Journa o Personalit an Socia Psycholog  Vol  No  Copyrigh  b th America i   Th Structur o Psychologica WellBein Revisite Caro D Ryf an Core Le M Keye Universit o WisconsinMadiso theoretica mode o psy

Journa o Personalit an Socia Psycholog Vol No Copyrigh b th America i Th Structur o Psychologica WellBein Revisite Caro D Ryf an Core Le M Keye Universit o WisconsinMadiso theoretica mode o psy - Description

Comparison wit othe frequentl use indicator positiv an negativ affect lif satisfaction demonstrate tha th latte neglec ke aspect o positiv functionin emphasize i theorie o healt an wellbeing Fo mor tha 2 years th stud o psychologica wellbein ha bee ID: 30513 Download Pdf

163K - views

Journa o Personalit an Socia Psycholog Vol No Copyrigh b th America i Th Structur o Psychologica WellBein Revisite Caro D Ryf an Core Le M Keye Universit o WisconsinMadiso theoretica mode o psy

Comparison wit othe frequentl use indicator positiv an negativ affect lif satisfaction demonstrate tha th latte neglec ke aspect o positiv functionin emphasize i theorie o healt an wellbeing Fo mor tha 2 years th stud o psychologica wellbein ha bee

Similar presentations

Download Pdf

Journa o Personalit an Socia Psycholog Vol No Copyrigh b th America i Th Structur o Psychologica WellBein Revisite Caro D Ryf an Core Le M Keye Universit o WisconsinMadiso theoretica mode o psy

Download Pdf - The PPT/PDF document "Journa o Personalit an Socia Psycholog ..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.

Presentation on theme: "Journa o Personalit an Socia Psycholog Vol No Copyrigh b th America i Th Structur o Psychologica WellBein Revisite Caro D Ryf an Core Le M Keye Universit o WisconsinMadiso theoretica mode o psy"— Presentation transcript:

Page 1
Journa o Personalit an Socia Psycholog 1995 Vol 69 No 4,719-72 Copyrigh 199 b th America ^i^ 01 ^*^!^] Th Structur o Psychologica Well-Bein Revisite Caro D Ryf an Core Le M Keye Universit o Wisconsin—Madiso theoretica mode o psychologica well-bein tha encompasse 6 distinc dimension o wellnes (Autonomy Environmenta Mastery Persona Growth Positiv Relation Wit Others Purpos i Life Self-Acceptance wa teste wit dat fro a nationall representativ sampl o adult (i - 1,108) age 2 an older wh participate i telephon interviews Confirmator facto analyse provide suppor fo th propose

6-facto model wit a singl second-orde supe factor Th mode wa superio i fit ove single-facto an othe artifactua models Ag an se difference o th variou well-bein dimension replicate prio findings. Comparison wit othe frequentl use indicator (positiv an negativ affect lif satisfaction demonstrate tha th latte neglec ke aspect o positiv functionin emphasize i theorie o healt an well-being Fo mor tha 2 years th stud o psychologica well-bein ha bee guide b tw primar conception o positiv func tioning On formulation traceabl t Bradburn' (1969 sem ina work distinguishe betwee positiv an negativ affec

an denne happines a th balanc betwee th two Concep tua an methodologica refinement buil o thi earl opera tionalizatio o well-being Fo example th postulate inde pendenc o positiv an negativ affec wa challenge an linke wit th failur t distinguis betwee th intensit an th frequenc o affec (Diener Larsen Levine & Emmons 1985) Frequenc o positiv an negativ affec tend t corre lat negatively wherea intensit correlation ar generall pos itive Thes conflictin relation wer sai t suppres th asso ciatio betwee positiv an negativ affect thereb creatin illusio tha th component ar independent O th two frequenc

ha bee promote a th bette indicato o well bein becaus i ca b bette measure an i mor strongl relate t long-ter emotiona well-bein tha intensit i (Diene & Larsen 1993 Diener Sandvik & Pavot 1991) Othe initiative hav focuse o measuremen issues callin fo mor vali an reliabl indicator o positiv an negativ affec (Watson Clark & Tellegen 1988 an suggestin tha measuremen erro obscure th bipolarit o positiv an neg ativ affec (Green Goldman & Salovey 1993) Th secon primar conception whic ha gaine promi nenc amon sociologists emphasize lif satisfactio a th ke indicato o well-being Viewe a a cognitiv

component lif Caro D Ryff Departmen o Psychology Universit o Wisconsin Madison Core Le M Keyes Departmen o Sociology Universit o Wisconsin—Madison Thi researc wa supporte b th Joh D an Catherin T MacArthu Foundatio Researc Networ o Successfu Midlif Development Correspondenc concernin thi articl shoul b addresse t Caro Ryff Departmen o Psychology Brodge Psycholog Building Uni versit o Wisconsin 120 Wes Johnso Street Madison Wisconsin 53706 o t Caro D Ryff Institut o Aging 224 Medica Science Center Universit o Wisconsin—Madison 130 Universit Avenue Madison Wisconsi 53706 satisfactio wa see t

complemen happiness th mor affec tiv dimensio o positiv functionin (e.g. Andrew & McKennell 1980 Andrew & Withey 1976 Bryan & Veroff 1982 Campbell Converse & Rodgers 1976) Stil othe stud ie parse well-bein accordin t globa question abou overal lif satisfactio an domain-specifi question abou work in come socia relationships an neighborhoo (Andrews 1991 Diener 1984) Interes i thes investigation frequentl cen tere o socia change—whethe qualit o lif i Americ mean somethin differen fro on er t th nex an whethe reporte level o well-bein an thei correlate varie ove tim (se als Bryan & Veroff 1982)

