The energy and identification continua of burnout and work engagement : Developmental profiles over eight years

Con en s lis s a ailable a ScienceDi ec Bu nou Resea ch jou nal homepage: www.else ie .com/loca e/bu n The ene gy and iden ifica ion con inua o bu nou and wo k engagemen : De elopmen al p ofiles o e eigh yea s Anne Mäkikangas ⁎ , Ka iina Hy önen, Ta u Feld Depa men o Psychology, Uni e si y o Jy äskylä, Finland ARTICLE INFO Keywo ds: Bu nou Wo k engagemen Pe son-cen e ed Longi udinal ABSTRACT Unde s anding o he mu ual de elopmen al dynamics be ween bu nou and wo k engagemen is limi ed due o he lack o longi udinal s udies wi h long ollow-ups and mul i-wa e da a. This s udy sough o iden i y sub- g oups o employees cha ac e ized by long- e m exhaus ion- igo (ene gy con inuum) and cynicism-dedica ion (iden ifica ion con inuum). A u he impo an aim was o in es iga e diffe ences be ween he iden ified sub- g oups in hei expe iences o p og ess in hei pe sonal wo k goals. Fi e-wa e, eigh -yea ollow-up da a among Finnish whi e-colla p o essionals (n= 168) we e s udied using La en P ofile Analysis (LPA). The analysis yielded h ee exhaus ion- igo subg oups: 1) “Low s able exhaus ion –high s able igo ”(n= 141), 2) “Fluc ua ing exhaus ion and igo ”(n= 19), and 3) “S able a e age exhaus ion –dec easing igo ”(n= 8). Th ee subg oups we e also ound o cynicism-dedica ion: 1) “Low s able cynicism –high s able dedica ion” (n= 124), 2) “Inc easing cynicism –dec easing dedica ion”(n= 27), and 3) “Dec easing cynicism –inc easing dedica ion”(n= 17). Exhaus ion and igo we e ound o be s able and mu ually exclusi e expe iences o he g ea majo i y o he pa icipan s. Howe e , mean changes we e also de ec ed –especially in igo –bu hese we e a e. A no able finding was ha he le els o and changes in cynicism and dedica ion showed opposi e ends in each subg oup: among he majo i y o he pa icipan s (74%), he le els o cynicism and dedica ion we e s able and in e sely ela ed, while among one- hi d hei le els simul aneously changed in he e e se di ec ion. The mos success ul p og ess in pe sonal wo k goals was ound in he g oups desc ibed by he iden- ifica ion con inuum, i.e., in he g oups o “Low s able cynicism –high s able dedica ion”and “Dec easing cynicism –inc easing dedica ion”. 1. In oduc ion Du ing ecen decades, inc easing a en ion has been paid o he ela ion be ween job bu nou and i s opposi e posi i e pole wo k engagemen ( o me a-analyses, see Cole, Wal e , Bedeian, & O’Boyle, 2012;C aw o d, LePine, & Rich, 2010). Despi e se e al yea s o p og ess in his esea ch domain, li le is known abou hei mu ual ela ion in he long e m, as he ew exis ing longi udinal s udies ha e u ilized a he sho ollow-up pe iods. In addi ion, he p e ious esea ch has offe ed a a he s a ic and gene ic pe spec i e o he de elopmen o bu nou and wo k engagemen which, heo e ically, should ins ead be ega ded as p og essi e and dynamic (Bakke & Deme ou i, 2007; Lei e , 1993; Maslach, 1982; Shi om, 2003) and indi idualized (Lei e & Maslach, 2016;Mäkikangas, Feld , Kinnunen, & Tol anen, 2012). These ea u es o he cu en bu nou -wo k engagemen li e a u e p omp a call o mo e esea ch on he long- e m de elopmen al p ofiles o bu nou and wo k en- gagemen . This s udy sough o deepen unde s anding o he connec ion be- ween bu nou and wo k engagemen by in es iga ing hei ela ion o e an eigh -yea pe iod, using a s a is ically inno a i e pe son-cen e ed app oach (Be gman, Magnusson, & El-Khou i, 2003; Mäkikangas & Kinnunen, 2016) which cap u es bu nou and wo k en- gagemen expe iences simul aneously wi hin he indi idual o e ime. In es iga ion o change and s abili y p ofiles yields a mo e ealis ic iew o he empo al and de elopmen al dynamism o an employee’s expe ience o well-being. This app oach also leads o a be e heo e- ical unde s anding, as i enables us o iden i y ypical and a ypical long- e m de elopmen al pa hs and di ec ions o change in bu nou and wo k engagemen . This s udy makes h ee p ima y con ibu ions o he exis ing pe son-cen e ed bu nou -wo k engagemen s udies (Mäkikangas e al., 2012). Fi s , we examine he long- e m de elopmen o bu nou and wo k engagemen by using a da ase wi h mo e measu emen s and a longe ollow-up han p e iously. Thus, he ecommended c i e ion o a longi udinal s udy, i.e., h ee o mo e measu emen s, is h p://dx.doi.o g/10.1016/j.bu n.2017.04.002 Recei ed 28 Sep embe 2016; Recei ed in e ised o m 4 Ap il 2017; Accep ed 14 Ap il 2017 ⁎ Co esponding au ho a : Depa men o Psychology, P.O. Box 35, 40014 Uni e si y o Jy äskylä, Finland. E-mail add ess: anne.makikangas@jyu.fi(A. Mäkikangas). Bu nou Resea ch 5 (2017) 44–54 2213-0586/ © 2017 The Au ho s. Published by Else ie GmbH. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/BY-NC-ND/4.0/). MARK me (Kelloway & F ancis, 2013). Second, ins ead o in es iga ing he in e connec ions o sepa a ely analyzed bu nou and wo k engage- men p ofiles, long- e m p ofiles o bu nou and wo k engagemen a e assessed simul aneously (c . Mäkikangas e al., 2012). Consequen ly, he p esen s udy p oduces a be e unde s anding o he diffe en na u ally occu ing longi udinal bu nou -wo k engagemen p ofiles in a gi en s udy popula ion. Thi d, his s udy con ibu es o he li - e a u e by showing o he fi s ime how employees’app aisals o hei goal p og ess a y o e ime in diffe en longi udinal bu nou - wo k engagemen p ofiles. Pe sonal wo k goals a e impo an om bo h he indi idual and o ganiza ional pe spec i es since goals guide he beha io al o ien a ions o indi iduals (Pomaki, Maes, & e Does , 2004). Pe sonal wo k goals can also eflec how employees adjus o changing wo k con ex s, which equi e inc easing p oac i i y and adap abili y (e.g., B iscoe & Hall, 2006; Hy önen e al., 2016; Inkson, 2006). In his sec ion, he concep s o bu nou and wo k engagemen a e in oduced along wi h empi ical and s a is ical e idence, based on bo h a iable- and pe son-cen e ed esea ch, on hei ela ionship. As he opic o his s udy is on he long- e m de elopmen o wo k engagemen and bu nou , bo h he heo e ical iews and empi ical findings ela ing o hei sepa a e and simul aneous long- e m de elopmen a e in ocus. Finally, we in es iga e diffe ences be ween he bu nou -wo k engage- men p ofiles in employees’app aisals o hei pe sonal wo k goal p og