__ working __ preparing food __ taking care of your children __ pray/worship/meditate __ watching TV __ computer/internet/email __ on the phone __ exercising
Who were you interacting with (including on the phone, etc)? Check all that apply.
__ no one __ spouse/significant other __ friends __ co-workers __ clients/ students __ my children __ parents/relatives __ boss __ other people
__ socializing __ nap/resting __ relaxing __ intimate relations
__ other (please specify)
How did you feel during this episode? Not at all Impatient for it to end . . . . . . . . . Happy . . . . . . . . . . . . . . . . . . . . . . Frustrated/annoyed . . . . . . . . . . .
Depressed/blue . . . . . . . . . . . . . . Competent/capable . . . . . . . . . . . Hassled/pushed around . . . . . . . Warm/friendly . . . . . . . . . . . . . . . Angry/hostile . . . . . . . . . . . . . . . .
Worried/anxious . . . . . . . . . . . . . Enjoying
myself . . . . . . . . . . . . . . Criticized/put down . . . . . . . . . . . Tired . . . . . . . . . . . . . . . . . . . . . . . 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 5 5 5 Very much 6 6 6 6 6 6 6 6 6 6 6 6
Initially tested with diverse convenience samples
?C Full day takes about 1 hour
Currently adapted for representative surveys conducted by Gallup and WHO
?C Different parts of the day for different R??s ?C Detailed reports on random subset of episodes ?C 10-15 minutes
Feasible task, independent of R??s education
Feelings as a function of episode characteristics ? Duration weighted to characterize day ? Provides rich descriptions of the subjective experience related to different activities, situations, interaction partners, etc. ? Can be related to background variables
?C e.g., SES, work conditions, personality
Gold standard for measuring momentary experience is ESM ? Against ESM data from the same person for the same day ? Against ESM profiles from other persons for the same type of episode ? Non-obvious ESM patterns
?C Can we reproduce non-obvious patterns observed in ESM? ?C E.g.,
People predict a linear increase in tiredness over the course of the day. ? ESM shows
?C you start out tired ?C tiredness drops over the morning ?C is lowest at lunch ?C increases after that
??Tired?? in DRM and ESM
DRM replicates this pattern, which R??s can??t report introspectively (Kahneman et al., Science, 2004)
People do not predict diurnal patterns for negative affect, which they see as situation driven. ? ESM shows diurnal pattern ? DRM replicates it ?C without introspective insight.
Negative Affect in DRM and ESM
More examples in Kahneman et al., Science, 2004 ? DRM replicates non-obvious ESM findings ? ESM and DRM profiles show good correspondence
Some Illustrative Observations
Activities & Affect
Activity intimate relations relaxing socializing eating pray/meditate watching TV preparing food shopping childcare internet/email doing housework working commuting Net Affect 4.26 3.71 3.59 3.35 3.26 3.17 2.90 2.81 2.55 2.50 2.39 2.25 2.10 % time 1.6 14.3 14.8 14.7 2.8 15.2 7.7 3.0 7.9 11.8 14.7 40.5 10.5 Net affect = positive minus negative (possible range -6/6) N = 909 employed women 24% African-American 22% Hispanic 49% White Household income $54.7k
Features More Correlated with Job Satisfaction Than Job Affect (DRM)
Satisfaction > Affect Current job compared to previous Low status job (-) Overqualified for my job (-) Requires specialized education/training Can make decisions that help people Household income Excellent benefits .31 .23 .20 .20 .20 .20 .17 .17 .11 .12 .09 .11 .04 .09
Features Less Correlated with Job Satisfaction Than Job Affect (DRM)
Satisfaction < Affect Affective Disposition Constant pressure to work fast ( -) Had trouble staying awake, etc (-) Constant attention to avoid mistakes (-) Can chat with other workers while on job Exposed to offensive noise (-) .25 .19 .08 .04 .04 .04 .42 .28 .18 .14 .10 .10
Global vs. Episodic:
How Much Do People Enjoy Their Kids?
Juster and colleagues, 1975 survey:
?C 28 activities rated from "dislike very much" (0) to "enjoy a great deal" (10).
Marital Satisfaction (4 studies)
Activities with one's children top the list (ranks 1-4) ? Grocery shopping and cleaning the house at the bottom (ranks 27 & 28) ? But??.
Top-down vs. Bottom-up
How much do you enjoy ??(10 = pos)
Global and episodic reports result in different rankings ? Likely key contributors:
?C Organization of memory ?C Memory bias ?C Social desirability
Positive minus negative Note: Numbers not comparable ! Socializing Eating 8.6 8.3 7.8 5.4 4.7 Watching TV Preparing food Grocery shopping Taking care of your children Housework 2.6 2.5 3.6 3.3 3.2 2.9 2.8
Socializing Taking care of your children Eating Watching TV Grocery Shopping Housework
Organization of Memory
Activities are encoded and represented by their key goals and features
?C Cooking while the kids are present is still ??cooking??
Global reports are based on memorable episodes
?C Memorable episodes are rarely typical ?Cthey are more extreme.
What??s recalled for ??kids?? are kid-focused activities
?C Reading stories, watching movies??
Activities that are not kid-focused are missed
?C But make up most of the time spent with kids
The DRM??s episodic reports are based on the specific episodes of yesterday
?C Avoids memory bias
Negative global judgments are less acceptable than negative episodes
Global and episodic reports paint different pictures ? Assessments should be based on specific and recent episodes ? DRM does so and combines affect reports with time use data, providing a rich characterization of people??s daily experience
?C ??I don??t like being with my kids?? (global) ?C ??My kids were a pain last night?? (specific episode)
Event Reconstruction Method (ERM)
Does not focus on yesterday ? Asks for the last instance of an activity
?C E.g., last time you went out for dinner
Reproduces the DRM affect profiles
?C provided the episode is recent
Does not provide time use information
?C Which can be taken from other sources
Has lower respondent burden and can capture less frequent activities