Altogether prio endeavor hav grapple minimall wit th cor underlyin question Wha doe i mea t b wel psycho logically Tha is extan indicator hav bee perpetuate wit littl debat a t whethe the capture ke feature o huma wellness Bradburn' (1969 classi study fo example gav lit tl attentio t th fundamenta meanin o well-being Tha positiv an negativ affec emerge a independen dimension was i fact a serendipitou finding fro a stud conceive fo othe purposes Similarly lif satisfactio measure wer gener ate wit a concer fo practica application o researc find- ings no explicatio o essentia meaning o wellnes

(Saue & Warland 1982) Quality-of-lif researc ha als bee de scribe a bein dat drive rathe tha base o a clea con ceptua framewor (Headey Kelley & Wearing 1993) Water man's (1993 distinctio betwee eudaimoni an hedoni conception o happines provide a notabl exceptio i thi largel atheoretica climate. Th absenc o theory-base formulation o well-bein i Whe i come t th tas o explainin th process o well-being tha is ho individual com t posses o no posses thi quality th theoretica terrai i muc riche (se Diener 1984 fo a review) Ou ow wor ha targete adults lif experience an thei interpretation the (e.g.

throug socia compariso processes a ke explanator factor (e.g. Heidric & Ryff 1993a Ryf & Essex 1992 Ryff Lee Essex & Schmutte 1994) Her w emphasiz th nee fo theor i th prio task namely dennin th essentia natur o well-being th indicator o whic serv a outcom variable i suc proces studies 71
Page 2
72 CARO D RYF AN CORE LE M KEYE puzzlin give abundan account o positiv functionin i subfield o psycholog (se Ryff 1985 1989a) Fro develop menta psychology Erikson' (1959 psychosocia stages Buhl er' (1935 basi lif tendencies an Neugarten' (1973 per sonalit change articulat wellnes a

trajectorie o continue growt acros th lif cycle Clinica psychologist offe furthe description o well-bein throug Maslow' (1968 conceptio self-actualization Allport' (1961 formulatio o maturity Rogers (1961 depictio o th full functionin person an Jung' (1933 accoun o individuation Th menta healt lit erature whic typicall elaborate th negativ en o psycho logica functioning nonetheles include som expositio o positiv healt (Birren&Renner 1980;Jahoda 1958) Th convergenc o thes multipl framework o positiv functionin serve a th theoretica foundatio t generat a multidimensiona mode o well-bein (Ryff

1989b 1995) In clude ar si distinc component o positiv psychologica functionin (se Appendix) I combination thes dimension encompas a breadt o wellnes tha include positiv evalua tion o onesel an one' pas lif (Self-Acceptance) a sens o continue growt an developmen a a perso (Persona Growth) th belie tha one' lif i purposefu an meaningfu (Purpos i Life) th possessio o qualit relation wit other (Positiv Relation Wit Others) th capacit t manag effectivel one' lif an surroundin worl (Environmenta Mastery) an a sens o self-determinatio (Autonomy) understan th natur o wellness descriptiv studie hav

focuse o ag an gende profiles Th origina validatio sam pl (Ryff 1989b compare youn (18-2 year old) midlif (30-6 year old) an old-age (6 year ol o older adult an foun incrementa ag profile fo Environmenta Master an Autonom (particularl fro youn adulthoo t midlife) decrementa ag profile fo Purpos i Lif an Persona Growt (particularl fro midlif t ol age) an n ag difference fo Self-Acceptanc an Positiv Relation Wit Others Mos o thes pattern wer replicate i anothe stud (Ryff 1991 involvin th sam thre ag groups I bot in vestigations wome score significantl highe tha me o Positiv Relation Wit Other an

Persona Growt (finding fo th latte dimensio approache statistica significanc i th first study) wit subsequen studie replicatin thes se difference (Ryf e al. 1994 Ryff Lee & Na 1993) Th propose multidimensiona structur o well-bein ha not however bee investigate wit analyti procedure tha tes th fit o th theoretica mode wit empirica data Hig correlation amon certai aspect o well-bein (Ryff 1989b underscor th nee fo theory-guide structura analyses I addition th propose theoretica mode ha no bee assesse a nationall representativ sample—prio wor ha bee conducte primaril wit selective communit

samples Dat fro representativ sample ar neede t tes th generaliz abilit o prio pattern o ag an se differences Thus th objective o thi stud wer threefold (a t test wit a nation all representativ sample th propose multidimensiona mode o well-being (b t examin th replicativ consistenc ag an se difference o th variou indicator o well-be ing an (c t compar th relationship betwee th theory base dimension o well-bein an thre prominen indica tor fro prio researc (i.e. happiness lif satisfaction depression) Metho Sample Dat ar reporte fro a nationa probabilit sampl o noninstitu tionalized