ess. P og ess in pe sonal wo k goals e e s o he ex en o which an indi idual pe cei es ha hei meaning ul, pe sonally salien wo k- ela ed goals a e p og essing and will be ealized o e ime (Klug & Maie , 2015; Wiese & F eund, 2005). Thus, he las pa o his in oduc ion will ocus on he pe sonal wo k goal li e a u e. 2. Bu nou and wo k engagemen : concep s and associa ions Job bu nou has se e al defini ions, o which he mos u ilized and p ominen in he wo k psychological li e a u e concep ualizes job bu nou as a wo k- ela ed s ess synd ome comp ising h ee dimen- sions: exhaus ion, cynicism and educed p o essional efficacy (see Lei e , Bakke , & Maslach, 2014;Maslach, Jackson, & Lei e , 1996; Schau eli, Lei e , & Maslach, 2009). Exhaus ion is cha ac e ized by eelings o being deple ed o one’s emo ional and physical esou ces and eelings o a igue; cynicism is cha ac e ized by dis ancing onesel om one’s wo k and holding a nega i e iew o one’s job, and educed p o essional efficacy is cha ac e ized by he belie ha one is incapable o ulfilling one’s asks and esponsibili ies a wo k (Maslach e al., 1996). Se e al defini ions ha e also been offe ed o wo k engagemen (see e.g., Mäkikangas e al., 2012). The mos commonly used o hese con- cep ualizes wo k engagemen as a posi i e, ulfilling wo k- ela ed cogni i e-affec i e s a e o mind, cha ac e ized by he dimensions o igo , dedica ion and abso p ion (Schau eli, Salano a, González- Romá, & Bakke , 2002). Vigo e e s o a high le el o ene gy and men al esilience a wo k; dedica ion e e s o a sense o mean- ing ulness and en husiasm owa ds one’s job, and abso p ion is cha - ac e ized by a s a e o ull concen a ion and imme sion in one’s job (Schau eli e al., 2002). As wo k engagemen was concei ed as he opposi e pole o bu nou and as a heo e ical ool o unde s anding wo k- ela ed well-being as a posi i e expe ience (Maslach & Lei e , 1997; Schau eli e al., 2002), hey sha e simila defini ional oo s and hus can be conside ed “ ellow- a ele s”. In he ci cumplex model o employee well-being (Bakke & Oe lemans, 2011; see also Mäkikangas e al., 2015), in ac- co dance wi h hei le els o pleasu e and ac i a ion, bu nou and wo k engagemen a e classified as opposi e s a es o well-being. Tha is, job bu nou is cha ac e ized by low ac i a ion and displeasu e, and wo k engagemen by high ac i a ion and pleasu e. Consequen ly, in he heo e ical sense, bu nou and wo k engagemen a e seen as e- p esen ing he e e se sides o employee well-being. Du ing he las en yea s, he heo e ical in e es has u ned o he co e dimensions o bu nou and wo k engagemen and hei ela ion- ship. In his mo e fine-g ained heo e ical app oach –which is ne e - heless oo ed in empi ical findings –exhaus ion and igo a e seen as he wo poles o he ene gy con inuum, and cynicism and dedica ion as he wo poles o he iden ifica ion con inuum (Deme ou i, Mos e , & Bakke , 2010;González-Romá, Schau eli, Bakke , & Llo e , 2006;Mäkikangas e al., 2012). In p ac ice, posi ioning cynicism and dedica ion as pola opposi es indica es ha employees can hold ei he nega i e o posi i e a i udes, bu no bo h simul aneously, owa ds hei wo k, whe eas he ene gy con inuum indica es ha men al ene gy and esiliency and eelings o a igue canno be expe ienced a he same ime. Empi ical findings ha e la gely suppo ed hese heo e ical pe - spec i es. To al sco es o bu nou and wo k engagemen ha e shown a nega i e co ela ion ( =−0.55 and −0.48) ( o me a-analyses, see Cole e al., 2012; C aw o d e al., 2010)and exis ence o he ene gy and iden ifica ion con inua has been demons a ed (Deme ou i e al., 2010; González-Romá e al., 2006). Howe e , he ela ionships wi hin hese con inua ha e showed somewha diffe ing esul s. Tha is, he iden i- fica ion con inuum has been ound o o m a s ong scale (scaling coefficien 0.50 in h ee samples) whe eas he ene gy con inuum o med only a weak o mode a e scale (coefficien s 0.36, 0.39 and 0.40) (González-Romá e al., 2006). Simila conclusions could also be d awn om confi ma o y ac o analysis (Deme ou i e al., 2010). These CFA esul s showed ha cynicism and dedica ion o med a bipola dimen- sion, and hus ep esen ed he opposi e ends o he same con inuum. The esul s o he ene gy dimension, howe e , imply ha exhaus ion and igo ep esen ed mo e independen s a es, and hus no pu e op- posi es. Tha is, al hough hey we e highly nega i ely co ela ed, sug- ges ing ha employees sco ing high on exhaus ion, ypically sco e low on igo , hey we e no mu ually exclusi e in o he combina ions. Consequen ly, he esea ch e idence sugges s ha he ene gy-iden ifi- ca ion dimension is a aluable ool in bu nou -wo k engagemen e- sea ch, and hus i was also used in he p esen s udy. Howe e , he c oss-sec ional and a iable-o ien ed in es iga ions p esen ed abo e a e no in o ma i e abou he possible longi udinal ela ion be ween bu nou and wo k engagemen , and how i mani es s a he in a-indi idual le el, issues which will be discussed nex . 3. Theo e ical and empi ical pe spec i es o long- e m de elopmen o bu nou and wo k engagemen Despi e decades o esea ch, he empo al and de elopmen al as- pec s o bu nou and wo k engagemen , ha is, whe he hey a e s able o change o e ime, emain con o e sial. Below, longi udinal pe - spec i es on bu nou and wo k engagemen a e fi s p esen ed sepa- a ely, a e which hei possible simul aneous de elopmen is dis- cussed. Theo e ically, bu nou is seen as a p og essi e and dynamic psy- chological esponse o job s ess (Maslach, 1982; Shi om, 2003). Con- sequen ly, change pa e ns indica ing a cons an a e (i.e., linea in- c ease o dec ease), a a iable a e (i.e., cu ilinea pa e n) o displaying combina ions o a es o change and di ec ion a e all heo- e ically possible (see Dun o d, Shipp, Boss, Ange meie , & Boss, 2012). The empo al p ocess could also mani es as s abili y. Such empo al a ie y has been demons a ed by bu nou ajec o y s udies (E olah i, Hul ell, & Collins, 2013;Hul ell, Melin, & Gus a sson, 2013; Mäkikangas e al., 2012;Rudman & Gus a sson, 2011). Howe e , a ecen sys ema ic e iew mapping he findings o he bu nou ajec o y s udies indica es ha bu nou ypically de elops along a s able pa h. Be ween one- hi d and h ee-qua e s o he indi iduals s udied be- longed o s able bu nou ajec o ies while only a mino i y expe ienced changes in bu nou (Mäkikangas & Kinnunen, 2016). Wo k engagemen , in u n, was o iginally defined as a ela i ely s able affec i e-cogni i e s a e, a he han a momen a y expe ience A. Mäkikangas e al. Bu nou Resea ch 5 (2017) 44–54 45 o a s a e dependen on any pa icula objec , pe son o e en (Schau eli e al., 2002). Rank-o de s abili y findings ha e suppo ed his iew o wo k engagemen ( o a e iew, see Mäkikangas, Kinnunen, Feld , & Schau eli, 2016).A ecen longi udinals udy u - he demons a ed ha , o wo k engagemen , he p opo ion o he a iance o s abili y is la ge han he p opo ion o he a iance o change (Seppälä e al., 2015). Howe e , he mean le els o wo k en- gagemen a e no fixed (Mäkikangas e al., 2016;Seppälä e al., 2015) and dia y esea ch has indica ed day- o-day fluc ua ions in i s le el (Mäkikangas e al., 2014; Xan hopoulou & Bakke , 2012). O e longe ime pe iods, se e al ajec o ies o wo k engagemen ha e been ound (Mäkikangas e al., 2012;Mäkikangas, Schau eli, Tol anen, & Feld , 2013). Howe e , he esul s o hese ajec o y s udies esembled hose o he bu nou ajec o y esea ch: wo k en- gagemen was ypically s able in i s de elopmen , as s abili y was obse ed in 61–68% o he employees s udied, whe eas pa e ns o change occu ed in he emaining one- hi d. While s udies on he sepa a e de elopmen o bu nou and wo k engagemen u ilizing a pe son-cen e ed app oach a e a e, s udies in- es iga ing hei simul aneous de elopmen wi hin indi iduals a e e en a e (Mäkikangas e al., 2012; Mäkikangas e al., 2014). Based on he ci cumplex model (Bakke & Oe lemans, 2011) and heo izing on he ene gy-iden ifica ion con inua (Deme ou i e al., 2010), bu nou and wo k engagemen can be expec ed o be de elopmen ally he opposi e, pa icula ly in he case o cynicism and dedica ion. The findings o he wo p e ious pe son-cen e ed s udies ha e suppo ed hese pe spec i es. Fi s , in he s udy by Mäkikangas e al. (2012), he pa e ns o exhaus ion and igo ( he ene gy con inuum) and cynicism and dedica ion ( he iden ifica ion con inuum) and he associa ions be ween he wo hypo hesized opposi es we e in es iga ed o e wo yea s. The esul s showed no s a is ical dependency be ween he exhaus ion and igo ajec o ies. Howe e , close inspec ion o he esul s demons a ed ha o e one- hi d o he pa icipan s belonged simul aneously o he g oups o low s able exhaus ion and high s able igo (Mäkikangas e al., 2012). Fu he mo e, he esul s e ealed ha he mean le els and changes in cynicism and dedica ion o e ime we e in opposi e di ec ions a he indi idual le el. Fo example, when an employee’s cynicism inc eased o e ime, he le els o dedica ion si- mul aneously dec eased and ice e sa. La e , in a dia y design (Mäkikangas e al., 2014), he de elopmen o exhaus ion and igo was in es iga ed simul aneously du ing one wo king week. This s udy – which canno , howe e , be conside ed a p ope longi udinal s udy – showed ha while day-le el expe iences o exhaus ion and igo we e mainly mu ually exclusi e, a small p opo ion o he esponden s e- po ed expe iencing exhaus ion and igo simul aneously h oughou he wo king week, bu only on a e age le els. Consequen ly, based on bo h he heo e ical (Bakke & Oe lemans, 2011) and empi ical e i- dence (Deme ou i e al., 2010; González-Romá e al., 2006; Mäkikangas e al., 2012; Mäkikangas e al., 2014) p esen ed abo e, i is expec ed ha : Hypo hesis 1. A long- e m p ofile exis s in which he mean le els o exhaus ion and igo (ene gy con inuum) a e in e sely ela ed. This p ofile comp ises he majo i y o employees. Hypo hesis 2. A long- e m p ofile exis s in which he mean le els o cynicism and dedica ion (iden ifica ion con inuum) a e in e sely ela ed. This p ofile comp ises he majo i y o employees. Hypo hesis 3. Long- e m change p ofiles in ene gy and iden ifica ion con inua exis bu hese p ofiles will comp ise a small mino i y o employees. 4. Goal p og ess and he long- e m de elopmen o bu nou and wo k engagemen In line wi h Aus in & Vancou e (1996), we conside pe sonal wo k goals as in e nal ep esen a ions o desi ed s a es, which can be ou - comes, e en s, o p ocesses. These desi ed s a es can ange om bio- logical needs o complex cogni i e aspi a ions (e.g., p omo ion). The wo p ima y app oaches o pe sonal goals a e 1) analysis o goal con- en s and 2) goal app aisals (Li le, 2007). The con en s o pe sonal goals desc ibe an indi iduals’o ien a ion owa ds he u u e, e ealing hei wan s, wishes, conce ns and he in en ions. In he wo k con ex , employees’pe sonal wo k goals ha e been ound o be ela ed o hei compe ence, p og ession, well-being a wo k, job change, job secu i y, finance and success o he o ganiza ion (Hy önen, Feld , Salmela-A o, Kinnunen, & Mäkikangas, 2009). Goal app aisals, in u n, e e o cog- ni i e and affec i e dimensions such as p og ession, impo ance, commi men , o difficul y o goal con en s ( o a e iew, see Aus in & Vancou e , 1996). Bo h he con en s and app aisals o pe - sonal wo k goals ha e shown associa ions wi h wo k- ela ed well-being and a i udes, e en a e con olling o ac o s ela ed o he psycho- social wo k en i onmen (Hy önen e al., 2009; Maie & B uns ein, 2001; Pomaki e al., 2004; e Does Maes, Gebha d , & Koelewijn, 2006), hus demons a ing hei ele ance in he occupa ional heal h psychology con ex . In his s udy, we ocused on employees’goal app aisals, and pa i- cula ly on goal p og ess, since i has shown a s onge associa ion wi h low bu nou and high wo k engagemen (Hy önen e al., 2009), and wi h subjec i e well-being ( o a e iew, see Klug & Maie , 2015), han he o he dimensions o goal app aisal. As he con en o pe sonal wo k goals ha e shown change o e ime (Hy önen, Feld , Tol anen, & Kinnunen, 2010), app aisals o pe sonal wo k goal p og ess offe a mo e meaning ul esea ch a ge in he long- e m pe spec i e adop ed he e. P e ious esea ch has also epo ed ha , besides a lowe le el o bu nou and highe le el o wo k engagemen , posi i e app aisals o goal p og ess we e ela ed o highe job sa is ac ion (Maie & B uns ein, 2001; Wiese & F eund, 2005), o ganiza ional com- mi men (Maie & B uns ein, 2001) and posi i e affec (Wiese & F eund, 2005). These findings a e in line wi h he social cogni i e ca ee heo y which pos ula es ha goals and goal-di ec ed beha io s a e likely o lead o job and li e sa is ac ion (Len & B own, 2006; Len & B own, 2008). Goal-di ec ed beha