English-speakin adults age 2 o older residin i th 4 contiguou state i th Unite States an whos household include leas on telephone Household wer selecte wit rando digi di alin procedures A adul fro eac househol wa the selecte ran doml an interviewe fo approximatel 3 mi b telephone Th respons rat wa approximatel 62% Dat wer weighte t correc fo overrepresentatio o household wit mor tha on telephon line fo underrepresentatio o adult age 2 o olde livin i household wit mor tha on adult an t matc censu burea estimate o th proportio o English-speakin adults age 2 o older residin i th majo geographica

region (i.e. Northeast Midwest South an West th Unite States Th tota sampl siz wa 1,108 o whic 59 wer female 87 wer Caucasian an th averag ag wa 45. year (SD - 14. years) Mos respondents abou 70% wer married Full one-thir ha graduate fro hig schoo only 26 ha som colleg background an 16 wer colleg graduates Jus ove hal o th sampl reporte a annua househol incom betwee $10,00 an $39,999 o whic 19 re porte a incom betwee $ 10,00 an $ 19,999,2 % betwee $20,00 an $29,999 an 18 reporte a househol incom o betwee $30,00 an $39,999 Fo analyti purposes w divide respondent int thre ag groups youn adult (

= 133 wer betwee th age o 2 an 29 midlif adult ( = 805 wer betwee th age o 3 an 64 an olde adult (n = 160 wer 6 o older Abou 60 o th sampl i eac ag grou wa female Thoug eac ag grou consiste mostl o Caucasians th racia homogeneit increase i th grou o olde adults Mor tha hal th youn adult eithe ha som colleg o graduate fro college wherea onl 44 o th middle-aged an stil fewe o olde adult (33%) ha som colleg o graduate fro college Compare wit th youn adults abou 70 o th middle-aged a wel a th olde adults wer married Th moda incom rang fo youn adult wa betwee $30,00 an $39,999 $20,000-$29,99 fo

middle-age adults an 10,000- 19,99 fo th olde adults Dat fro tw additiona sample (Ryff 1989b Ryffe al. 1994 ar include fo comparativ purposes Bot wer communit voluntee samples wit th forme consistin o young midlife an old-age adult (se previou definition o ag ranges an th latte includin onl midlif adults Detaile description o thes sample ar availabl Ryff 1989b an Ryffe al (1994) Measures and Procedure generate definition o th si dimension o psychologica well bein (se Tabl 1 fro th multipl theoretica account o positiv functioning I th initia validatio stud (Ryff 1989b) eac dimen sio wa

operationalize wit a 20-ite scal (tha showe hig interna consistenc an test-retes reliabilit a wel a convergen an discrim inan validit wit othe measures) T accommodat tim an cos restriction o a nationa survey w chos onl 3 o th origina 2 item measur eac construct Becaus al paren scale ha multifactoria structures w selecte item fro subfactor withi eac longe scal t maximiz th conceptua breadt o th shortene scales Th short ene scale correlate fro .7 t .8 wit 20-ite paren scales Eac scal include bot positivel an negativel phrase items Th re spons scal wa a 6-poin continuum rangin fro completely dis-

agree t completely agree. Interviewer administere th item usin a
Page 3
STRUCTUR O WELL-BEIN 72 Tabl 1 Descriptive Statistics of Theoretically Grounded Scales of Psychological Weil-Being Scal Self-Acceptanc Environmenta Master Positiv Relation Purpos i Lif Persona Growt Autonom SD .5 14. 3. .4 .4 14. 2. .4 .3 .5 14. 3. .2 .1 .1 .3 14. 3. .1 .2 .1 .3 .4 15. 2. .2 .2 .1 .1 .2 .3 15. 2. Note. Al correlatio coefficient ar statisticall significan a th .0 level unfoldin techniqu (Groves 1989)—firs askin whethe th respon den agree o disagree wit th statemen an the askin ho muc (strongly

moderately o slightly) Fo comparativ purposes th intervie als include single-ite in dicator o happines an lif satisfaction Fo th former respondent answere ho muc o th tim durin th pas mont ( = none 1 = some 2 = most 3 = all the fel happ (mea scor = 1.7 SD = 0.64) Fo th latter respondent summarize ho thing wer goin i thei lif o a scal fro 0 (the worst possible life you could imagine) t 1 th best possible life you could imagine; mea scor = 7.7 SZ = 1.6) Depressio wa als measure wit eigh item indicatin ho muc o th tim ( = none 1 = some 2 = most 3 = all durin th pas mont respondent fel (1 ful o

life (2 wor out (3 tired (4 downhearte an blue (5 cal an peaceful (6 nervous (7 ha a lo o energy an (8 wer s dow i th dump tha nothin coul chee yo u (mea scor = 8.1 SD = 3.4) Facto analyse reveale tw underlyin dimensions whic wer labele Dysfunctiona Energ an Dysfunc tiona Affect Fro th tw prio investigation (Ryff 1989b Ryffe al. 1994) dat ar reporte o th si measure o well-bein (20-ite scale i th first stud an 14-ite scale i th second i relatio t positiv an negativ affect affec balanc (Bradburn 1969) th Lif Satisfac tio Inde (Neugarten Havighurst & Tobin 1961) th ZungDepres- sio Scal (Zung

1965) th Cente fo Epidemiologi Stud Depres sio Scal (CES-D Radloff 1977) an single-ite indicator o overal happines an lif satisfaction (Se publishe studie fo details. Result Item and Scale Analyses Preliminar analysi (result no shown indicat tha th 1 item continu t mee psychometri criteria wit eac ite correlatin strongl an positivel wit onl it ow scale. Scal intercorrelation ar modes (se Tabl 1) rangin fro .1 (e.g. Purpos i Lif an Autonomy t .4 (Self-Acceptanc an Environmenta Mastery) Estimate o interna consistenc (alpha) coefficient wer lo t modest rangin fro .3 (Purpos i Life t .5 (Positiv

Relation Wit Others) Th alph coefficien i a conservativ estimat o interna reliabilit fo mos (congeneri indicators scale (Bollen 1989) Th modes alph coefficient likel reflec th smal numbe o in dicator pe scal an th fac tha item wer chose t repre sen th conceptua breadt withi eac construc (se ratio nal fo ite selectio i Measures and Procedure section rathe tha t maximiz interna consistency Age and Sex Differences in Weil-Being Mean-leve analyses despit dramati reduction i dept o measurement replicate man prio findings (se Figur 1) Th subscale o Purpos i Life F(2, 921 = 19.8 p < .001 an Persona

Growth F 2,921 = 16.4 p < .001 continue t sho decrementa ag profile (wit score o th oldes respon dent significantl lowe tha thos o th tw younge ag groups Environmenta Mastery F(2, 921 = 3.05 p < .05 als continue t sho ag increment (wit bot olde group scorin significantl highe tha youn adults) an self-accep tanc showe n ag differences Autonom agai showe ag increments F(2, 921 = 4.97 p < .01 bu onl fro youn adulthoo t midlife Althoug th tw previou studie indi cate n ag difference i Positiv Relations thes dat showe incrementa score wit age F( 2,921 = 7.12 p < .00 (olde respondent score highe tha

bot younge ag groups) Th onl scal tha showe significan se difference wa Positiv Relation Wit Others F( 1,921 = 8.94 p < .01 wit wome agai scorin highe tha men Structural Analyses estimate th parameter an fit o th measuremen model wit LISRE 7. (Joresko & Sorbom 1989) I al model th measuremen erro wa assume t occu ran domly—ou goa wa no t improv th fit o th measuremen mode bu t determin whethe th "clean mode (i.e. n correlate measuremen error fit th dat well Beside th the oretica model severa othe model capturin distinc an per hap mor parsimoniou explanation fo th structur o th dat als wer

assessed Fo example w estimate a unidimen siona mode i whic al indicator loade o a singl facto o well-being W als explore artifact o measuremen b esti Nearl al o th item-to-scale coefficient ar statisticall signifi cant a reflectio o th larg sampl siz and hence statistica power T b consisten wit othe analyse i th article w use listwis deletio o missin dat whe computin alph coefficients Th sampl siz afte listwis deletio i n = 928 compare wit a tota sampl siz 1,108 Contrast betwee eac ag grou wer mad wit Tukey' Honestl Significan Difference procedure wit alph se a .05
Page 4

RYF AN CORE LE M KEYE Youn Midlif Olde Dimension o Well-Bein I-Sel Acceptanc O Positiv Relation A Persona Growt A Purpos i Lif #Env Master O Autonom Figure 1. Ag difference o th si 3-ite measure o psychologica well-being Env Master = Envi ronmenta Mastery matin tw type o models eac suggestin tha respondent answe question t portra a positiv self-image Th first arti factua mode extende th argumen tha a singl dimensio (i.e. well-being explaine th structur o th dat an furthe specifie tha negativel worde item loade o a artifac di mension A secon artifactua mode suggest tha peopl en gag i a

agreement-disagreemen bia t portra a positiv self-image The agre wit al positivel worded an disagre wit al negativel worded items Thus w estimate a mode wit tw laten artifac constructs on fo positivel worde item an a secon fo negativel worde items Next w com pare th fit o th theoretica mode wit th singl facto an artifactua model wit a super-facto mode i whic th si laten construct ar effect o a second-orde laten construc calle psychological well-being. Descriptiv analysi indicate tha th margina distribution nearl al item ar skewe substantiall (result no shown) therefor use weighte leas square

estimation whic produce distribution-fre an asymptoticall unbiase an efficien estimates I theory weighte leas square i superio maximu likelihoo estimatio fo fitting model wit indi cator whos distribution ar non-norma an skewed espe ciall a sampl siz increase (Bollen 1989 Boomsma 1983 Joresko & Sorbom 1989 . W produce th variance-covar ianc matri an asymptoti covarianc matri o th estimate variance an covariance usin PRELI (Joresko & Sorbom 1988) Usin listwis deletio o missin data w purge 18 respondent fro th sample Th conclusion usin mean-sub stitutio o missin dat wer unchanged W presen onl th

result usin listwis deletio o missin data Tabl 2 display thre indice o th overal fit o eac model Smal an statisticall nonsignifican value o th chi-squar statisti indicat good-fittin models However non o th model fit th dat accordin t chi-square whic i sensitiv t sampl siz an sampl variabilit (Bollen 1989 Raykov Tomer & Nesselroade 1991) W therefor repor th Bayesia informatio criterio (BIC Raftery 1993) whic i increas ingl likel t indicat goo fitting model a sampl siz in creases Positiv value o BI indicat poor-fittin models an negativ value indicat good-fittin models Moreover smalle (i.e.

increasingl negative value o BI indicat increasingl better-fittin models Last th adjuste goodness-of-fi inde (AGFI als indicate th overal fit o th model Th AGF varie betwee 0. an 1 value tha approac unit indicat bette models A a rul o thumb value o AGF .9 o highe indicat very-good-fittin models Comparin th overal fit indice fo th first fou model i Tabl 3 reveal tha th six-facto mode i th best-fittin model Th chi-squar valu i lower AGF i higher an BI i increasingl negative Th six-facto model i fact reveal a vas improvemen i fit ove eac o th precedin models par ticularl th single-facto model I

turn Mode 5—th super facto model—propose tha th si factor ar a functio of o ar cause by anothe laten construct Tha is eac o th si factor belong t a singl conceptua domai calle well-being. Ther ar trade-off t eac estimatio technique On mus b con cerne wit th robustness o maximu likelihoo estimate an th consistency o asymptotic distribution-fre estimates Th conclusion thi study however remai th sam whe eithe typ o estimato i use (result o maximu likelihoo estimatio ar availabl fro Caro D Ryf o request)
Page 5
STRUCTUR O WELL-BEIN 72 Tabl 2 Confirmatory Factor Analyses: Indices of Fit