io no only p o ides s uc u e and mean- ing ul ac i i ies in li e bu also b ings people in o con ac wi h o he s. Mo e specifically, he social cogni i e ca ee heo y sugges s ha as- se ing pe sonal agency h ough goal se ing, being commi ed o hese goals, and making p og ess in one’s pe sonally salien goals a e cen al in p omo ing wo k and li e sa is ac ion (Len & B own, 2008). Howe e , he di ec ion o he ela ionship be ween p og ess in pe sonal wo k goals and wo k- ela ed well-being is no heo e ically clea cu , as hei associa ion can be explained in wo ways. On he one hand, a o able app aisals o goal p og ess may sus ain well-being a wo k because an employee who p og esses owa ds he a ainmen o pe sonal goals has less o p ocess cogni i ely and emo ionally, less conce ns abou unfinished asks, and ewe disc epancies in goal a - ainmen (Sy ek & An oni, 2014;Sy ek, Weigel , Pei e , & An oni, 2017). All hese expe iences dec ease umina i e hough s and hus os e he employee’s well-being a wo k. On he o he hand, he as- socia ion be ween goal p og ess and well-being may wo k he o he way a ound. Lowe well-being (e.g., high bu nou , low wo k engage- men ) may ax an employee’s ene gy and mo i a ion o pu sue his o he pe sonal wo k goals, which in u n hinde s p og ess owa ds alued pe sonal wo k goals. The e o e, we in es iga ed he diffe ences be- ween bu nou -wo k engagemen p ofiles in employees’app aisals o hei pe sonal wo k goal p og ess a each measu emen poin wi hou posi ing any causal di ec ion be ween hem. Based on he heo e ical (Len & B own, 2006; Len & B own, 2008) and empi ical esea ch on goal p og ess (e.g., Hy önen e al., 2009; Maie & B uns ein, 2001), we expec ed ha : Hypo hesis 4. Employees in he p ofiles ep esen ing high well-being A. Mäkikangas e al. Bu nou Resea ch 5 (2017) 44–54 46 in he ene gy and iden ifica ion con inua will epo posi i e app aisals o goal p og ess ac oss he s udy pe iod. Hypo hesis 5. Employees in he p ofiles ep esen ing changes in well- being in he ene gy and iden ifica ion con inua will show also changes in goal p og ess; i.e., hose in he inc easing well-being p ofiles app aisals o goal p og ess will show an inc ease whe eas hose in he dec easing well-being p ofiles app aisals o goal p og ess will show a dec ease. 5. Me hod 5.1. Pa icipan s and p ocedu e This ques ionnai e s udy u ilized fi e measu emen poin s: sp ing 2006 (T1), sp ing 2008 (T2), sp ing 2010 (T3), sp ing 2012 (T4) and sp ing 2014 (T5). The sample was selec ed om he membe ship eg- is e s o wo Finnish na ional ade unions (T ade Union P o and he Union o P o essional Enginee s). The o iginal sample con ained 1 904 membe s, all o whom we e below age 36 and whose p o essional i le indica ed a managemen o leade ship posi ion. The ini ial a ge o he esea ch p ojec was o in es iga e young manage s’occupa ional well- being in he ea ly s ages o hei ca ee s. Ques ionnai es we e sen o pa icipan s’home add esses and 933 we e e u ned. The 186 e- sponden s who u ned ou no o be in managemen o in employmen we e excluded om he final sample (N= 747). The esponse a e was 43.4% in 2006 ( o mo e de ail and a i ion analyses, see Feld e al., 2016; Hy önen e al., 2009). A T2 in 2008, ques ionnai es we e sen o 621 o he 747 pa ici- pan s who had pa icipa ed a T1. The emaining 126 pa icipan s had indica ed a T1 ha hey wished o wi hd aw om he esea ch and we e he e o e excluded om he T2 ollow-up. A T2, 433 ques- ionnai es we e e u ned, yielding a esponse a e o 69.7% o he a ailable sample and 58.0% o he ini ial 2006 sample (N= 747). A T3 in 2010, as 26 pa icipan s had declined u he pa icipa ion a T2, ques ionnai es we e sen o 595 pa icipan s, o whom 380 esponded, yielding a esponse a e o 63.9% o he a ailable sample and 50.8% o he ini ial sample. A T4 in 2012, as 20 pa icipan s had indica ed a T3 ha hey no longe wished o be con ac ed, ques ionnai es we e sen o 575 pa icipan s, o whom 333 esponded, yielding a esponse a e o 57.9% o he a ailable sample and 44.6% o he ini ial sample. A T5, as 13 pa icipan s had indica ed a T4 ha hey no longe wished o be con ac ed, he ques ionnai e was sen o 562 pa icipan s, o whom 289 pa icipan s esponded, yielding a esponse a e o 51.4% and 38.7% o he ini ial sample. The p esen pa icipan s (n=168; 85.1% men) include all hose esponden s who we e employed and esponded o bo h he bu nou and wo k engagemen scales a each o he fi e measu emen s. Pa icipan s we e on a e age 31.1 yea s old when he s udy s a ed a T1 ( ange 25–36 yea s; SD = 3.2 yea s). A T1, he majo i y had a lowe uni e si y deg ee in enginee ing (67.1%) and 2.3% had a highe uni e si y deg ee. Technicians o med 6.6% o he pa icipan s and 22.8% had o he echnical aining. Only 1.2% o pa icipan s had no p o essional aining. The majo i y o pa icipan s (94%) we e in pe - manen employmen . A T1, 6.7% we e in op managemen , 53.7% in middle managemen and 39.6% in lowe managemen . A T5, 69% o pa icipan s epo ed con inuing o wo k in a managemen o leade - ship posi ion. 5.2. Measu es Bu nou was measu ed using he Be gen Bu nou Indica o (Nää änen, A o, Ma hiensen, & Salmela-A o, 2003; see also Feld e al., 2014), which is based on he same defini ion o job bu nou as he MBI (Maslach e al., 1996). The p esen s udy used he 9-i em e sion, which has been shown o ha e be e alidi y han he o iginal 15-i em scale (Salmela-A o, Ran anen, Hy önen, Tillemann, & Feld , 2011). The BBI-9 consis s o h ee subscales; he e, howe e , only he dimensions measu ing exhaus ion ( h ee i ems; e.g., “I am snowed unde wi h wo k”) and cynicism ( h ee i ems; e.g., “I eel ha I ha e g adually less o gi e”) we e used. Responses we e a ed on a six-poin esponse scale anging om 1 (comple ely disag ee) o6(comple ely ag ee). Wo k engagemen was measu ed wi h he U ech Wo k Engagemen Scale-sho o m (UWES-9; Schau eli, Bakke , & Salano a, 2006). In his s udy, he h ee i ems measu ing igo (e.g., “A my job, I eel bu s ing wi h ene gy”) and he h ee measu ing dedica ion (e.g., “My job in- spi es me”) we e used. The i ems we e sco ed on a 7-poin a ing scale (1 = ne e ,7=daily). Pe sonal wo k goals we e asked a each measu emen ime wi h an open-ended i em: “W i e down you mos impo an pe sonal goal e- la ing o you wo k o ca ee ”(Hy önen e al., 2009; Hy önen e al., 2010). The i em was ollowed by a blank space whe e pa icipan s could w i e down hei goal. The w i en goals we e ela ed o com- pe ence, p og ession, well-being, job change, job secu i y, o ganiza ion, and finance a each measu emen poin (Anonen & Hemilä, 2017; Hy önen e al., 2009;Hy önen e al., 2010; Hy önen e al., 2016). Nex , again a each measu emen poin , pa icipan s we e asked o app aise hei goal p og ess wi h he single-i em ques ion “How a ha e you p og essed in eaching his goal?”(see Hy önen e al., 2009). Answe s we e gi en on a 5-poin scale om 1 (no a all) o5( e y much). The app aisal o pe sonal wo k goal p og ess was analyzed in his s udy. The p esen s udy ocused solely on app aisals o goal. The co ela ions be ween he s udy a iables oge he wi h he C onbach alphas and desc ip i e in o ma ion on he a iables a e p e- sen ed in Table 1. 5.3. A i ion analyses The baseline a i ion analysis showed ha , a T1, pa icipan s (N= 747) did no diffe in gende om non- esponden s (N= 971), χ 2 (1) = 0.70, ns. ( o mo e de ails, see Hy önen e al., 2009). Da a on he age o non- esponden s was only a ailable o he membe s o T ade Union P o; hese esponden s (n= 331) did no diffe in age om non- esponden s (n= 379) as de e mined by independen samples - es , (708) = 1.53, ns. Nex , he final pa icipan s (n=168) we e compa ed o he pa icipan s who did no comple e all he ques ionnai es (n=579) in backg ound a iables (gende , age) and in he a iables o exhaus ion, cynicism, igo and dedica ion a T1. The χ 2 es s indica ed no diffe ences in he gende dis ibu ion, χ 2 (1) = 0.03, ns. Simila ly, he independen - es indica ed no diffe ences in age, (738) = 0.17, ns.Nodiffe ences we e ound in ela ion o ei he exhaus ion, (740) = 0.01, ns, o cynicism, (740) = 1.17, ns. Sligh ly highe alues o igo we e epo ed among pa icipan s (M= 5.71) han non-pa ici- pan s (M= 5.51) a T1, (742) = 4.52, p<0.05, bu no diffe ences we e ound o dedica ion, (742) = 0.35, ns. 5.4. S a is ical analysis La en p ofile analysis (LPA), which is a ype o fini e mix u e modeling (Mu hén & Mu hén, 1998–2012;S e ba, 2013), was used o in es iga e p ofiles based on he le els o and changes in bo h bu nou and wo k engagemen om Time 1 o Time 5. LPA iden ifies la en p ofiles (e.g., subpopula ions) om he obse ed da a and es ima es he pa ame e s o hese p ofiles (Mu hén & Mu hén, 1998–2012). In his s udy, he pa ame e s o he p ofile solu ions we e es ima ed using he maximum likelihood es ima ion wi h obus s anda d e o s (MLR) (Mu hén & Mu hén, 1998–2012). The wi hin-p ofile model was speci- fied so ha a iances o bu nou and wo k engagemen we e con- s ained o be equal ac oss he p ofiles. Fu he mo e, bu nou and wo k engagemen we e allowed o co ela e wi h one ano he . The LPAs we e pe o med using Mplus ( e sion 7) (Mu hén & Mu hén, 1998–2012). A. Mäkikangas e al. Bu nou Resea ch 5 (2017) 44–54 47 Va ious c i e ia we e used o de e mine he adequa e numbe o la en p ofiles (Mu hén, 2003; Nylund e al., 2007): (a) he Akaike In- o ma ion C i e ion (AIC); (b) he Bayesian In o ma ion C i e ion (BIC); (c) he Boo s ap Likelihood Ra io Tes (BLRT). The model wi h he smalles AIC and BIC alues is conside ed o be supe io . The BLRT es compa es solu ions wi h diffe en numbe s o la en p ofiles. Fo he es , a p- alue lowe han 0.05 sugges s ha he solu ion wi h kp ofiles fi s he da a be e han he solu ion wi h k-1 p ofiles. Fu he mo e, he dis inc i eness o he p ofiles was assessed using en opy and a e age la en class pos e io p obabili ies (A ePP). En opy illus a es how accu a e he o e all classifica ion is, and A ePP e alua es he ce ain y o placing an obse a ion in a pa icula class using pos e io p ob- abili ies. Using he mos likely la en membe ship, A ePP is calcula ed o each o he classes. Bo h he en opy and A ePP alues ange om 0 o 1; o en opy, alues close o 1 indica e a clea classifica ion (Celeux & So omenho, 1996). Fo he cases in he mos likely la en class, an A e PP o g ea e han.70 indica es ha he solu ion ound can be in e p e ed using he mean ajec o ies (Nagin, 2005). When de- e mining he numbe o p ofiles, he heo e ical in e p e abili y o he solu ion was also conside ed. A Gene al Linea Model (GLM) o epea ed measu es was used o es whe he he bu nou -wo k engagemen p ofiles diffe ed in pe sonal wo k goal p og ess ac oss ime. In hese analyses, he p ofile solu ion o he exhaus ion- igo /cynicism-dedica ion was ea ed as a fixed ac o and ime as a epea ed measu e. The GLM and p elimina y analyses, i.e., he means, s anda d de ia ions, co ela ions and eliabili ies (C onbach’s alphas) o he s udy a iables, we e pe o med using IBM SPSS S a is ics 22. 6. Resul s 6.1. P ofiles o bu nou and wo k engagemen The esul s o he la en p ofile analysis o he simul aneously es- ima ed bu nou and wo k engagemen p ofiles a e p esen ed in Table 2. As shown, o he ene gy con inuum (i.e., exhaus ion- igo ), he h ee-p ofile solu ion was suppo ed by he BIC alue, which has p o en o be he mos consis en goodness-o -fi indica o o la en p ofiles (Nylund e al., 2007; Mu hén, 2006). In addi ion, bo h he en opy alue (.98) and he A ePPs ( ange.98–.99) we e e y high, il- lus a ing he dis inc i eness o he p ofiles in he h ee-p ofile solu ion. The same solu ion was also eplica ed, indica ing ha he solu ion was global ins ead o local (see Jung & Wick ama, 2008). The ou -p ofile solu ion, which also showed a good en opy alue, did no , howe e , eplica e and one o he ou p ofiles was small, comp ising only 1% o he pa icipan s. The e o e, he h ee-p ofile solu ion was chosen o he subsequen analyses. Fig. 1 shows he h ee p ofiles o simul aneously es ima ed exhaus ion and igo a he fi e measu emen poin s. P ofile 1, labeled “Low s able exhaus ion –high s able igo ”, was he la ges o he p ofiles and con ained 84% o he pa icipan s (A ePP = .99). This p ofile showed s able low le els o exhaus ion and s able high le els o igo ac oss he fi e measu emen s. P ofile 2, i led “Fluc ua ing exhaus ion and igo ”(A ePP = .98), accoun ed o 11% Table 1 Co ela ions o he S udy Va iables. MSD123456789101112131415161718192021222324 1. Ex 1 2.86 1.03 .70 2. Ex 2 2.86 1.08 .38 .71 3. Ex 3 2.84 1.06 .33 .61 .71 4. Ex 4 2.82 1.06 .32 .50 .57 .70 5. Ex 5 2.83 1.16 .40 .56 .53 .56 .77 6. Cy 1 2.50 0.91 .52 .34 .26 .24 .29 .82 7. Cy 2 2.55 1.11 .31 .54 .34 .19 .37 .69 .83 8. Cy 3 2.62 