Based on Weighted Least Squares Estimation Mode df AGF BI Single-facto Single-facto an negativ ite artifact Two-factor negativ an positiv ite artifac Six-facto Second-order singl super-facto 531. 442. 488. 339. 378. 13 12 13 12 12 .8 .8 .8 .8 .8 -38.1 -93.7 -47.7 -167.6 -166.0 Note, n = 928 AGF = adjuste goodness-of-fi index BI = Bayesia informatio criterion Th correlatio betwee th laten constructs overal well-bein an th negativ ite artifact i con straine t b 0 Thi mode i mor restrictiv an parsimoniou tha th six facto model Comparin th BI statistic th super-facto mode fits th dat bette tha th

six-facto model Accordin Rafter (1993) th BI statisti shoul increas a leas 1 point fo eac degre o freedo los t conclud confidentl tha th mor comple mode (i.e. th mode wit fewe degree freedom—Mode 4 fits th dat bette tha th mor parsi moniou mode (i.e. Mode 5) Th differenc i th BI sta tisti i onl abou 2 wit a differenc o 9 degree o freedom i importan t emphasiz tha th super-facto mode i no th sam a th single-facto model Th super-facto mode say tha si factor fit th dat an tha thes si factor mea sur a singl laten construc calle psychologica well-being Parsimon exists bu a a highe order Thus ther i

a hierar chica structur i whic genera well-bein ha it effect throug th si conten domain specifie a prior b guidin psychologica theory Th ag analyse furthe clarif tha lif cours effect o well-bein canno b explaine wit a genera facto becaus thes ag effect ar no uniform—differen di mension o well-bein sho differen ag profiles Tabl 3 contain th estimate o th correlation betwee th unmeasured laten constructs Th estimate correlation be twee th laten construc ar importan fo descriptiv reason and i thi case t highligh th necessit o theory-drive structura analysis Mos o th correlation ar modest aroun

.50 However th correlatio betwee Self-Acceptanc an En Tabl 3 Correlations Among the Latent Constructs, From Weighted Least Squares Estimates of the Six-Factor Model Construc Self-Acceptanc Positiv Relation Environmenta Master Persona Growt Autonom Purpos i Lif .6 .8 .5 .5 .5 .6 .3 .2 .3 .5 .5 .3 .5 .6 .3 Note, n = 928 Al t values—tha is th rati o th estimate coeffi cien t it estimate asymptoti standar error—ar statisticall sig nifican a th .0 level vironmenta Master i ver high Suc a hig correlatio be twee laten construct coul indicat redundanc o share source o variance W note however tha

despit th apparen overlap Environmenta Master an Self-Acceptanc exhibi differen ag profiles Thus possibl redundanc i structura analyse i contraste wit apparen distinctnes i lif cours analyses illustratin th importanc o theor an multipl mode o testin i i assessin th structur o th well-bein domain Correlations With Other Measures Tabl 4 provide zero-orde correlation fro thre separat studie o eac o th si scale o well-bein wit othe promi nen indicator o well-being namely happiness lif satisfac tion an depression Acros thes dat sets measure o happi nes (affec balanc o single-ite indicators sho

modes t stron association wit Self-Acceptanc an Environmenta Mastery somewha weake link wit Purpos i Life an stil weake tie t Positiv Relation Wit Others Persona Growth an Autonomy Parallel althoug generall stronger pattern o associatio ar eviden fo lif satisfactio (measure bot a a multi-ite scal an a single-ite indicators)—stronges re lation ar eviden fo self-acceptanc an environmenta mas tery wit th remainin coefficient showin wea t modes associations Finally th multipl indicator o depressio sho consistentl negativ association wit al dimension o well being wit th stronges pattern agai eviden

fo Self-Accep tanc an Environmenta Mastery Discussio Th purpos o th presen stud wa t tes th propose theoretica structur o a multidimensiona mode o psycho logica well-being Draw fro point o convergenc i prio theorie o lif cours development clinica account o positiv functioning an menta healt conceptions th mode include si distinc component o psychologica wellness Self-Accep tance Environmenta Mastery Purpos i Life Positiv Re lation Wit Others Persona Growth an Autonomy Con firmatory facto analyse wit dat fro a nationall represen tativ sampl supporte th propose multidimensiona
Page 6

72 CARO D RYF AN CORE LE M KEYE Tabl 4 Correlations Between Theory-Based Scales oj Well-Being and Prior Measures Stud Ryff 1989b) Ryffe al (1994) Presen stud Prio measure Happines Affec balanc Negativ affec Positiv affec Satisfactio Lif Satisfactio Inde Depressio Zun Depressio Scal Happines Singl item globa Satisfactio Singl item globa Depressio CES- Happines Singl item amoun durin pas mont Satisfactio Singl item rat lif overal Depressio Dysfunctiona energ Dysfunctiona affec .5 -.4 .4 .7 -.5 .5 .6 -.7 .3 .4 -.3 -.4 .3 -.1 .2 .4 -.3 .3 .4 -.4 .2 .3 -.2 -.3 Ne scale .4 -.2 .4 .5 -.6 .4 .5 -.5 .1