1.18 .21 .41 .54 .22 .23 .45 .54 .86 9. Cy 4 2.46 1.17 .15 .34 .35 .44 .25 .39 .32 .47 .84 10. Cy 5 2.44 1.19 .17 .28 .25 .14 .38 .33 .40 .42 .47 .86 11. Vi 1 5.71 0.97 -.22 -.03 .02 .04 -.01 -.43 -.21 -.10 -.10 -.20 .82 12. Vi 2 5.79 0.98 -.15 -.23 -.13 -.01 -.24 -.36 -.57 -.24 -.24 -.40 .42 .86 13. Vi 3 5.66 1.12 -.11 -.20 -.25 -.07 -.08 -.27 -.31 -.54 -.44 -.33 .39 .52 .88 14. Vi 4 5.54 1.26 -.09 -.13 -.15 -.17 -.08 -.23 -.21 -.31 -.68 -.39 .34 .46 .68 .90 15. Vi 5 5.53 1.26 -.10 -.21 -.16 -.13 -.27 -.21 -.27 -.26 -.40 -.72 .35 .52 .52 .62 .91 16. De 1 5.64 1.16 -.15 .04 .02 .10 .04 -.44 -.20 -.08 -.05 -.12 .79 .36 .29 .27 .22 .87 17. De 2 5.78 1.15 -.07 -.24 -.13 -.02 -.19 -.32 -.57 -.27 -.23 -.34 .33 .86 .49 .43 .45 .33 .89 18. De 3 5.58 1.24 -.06 -.18 -.22 -.06 -.07 -.26 -.32 -.56 -.47 -.37 .35 .48 .85 .63 .48 .32 .53 .91 19. De 4 5.49 1.45 -.08 -.15 -.18 -.16 -.10 -.29 -.26 -.37 -.73 -.45 .29 .45 .67 .90 .62 .29 .46 .70 .93 20. De 5 5.45 1.40 -.06 -.10 -.06 -.03 -.13 -.22 -.24 -.20 -.38 -.71 .42 .53 .49 .61 .88 .35 .500 .54 .66 .94 21. P og. 1 3.48 0.92 -.12 -.05 .08 .06 .05 -.29 -.14 -.08 -.09 -.09 .38 .18 .22 .15 .05 .33 .13 .18 .18 .07 – 22. P og. 2 3.47 0.85 -.04 .00 .03 .07 .03 -.09 -.13 -.01 -.09 -.11 .22 .28 .24 .25 .20 .13 .32 .22 .23 .28 .07 – 23. P og. 3 3.34 0.89 -.02 -.10 -.20 -.06 .01 -.24 -.25 -.29 -.37 -.20 .13 .28 .43 .40 .19 .10 .32 .43 .44 .20 .31 .32 – 24. P og. 4 3.46 0.99 .01 -.09 -.01 -.05 -.11 -.16 -.19 -.21 -.46 -.32 .23 0.35 .42 .49 .35 .19 .32 .47 .52 .39 .32 .24 .29 – 25. P og. 5 3.48 1.01 -.08 -.01 -.03 -.01 -.10 -.16 -.06 -.05 -.29 -.36 .23 0.31 .31 .42 .45 .16 .26 .28 .39 .46 .11 .27 .41 .35 No e. = .15 −.19, p< .05; = 0.20–.26, p< .01; > .27, p<.001. Ex = Exhaus ion, Cy = Cynicism, Vi = igo , De = Dedica ion, P og. = P og ess in wo k goals. The numbe s 1–5 e e o he measu emen imes. The C onbach’s alphas a e shown on he diagonal (bold ace). Table 2 Fi Indices o La en P ofile Models o Bu nou and Wo k Engagemen wi h Diffe en Numbe s o La en P ofiles. Numbe o logL AIC BIC En opy BLRT La en p ofile p opo ionsla en p ofiles p alue Exhaus ion- igo 1−2189.6 4509.3 4712.3 ––168 2−2150.7 4453.4 4690.9 0.97 0.001 149/19 3−2122.1 4418.2 4690.0 0.98 0.001 141/19/8 4−2097.6 4391.2 4697.3 0.98 0.001 137/19/9/3 Cynicism-dedica ion 1−2188.3 4506.6 4709.7 ––168 2−2150.6 4453.2 4690.7 0.96 0.001 143/25 3−2119.4 4412.9 4684.7 0.96 0.001 27/124/17 4−2095.3 4386.7 4692.8 0.98 0.001 133/23/6/6 No e. logL = Log likelihood; AIC = Akaike In o ma ion C i e ion; BIC = Bayesian in- o ma ion c i e ion; BLRT = Boo s ap-Likelihood-Ra io Tes . A. Mäkikangas e al. Bu nou Resea ch 5 (2017) 44–54 48 1 2 3 4 5 6 7 Exhaus ion Vigo Time 1 Time 2 Time 3 Time 4 Time 5 P o ile 1: “Low s able exhaus ion - high s able igo ” (n= 141, 84%) 1 2 3 4 5 6 7 Exhaus ion Vigo Time 1 Time 2 Time 3 Time 4 Time 5 P o ile 2: “Fluc ua ing exhaus ion and igo ” (n= 19, 11%) 1 2 3 4 5 6 7 Exhaus ion Vigo Time 1 Time 2 Time 3 Time 4 Time 5 P o ile 3: “S able a e age exhaus ion - dec easing igo ” (n= 8, 5%) Fig. 1. Es ima ed means o he la en p ofiles o exhaus ion and igo . No e. Scale o exhaus ion: 1 = comple ely disag ee, 6 = comple ely ag ee; o igo : 1 = ne e , 7 = daily. A. Mäkikangas e al. Bu nou Resea ch 5 (2017) 44–54 49 1 2 3 4 5 6 7 Cynicism Dedica ion Time 1 Time 2 Time 3 Time 4 Time 5 P o ile 1: “Low s able cynicism - high s able dedica ion” (n= 124, 74%) 1 2 3 4 5 6 7 Cynicism Dedica ion Time 1 Time 2 Time 3 Time 4 Time 5 P o ile 2: “Inc easing cynicism - dec easing dedica ion” (n= 27, 16%) 1 2 3 4 5 6 7 Cynicism Dedica ion Time 1 Time 2 Time 3 Time 4 Time 5 P o ile 3: “Dec easing cynicism - inc easing dedica ion” (n= 17, 10%) Fig. 2. Es ima ed means o he la en p ofiles o cynicism and dedica ion. No e. Scale o cynicism: 1 = comple ely disag ee, 6 = comple ely ag ee; o dedica ion: 1 = ne e , 7 = daily. A. Mäkikangas e al. Bu nou Resea ch 5 (2017) 44–54 50 o he pa icipan s. In his p ofile, he le els o exhaus ion we e low, bu sligh ly fluc ua ing (i.e., inc easing o dec easing depending on he measu emen s). Vigo was a mean le els, bu also o some ex en un- s able. P ofile 3, labeled “S able a e age exhaus ion –dec easing igo ” (A ePP = .99), comp ised 5% o he pa icipan s. This p ofile showed a conside able dec ease in igo du ing he ollow-up, as he le els o exhaus ion –which we e highe han in he o he wo p ofiles – e- mained a he s able o e ime. Fo he iden ifica ion con inuum, i.e., o simul aneously es ima ed cynicism and dedica ion, he c i e ia used also suppo ed a h ee-p ofile solu ion (see Table 2). Fi s , and mos impo an ly, he h ee-p ofile solu ion was suppo ed by he BIC alue. The en opy alue (.96), o- ge he wi h he A ePPs ( ange.96–.99), we e also high, indica ing good classifica ion quali y and accu acy. The ou -p ofile solu ion was no us wo hy as he bes log likelihood alue was no eplica ed, im- plying p oblems in con e gence (Jung & Wick ama, 2008). (Jung & Wick ama, 2008). In ligh o all his e idence, he h ee- p ofile solu ion was selec ed o he subsequen analyses. Fig. 2 shows he esul s o he h ee-p ofile solu ion in mo e de ail. P ofile 1, i led “Low s able cynicism –high s able dedica ion”was he la ges o he h ee cynicism-dedica ion p ofiles and con ained 74% o he pa icipan s (A ePP = .99). These esponden s showed low s able le els o cynicism and high s able le els o dedica ion. P ofile 2, labeled as “Inc easing cynicism –dec easing dedica ion”(A ePP = .96), in- cluded 16% o he pa icipan s. These pa icipan s showed an inc ease in cynicism and a simul aneous dec ease in dedica ion o e ime. P o- file 3, named “Dec easing cynicism –inc easing dedica ion” (A ePP = .99) comp ised 10% o he pa icipan s and showed de el- opmen o e ime con a y o ha in P ofile 2: cynicism dec eased and dedica ion inc eased. To summa ize, ou Hypo heses 1 and 2 we e suppo ed since he le els o exhaus ion and igo we e in e sely ela ed, as also we e he le els o cynicism and dedica ion. Hypo hesis 3 was also suppo ed as long- e m changes in bo h he ene gy and iden ifica ion con inua long- e m changes we e de ec ed, al hough only among a mino i y o em- ployees. 6.2. P ofiles o bu nou and wo k engagemen : diffe ences in goal p og ess GLM o epea ed measu es was used o es whe he he bu nou - wo k engagemen p ofiles diffe ed by p og ess in pe sonal wo k goals ac oss ime. The GLM esul s a e p esen ed in Table 3. As shown, he in e ac ion effec be ween he la en exhaus ion- igo p ofiles and ime did no each significance, howe e , bo h he p ofile and ime effec s we e significan . Consequen ly, he p ofiles diffe ed significan ly in goal p og ess a he o e all mean le el. This diffe ence was seen a each measu emen poin : employees in he p ofiles “Fluc ua ing exhaus ion and igo ”and “S able a e age exhaus ion –dec easing igo ” epo ed less p og ess in hei pe sonal wo k goals han hose in he p ofile “Low s able exhaus ion –high s able igo ”(see Table 3). The ime effec indica ed ha a s a is ically significan dec ease in goal p og ess oc- cu ed be ween Time 1 and Time 3 a he o e all mean le el. In e - es ingly, howe e , he mean le els o goal p og ess dec eased among employees in he “S able a e age exhaus ion –dec easing igo ”p o- file, while emaining ela i ely s able among hose in he o he wo p ofiles (see Table 3). Ne e heless, he ela i ely la ge mean le el changes we e no s a is ically significan , p obably owing o he small size o he p ofile (n= 8) in compa ison wi h he o he p ofiles. The in e ac ion effec be ween he cynicism-dedica ion p ofiles and ime was significan , sugges ing diffe ences be ween he p ofiles in hei changes in goal p og ess o e ime (see Table 3). Among em- ployees in he p ofile “Inc easing cynicism –dec easing dedica ion”, goal p og ess showed a dec easing end. Con e sely, in he p ofile “Dec easing cynicism –inc easing dedica ion”, an inc ease in goal p og ess was obse ed, especially be ween Time 3 and Time 5. In u n, employees in he p ofile “Low s able cynicism –high s able dedica ion” epo ed high le els in goal p og ess ac oss he s udy pe iod. To summa ize, Hypo hesis 4 was suppo ed, as p og ess in pe sonal wo k goals was associa ed wi h high s able le els o bo h he ene gy and iden ifica ion con inua. Hypo hesis 5 was pa ially suppo ed, as changes in goal p og ess we e e iden only wi h changes in cynicism and dedica ion (i.e., he iden ifica ion con inuum). 7. Discussion This s udy augmen s knowledge on he bu nou -wo k engagemen ela ionship longi udinally on he wi hin-pe son le el. As bu nou and wo k engagemen display bo h dynamic and s a ic quali ies o e a ying ime ames ( o a e iew, see Mäkikangas e al., 2016), he pe son-cen e ed app oach applied he e can yield unique insigh s. Spe- cifically, we ound h ee in a-indi idual de elopmen al p ofiles o bo h he ene gy and iden ifica ion con inua. In line wi h he ea lie esea ch ( o e iews, see Mäkikangas & Kinnunen, 2016; Mäkikangas e al., 2016), employees ypically epo ed s able low le els o ex- haus ion and cynicism and s able high le els o igo and dedica ion h oughou he eigh -yea ime span. This indica es ha he ene gy (i.e., exhaus ion- igo ) and iden ifica ion (i.e., cynicism-dedica ion) con- inua display an in e se ela ionship. Alongside hese s able opposi- ions, so-called a ypical (i.e., ound among a small mino i y o pa ici- pan s) de elopmen al p ofiles we e also obse ed o bo h con inua. Namely, o he iden ifica ion con inuum, wo linea ly changing de- elopmen al p ofiles we e de ec ed in which he le els o and changes in cynicism and dedica ion showed simul aneous, bu in e se changes. Fo he ene gy con inuum, wo a ypical p ofiles we e mainly due o Table 3 Diffe ences in Goal P og ess in he Bu nou -Wo k Engagemen P ofiles: GLM o Repea ed Measu es. Bu nou -wo k engagemen p ofiles Time 1 M (S.E.) Time 2 M (S.E.) Time 3 M (S.E.) Time 4 M (S.E.) Time 5 M (S.E.) P ofile Effec F Time Effec F P ofile × Time Effec F 12.90 *** 3.04* 1.69 ns. 1. Low s able exhaus ion –high s able igo 3.53 (0.08) 3.57 (0.07) 3.42 (0.07) 3.65 (0.08) 3.61 (0.09) > 2, 3 2. Fluc ua ing exhaus ion and igo 3.46 (0.24) 2.80 (0.22) 2.73 (0.22) 2.93 (0.24) 2.86 (0.26) 3. S able a e age exhaus ion – dec easing igo 3.17 (0.38) 3.00 (0.34) 2.50 (0.35) 2.17 (0.38) 2.67 (0.42) 19.67*** 2.18 ns. 2.76** 1. Low s able cynicism –high s able dedica ion 3.62 (0.09) 3.63 (0.08) 3.48 (0.08) 3.72 (0.09) 3.65 (0.09) > 2, 3 2. Inc easing cynicism –dec easing dedica ion 3.29 (0.20) 2.81 (0.18) 2.91 (0.19) 2.67 (0.20) 2.52 (0.21) 3. Dec easing cynicism –inc easing dedica ion 3.07 (0.24) 3.27 (0.21) 2.73 (0.22) 3.20 (0.24) 3.73 (0.25) No e. Pai wise compa isons made by Bon e oni pos hoc es s. * p<.05. ** p<.01. *** p<.001 A. Mäkikangas e al. Bu nou Resea ch 5 (2017) 44–54 51 changing o fluc ua ing le els o igo . Mo eo e , we obse ed mean- ing ul diffe ences in pe sonal wo k goal p og ess be ween he iden ified p ofiles. Goal p og ess was pe cei ed as mos success ul in he g oups desc ibed by he iden ifica ion con inuum, i.e., he g oups o “Low s able cynicism –high s able dedica ion”and “Dec easing cynicism – inc easing dedica ion”. Below, hese findings and hei possible im- plica ion o he bu nou -wo k engagemen li e a u e a e desc ibed in de ail. Al hough ea lie esea ch has ques ioned he an ipodal na u e o exhaus ion and igo (Deme ou i e al., 2010; González-Romá e al., 2006) and shown hei possible co-occu ence on a e age le els (Mäkikangas e al., 2012; Mäkikangas e al., 2014), he p esen findings e ealed ha , among he g ea majo i y o employees (84%), s able low le els o exhaus ion combined wi h high le els o igo we e e iden in he long- e m. A simila an ipodal de elopmen al ela ionship was also obse ed in he p ofile labeled “Fluc ua ing exhaus ion and igo ”, whe e he le els o exhaus ion and igo we e la gely in e sely ela ed and hus, ypically, exclusi e expe iences. This s udy also e ealed a small, a ypical exhaus ion- igo p ofile, i.e., “S able a e age exhaus ion –dec easing igo ”. In his p ofile, he le els o exhaus ion we e sligh ly highe han in o he wo p ofiles, and he le els o igo showed a sha p down u n o e ime. This long- e m de elopmen al p ofile offe s e idence ha e en sligh ly ele a ed, bu p olonged exhaus ion can lead o he e osion o igo . This de elop- men al p ofile was no e iden o e a sho e ime span, i.e., wo yea s (Mäkikangas e al., 2012), sugges ing ha he slow p ocess o ene gy o e consump ion only becomes isible o e a long in e al. This so called isk p ofile, accoun ing o 5% o he pa icipan s, adds o ou unde s anding o he un a o able de elopmen o he ene gy con- inuum in he long e