.1 -.05 -.1 .2 -.ll .3 .3 -.4 .1 .2 -.2 .1 .1 -.1 -.1 .3 -.3 .2 .2 -.3 .3 .3 -.4 .0 .1 -.1 -.1 .6 -.5 .4 .6 -.6 .5 .6 -.6 .4 .3 -.4 -.5 Note. S = Self-Acceptance P = Positiv Relation Wit Others P = Purpos i Life P = Persona Growth A = Autonomy E = Environmenta Mastery CES- = Cente fo Epidemiologica Stud Depressio Scal (Radloff 1977) "Affec Balanc (Bradburn 1969) Negativ Affect an Positiv Affec ar th subscale tha compris Affec Balance Lif Satisfactio Inde (Neugarte e al. 1961) Zun Depressio Scal (Zung 1965) Th happines ite is "Al thing considered ho happ ar you? Th satisfactio ite is "Thinkin

abou you lif a a whole ho satisfie ar you? No statisticall significan a th .0 alph level Al othe coefficient ar statisticall significan a leas th .0 alph level structure Th mode tha bes fit th dat wa on o si pri mar factor joine togethe b a singl highe orde factor Thi mode showe dramati improvemen i fit ove suggeste al ternatives especiall th single-facto model Th theoretica formulatio o well-bein wa thus supporte a a multifacete domai encompassin positiv self-regard master o th sur roundin environment qualit relation wit others continue growt an development purposefu living an th capacit fo

self-determination Th dat als poin t th replicativ consistenc o ag an se difference o thes variou aspect o well-being Acros multipl investigation havin wid variatio i dept o mea suremen (20-item 14-item an 3-ite scales) declinin ag profile wer obtaine o Purpos i Lif an Persona Growth incrementa score wer eviden fo Environmenta Master an Autonomy an n ag difference wer obtaine fo Self Acceptance Pattern fo Positiv Relation varie betwee showin n ag difference o incrementa patterns Longitudi na dat ar obviousl neede t clarif whethe thes ag pro file represen maturationa changes o cohor differences

Whateve th "source o thes differences th result un derscor th diversit o lif cours an cohor profile o well being Finally acros al studies wome wer foun t scor highe tha me o Positiv Relation Wit Others Departin fro prio result includin cross-cultura compariso (Ryffe al. 1993) thes dat di no sho tha wome als ha highe score o persona growth Presumably i i th extensiv reduc tio i conten o th growt scale o th differenc i sampling tha account fo thi divergenc fro prio studies Compariso o th theory-base indicator o well-bein wit othe frequentl use measure indicate moderat t stron association betwee

tw scale (Self-Acceptanc an Environ menta Mastery an single an multi-ite scale o happiness lif satisfaction an depression However th remainin fou dimension o well-bein (Positiv Relation Wit Others Pur pos i Life Persona Growth Autonomy showe mixe o wea relationship wit thes prio indicators Continue em pirica relianc o thes earlie indice thu translate t neglec ke aspect o positiv functionin emphasize i theoretica accounts i importan t recogniz tha thes guidin theorie giv surprisingl littl commentar t happines o positiv affec a
Page 7
STRUCTUR O WELL-BEIN 72 definin featur o huma

wellness I fact i ha bee argue tha certai aspect o positiv functioning suc a th realiza tio o one' goal an purposes requir effor an disciplin tha ma wel b a odd wit short-ter happines (Waterman 1984) I addition philosophica account cautio agains a fo cu o happines a th ultimat goo i life Histor provide countles example o thos wh live ugly unjust o pointles live wh wer nonetheles happ (se Becker 1992) Eve whe presen i exemplar lives philosopher constru happi nes no a a en i itsel bu a byproduc o other mor nobl pursuit (Mill 1873/1989) Thes observations combine wit th finding tha mos peopl (eve

th disabled abused an unemploye people repor themselve t b happ (Diener 1993 Taylo & Brown 1988) rais question (theoretical philosophical empirical abou th scientifi at tentio lavishe o happines an positiv affect particularl a th expens o othe aspect o positiv functioning Compre hensiv account o psychologica well-bein nee als t prob people' sens o whethe thei live hav purpose whethe the ar realizin thei give potential wha i th qualit o thei tie t others an i the fee i charg o thei ow lives Apar fro expandin th substantiv meanin o psycholog ica well-being ou cal t reexamin th contour o positiv

functionin illustrate th complexit involve i definin an assessin structur withi a particula domain Ideally struc tura analyse begi wit a well-articulate theoretica frame work whic i the operationalize an teste wit procedure designe t asses th propose theory suc a confirmator facto analysis However i th process i become apparen tha final answer d no en wit examinatio o factoria structure I thi investigatio suc analyse indicate tha tw th si theoretica construct (Self-Acceptanc an Environ menta Mastery wer highl correlated I othe words data no theory—suggeste a possibl five-factor model whic woul

combin indicator o Self-Acceptanc an Environmen ta Mastery Examinatio o structur throug othe analyses namely lif cours profiles showed however tha Self-Accep tanc an Environmenta Master exhibite distinc ag pro files (th forme showin littl variatio b age th latte show in incrementa ag differences) Analyse o additiona grou difference (e.g. b socia class ethnicity o culture woul fur the infor understandin o th basi structur o well-being Stil othe pertinen informatio pertain t ho th differen dimension o positiv functionin var a outcome measures process-oriente studie (se Heidric & Ryff 1993a 1993b

Ryf & Essex 1992 Ryf e al. 1994 Schmutt & Ryff 1994 Twee & Ryff 1991 Va Riper Ryff & Pridham 1992) T th exten tha lif transition affec some bu no othe dimension wellness furthe evidenc i garnere fo th multidimen sionalit o th well-bein domain Finally assessment o struc tur nee t incorporat metho variation Al o th analyse described-thu fa rel o self-repor techniques whic ma in clud self-presentatio biases Augmentin thes dat wit ob servationa methods o dat obtaine fro significan other (e.g. spouses) woul strengthe th veracit o th hypothe size model I sum mappin th fundamenta structur o

psychologica well-bein i a multitas agenda requirin ongo in synthese o divers source o evidence Fo now w offe th provisiona conclusio tha ther i mor t bein wel tha feelin happ an satisfie wit life Reference Allport G W (1961) Pattern and growth in personality. Ne York Holt Rinehar & Winston Andrews F M (1991) Stabilit an chang i level an structur o subjectiv well-being Social Indicators Research, 25, 1-30 Andrews F M. & McKennell A C (1980) Measure o self-reporte well-being Social Indicators Research, 8, 127-156 Andrews F M. & Withey S B (1976) Social indicators of well-being: America's

perception of life quality. Ne York Plenum Becker L C (1992) Goo lives Prolegomena Social Philosophy and Policy, 9(2) 15-37 Birren J E. & Renner V J (1980) Concept an issue o menta healt an aging I J E Birre & R B Sloan (Eds.) Handbook of mental health and aging (pp 3-33) Englewoo Cliffs NJ Prentic Hall Bollen K A (1989) Structural equations with latent variables. Ne York Wiley Boomsma A (1983) On the robustness ofLISREL (maximum likeli- hood estimation) against small sample size and non-normality. Am sterdam Sociometri Researc Foundation Bradburn N M (1969) The structure of psychological

well-being. Chicago Aldine Bryant F B. & Veroff J (1982) Th structur o psychologica well being A sociohistorica analysis Journal of Personality and Social Psychology, 43, 653-673 Buhler C (1935) Th curv o lif a studie i biographies Journal of Applied Psychology, 19, 405-409 Campbell A. Converse P E. & Rodgers W L (1976) The quality of American life: Perceptions, evaluations, and satisfactions. Ne \brk Russel Sag Foundation Diener E (1984) Subjectiv well-being Psychological Bulletin, 95, 542-575 Diener E (1993) Most Americans are happy. Unpublishe manu script Departmen o Psychology Universit o

Illinois Diener E. & Larsen R J (1993) Th experienc o emotiona well being I M Lewi & J M Havilan (Eds.) Handbook of emotions (pp 405-415) Ne York Guilfor Press Diener E. Larsen R J. Levine S. & Emmons R A (1985) Intensit an frequency Dimension underlyin positiv an negativ affect Journal of Personality and Social Psychology, 48, 1253-1265 Diener E. Sandvik E. & Pavot W (1991) Happines i th frequency no th intensity o positiv versu negativ affect I F Strack M Argyle & N Schwart (Eds.) Subjective well-being: An interdisci- plinary perspective (pp 119-139) Elmsford NY Pergamo Press Erikson E

(1959) Identit an th lif cycle Psychological Issues, 1, 18-164 Green D P. Goldman S L. & Salovey P (1993) Measuremen erro mask bipolarit i affec ratings Journal of Personality and Social Psychology, 64, 1029-1041 Groves R (1989) Survey errors and survey costs. Ne York Wiley Headey B. Kelley J. & Wearing A (1993) Dimension o menta health Lif satisfaction positiv affect anxiet an depression So- cial Indicators Research, 29, 63-82 Heidrich S M. & Ryff C D (1993a) Th rol o socia compariso processe i th psychologica adaptatio o elderl adults Journal of Gerontology, 48, 127-136 Heidrich S M. & Ryff

C D (1993b) Physica an menta healt i late life Th self-syste a mediator Psychology and Aging, 8, 327 338 Jahoda M (1958) Current concepts of positive mental health. Ne York Basi Books
Page 8
72 CARO D RYF AN CORE LE M KEYE Joreskog K G. & Sorbom D (1988) PRELIS: A program for multi- variate data screening and data summarization (2n ed.) Moores ville IN Scientifi Software Joreskog K G. & Sorbom D (1989) LISREL 7: A guide to the pro- gram and applications (2n ed.) Chicago SPSS Jung C G (1933) Modern man in search of a soul(W. S Del & C F Baynes Trans.) Ne York Hartcourt Brac & World

Maslow A H (1968) Toward a psychology of being (2n ed.) Ne York Va Nostrand Mill J S (1989) Autobiography. London Pengui Books (Origina wor publishe 1873 Neugarten B L (1973) Personalit chang i lat life A developmen ta perspective I C Eisdorfe & M P Lawto (Eds.) The psychol- ogy of adult development and aging (pp.311-335) Washington DC America Psychologica Association Neugarten B L. Havighurst R. & Tobin S (1961) Th measure men o lif satisfaction Journal of Gerontology, 16, 134-143 RadlofF L (1977) Th CES- Scale A self-repor depressio scal fo researc i th genera population Applied