m. I is an example o a pa icula combina ion o exhaus ion and igo , which we we e now able o demons a e em- pi ically in he long- un ia a pe son-o ien ed app oach. The esul s o he iden ifica ion con inuum no only confi med ea lie findings (Deme ou i e al., 2010; González-Romá e al., 2006; Mäkikangas e al., 2012) bu also elabo a ed hem u he . Tha is, among he majo i y o pa icipan s (74%), he le els o cynicism and dedica ion we e s able and in e sely ela ed, and among one- hi d o he pa icipan s hei le els simul aneously changed in he e e se di- ec ion. The e o e, he esul s clea ly showed ha cynicism and ded- ica ion we e mu ually exclusi e well-being s a es; i.e., employees can hold ei he nega i e o posi i e a i udes owa ds hei wo k, bu no bo h. This ela ionship has been es ablished in bo h a iable- (Deme ou i e al., 2010; González-Romá e al., 2006) and pe son-cen- e ed s udies (Mäkikangas e al., 2012) and using a a ie y o bu nou scales (e.g., OLBI, MBI, BBI). Thus, he findings ela ing o he iden i- fica ion con inuum seems o be a he conclusi e. O e all, he de elopmen al p ofiles ob ained he e offe a clea e pic u e o he s abili y and change in bu nou and wo k engagemen han has been epo ed by p e ious long- e m pe son-cen e ed s udies (e.g., E olah i e al., 2013; Hul ell e al., 2013; Mäkikangas e al., 2012; Rudman & Gus a sson, 2011). The e a e many easons o his: fi s , he ime span was longe han p e iously used, in ac long enough o clea changes in well-being o be de ec ed. Second, s udies ha e ypically ocused on employees in he ea ly ca ee s age (Hul ell e al., 2013; Rudman & Gus a sson, 2011), a ime known o be a u bulen in many espec s and also a cause o changes in le els o well-being ( o e iew, see Mäkikangas e al., 2016). The de elopmen al p ofiles iden ified in he p esen s udy cap u ed employees in ea ly mid-ca ee , ha is, be- ween ages 30–40. Thi d, ou s a is ically inno a i e me hod enabled us o in es iga e bu nou and wo k engagemen simul aneously a he wi hin-pe son le el, an app oach which po en ially cla ifies he con en and numbe o he p ofiles ound. P e iously, de elopmen al ends ha e been in es iga ed sepa a ely (Mäkikangas e al., 2012). Ou esul s also showed ha he employees in he diffe en long- e m de elopmen p ofiles o bu nou and wo k engagemen diffe ed in hei p og ess owa ds a ainmen o hei pe sonal wo k goals ac oss he eigh -yea pe iod. P og ess in pe sonal wo k goals was mos closely ela ed o he cons uc s pe aining o he iden ifica ion con inuum. Tha is, p og ess in pe sonal wo k goals was epo ed h oughou he eigh yea s among hose who epo ed low s able le els o cynicism and high s able le els o dedica ion. Mo eo e , lowe goal p og ess was e iden among he pa icipan s in he p ofile “Inc easing cynicism – dec easing dedica ion”, whe eas highe goal p og ess was cha ac e is ic in he p ofile “Dec easing cynicism –inc easing dedica ion. High p o- g ess in pe sonal wo k goals was also e iden in he ene gy con inuum p ofile ‘Low s able exhaus ion –high s able igo ’. O e all, hese esul s sugges ha when employees a e able o ad- ance in hei pe sonally meaning ul s i ings, hey eel mo i a ed and willing o in es e en mo e effo in hei wo k. On he con a y, i hei pe sonal s i ings a e no g ee ed wi h some measu e o success, hey a e likely o ques ion he alue and meaning o wo k, and lose hei wo k mo i a ion. The associa ion be ween pe sonal goal p og ess and well-being is likely o be ecip ocal: p og ess uels well-being, bu well- being also de e mines how much ene gy is in es ed in goal pu sui . I is also likely ha p og ess in pe sonal wo k goals inc eases pe son-job fi , he eby acili a ing posi i e a i udes owa ds wo k, such as he eeling ha wo k is meaning ul ( o a me a-analysis, see K is o -B own, Zimme man, & Johnson, 2005). In sum, p og ess in wo k goals offe s a ui ul pe spec i e om which o unde s and he long- e m de elop- men o bu nou and wo k engagemen . Consequen ly, i is wo h pu pose ully inco po a ing pe sonal wo k goals and p og ess owa ds hei a ainmen in o job app aisals and p o essional de elopmen plans. 7.1. Limi a ions This s udy has i s limi a ions. Fi s , all he pa icipan s we e em- ployed on each measu emen occasion. This selec ion c i e ion may, howe e , ha e biased he findings on he de elopmen o bu nou and wo k engagemen in a mo e posi i e di ec ion and also biased he esul o he a i ion analysis (i.e., igo ous employees we e o e ep esen ed in he longi udinal sample). Second, he p esen sample consis ed mos ly o men who we e ypically employed in manage ial du ies, which clea ly limi s he gene alizabili y o he esul s. Since manage s ypically ha e a high le el o job- ela ed well-being (Kinnunen, Feld , & Mäkikangas, 2008;Mäkikangas e al., 2012), he p esen e- sul s, indica ing e y posi i e de elopmen al ends o wo k engage- men and bu nou , canno be applied o all occupa ional s a us g oups. Hence, he long- e m de elopmen o bu nou and wo k engagemen should also be in es iga ed wi h samples d awn om a wide a ie y o occupa ions and indus ies. The longi udinal de elopmen al p ofiles ound in his s udy also need o be es ed and eplica ed in o he con- ex s, as du ing he s udy pe iod many o he labo ma ke - ela ed changes occu ed. Fo example, an economic down u n began in Fin- land du ing he second hal o 2008 and subsequen ly in ensified. Bu nou -wo k engagemen de elopmen should be also in es iga ed ac oss o he age anges, as he p esen p ofiles we e in es iga ed when he pa icipan s we e aged be ween 30 and 40. Finally, i has o be emembe ed ha he esul s on goal p og ess and de elopmen al p o- files o bu nou -wo k engagemen do no imply causal ela ions; hey only show simul aneous associa ions and pa allel changes. 7.2. Conclusions The main esul s e ified he ea lie esul s by showing ha cyni- cism and dedica ion mani es opposi e s a es o well-being also o e he long e m. The implica ions o p ac ice o he p esen findings hus concu wi h hose o p e ious s udies (Deme ou i e al., 2010; Mäkikangas e al., 2012), indica ing ha alid measu emen s o he iden ifica ion dimension could be made by using a single scale. Ou esul s u he cla ified he de elopmen o he ene gy con inuum: y- pically, exhaus ion and igo ep esen ed opposi e well-being s a es. A A. Mäkikangas e al. Bu nou Resea ch 5 (2017) 44–54 52