Psychological Measure- ment, 7,385-401 Raftery A E (1993) Bayesia mode selectio i structura equation models I K A Bolle & J S Lon (Eds.) Testing structural equa- tion models (pp 163-180) Newbur Park CA Sage Raykov T Tomer A. & Nesselroade J R (1991) Reportin struc tura equatio modelin result i Psychology and Aging: Som pro pose guidelines Psychology and Aging, 6, 499-503 Rogers C R (1961) On becoming a person. Boston Houghto Mifflin Ryff C D (1985) Adul personalit developmen an th motivatio fo persona growth I D Kleibe & M Maeh (Eds.) Advances in motivation and achievement: Vol. 4. Motivation

and adulthood (pp 55-92) Greenwich CT JA Press RyfF C D (1989a) Beyon Ponc d Leo an lif satisfaction Ne direction i ques o successfu aging International Journal of Be- havioral Development, 12, 35-55 Ryff C D (1989b) Happines i everything o i it Exploration o th meanin o psychologica well-being Journal of Personality and Social Psychology, 57, 1069-1081 Ryff C D (1991) Possibl selve i adulthoo an ol age A tal o shiftin horizons Psychology and Aging, 6, 286-295 Ryff C D (1995) Psychologica well-bein i adul life Current Di- rections in Psychological Science, 4, 99-104 Ryff C D. & Essex M J

(1992) Th interpretatio o lif experienc an well-being Th sampl cas o relocation Psychology and Aging, 7,507-517 Ryff C D. Lee Y H. Essex M J. & Schmutte P S (1994) M childre an me Midlif evaluation o grow childre an self Psy- chology and Aging, 9, 195-205 Ryff C D. Lee Y H. & Na K C (1993 November) Throug th len o culture Psychologica well-bein a midlife Pape presente th meeting o Gerontologica Societ o America Ne Orleans LA Sauer W J. & Warland R (1982) Moral an lif satisfaction I J Mange & W A Peterso (Eds.) Research instruments in social gerontology: Vol. 1, Clinical and social psychology

(pp 195-240) Minneapolis Universit o Minnesot Press Schmutte P S. & Ryff C D (1994) Success socia comparison an self-assessment Parents midlif evaluation o sons daughters an self Journal of Adult Development, 1, 109-126 Taylor S E. & Brown J D (1988) Illusio an well-being A socia psychologica perspectiv o menta health Psychological Bulletin, 103, 193-210 Tweed S. & Ryff C D (1991) Adul childre o alcoholics Profile wellnes an distress Journal of Studies on Alcohol, 52, 133-141 Va Riper M. Ryff C D. & Pridham K (1992) Parenta an famil well-bein i familie o childre wit Dow syndrome A compara tiv

study Research in Nursing and Health, 15, 227-235 Waterman A S (1984) The psychology of individualism. Ne York Praeger Waterman A S (1993) Tw conception o happiness Contrast o persona expressivenes (eudaimonia an hedoni enjoyment Journal of Personality and Social Psychology, 64, 678-691 Watson D. Clark L A. & Tellegen A (1988) Developmen an val idatio o brie measure o positiv an negativ affect Th PANA scales Journal of Personality and Social Psychology, 54, 1063-1070 Zung W K (1965) A self-ratin depressio scale Archives of General Psychiatry, 12, 63-70 (Appendix follows on next page)

STRUCTUR O WELL-BEIN 72 Appendi Definition o Theory-Guide Dimension o Weil-Bein Self-Acceptanc Hig scorer possesse a positiv attitud towar th self acknowledge an accept multipl aspect o self includin goo an ba qualities feel positiv abou pas life Lo scorer feel dissatisfie wit self i disappointe wit wha ha occurre i pas life i trouble abou certai persona qualities wishe t b differen tha wha h o sh is Positiv Relation Wit Other Hig scorer ha warm satisfying trustin relationship wit others i concerne abou th welfar o others capa bl o stron empathy affection an intimacy understand giv an

tak o huma relationships Lo scorer ha fe close trustin relationship wit others finds i difficul t b warm open an concerne abou others i isolate an frustrate i interpersona relationships no willin t mak compromise t sustai importan tie wit others Autonom Hig scorer i self-determinin an independent abl t resis socia pressure t thin an ac i certai ways regulate behavio fro within evaluate sel b persona standards Lo scorer i concerne abou th expectation an evaluation o others relie o judgment o other t mak importan decisions conform t socia pressure t thin an ac i certai ways Environmenta Master

Hig scorer ha a sens o master an competenc i managin th environment control comple arra o externa activities make effectiv us o surroundin opportunities abl t choos o creat context suitabl t persona need an values Lo scorer ha difficult managin everyda affairs feel unabl t chang o improv surroundin context i unawar o surroundin opportunities lack sens o contro ove externa world Purpos i Lif Hig scorer ha goal i lif an a sens o directedness feel ther i meanin t presen an pas life hold belief tha giv lif purpose ha aim an objective fo living Lo scorer lack a sens o meanin i life ha fe goal o

aims lack sens o direction doe no se purpos i pas life ha n outlook o belief tha giv lif meaning Persona Growt Hig scorer ha a feelin o continue development see sel a growin an expanding i ope t ne experiences ha sens o realizin hi o he potential see improvemen i sel an behavio ove time i changin i way tha reflec mor self-knowledg an effectiveness Lo scorer ha a sens o persona stagnation lack sens o improvemen o expansio ove time feel bore an unintereste wit life feel unabl t develo ne attitude o behaviors Receive Apri 15,199 Revisio receive Februar 5,199 Accepte Februar 7,199