<_SYM psyRA_NP JAPP_NP 2012_CD 3_CD >_SYM 
<_SYM Longitudinal_JJ Relationships_NNS Between_IN Core_NP Self-Evaluations_NP and_CC Job_NP Satisfaction_NP >_SYM 
<_SYM Chia-Huei_NP Wu_NP Mark_NP A_NP ._SENT Griffin_NN >_SYM 
Abstract_JJ 
Core_JJ self-evaluations_NNS (_( CSE_NP )_) have_VHP been_VBN proposed_VVN as_IN a_DT static_JJ personality_NN trait_NN that_WDT influences_VVZ individuals�work_NN experiences_NNS ._SENT However_RB ,_, CSE_NNS can_MD also_RB be_VB influenced_VVN by_IN work_NN experiences_NNS ._SENT Based_VVN on_IN the_DT corresponsiveprinciple_NN of_IN personality_NN development_NN ,_, this_DT study_NN incorporated_VVD both_DT dispositional_JJ and_CC contextual_JJ perspectives_NNS to_TO examine_VV longitudinal_JJ reciprocal_JJ relationships_NNS between_IN CSE_NN and_CC job_NN satisfaction_NN ._SENT Longitudinal_JJ data_NNS from_IN 5_CD ,_, 827_CD participants_NNS in_IN the_DT British_NP Household_NP Panel_NP Survey_NP from_IN 1997_CD to_TO 2006_CD were_VBD analyzed_VVN ._SENT A_DT series_NN of_IN structural_JJ equation_NN models_NNS revealed_VVD that_IN/that job_NN satisfaction_NN and_CC the_DT growth_NN of_IN job_NN satisfaction_NN in_IN previous_JJ years_NNS positively_RB predicted_VVD CSE_NN in_IN a_DT later_JJR year_NN ._SENT In_IN turn_NN ,_, CSE_NP contributed_VVD to_TO higher_JJR job_NN satisfaction_NN and_CC growth_NN of_IN job_NN satisfaction_NN in_IN following_VVG years_NNS ._SENT This_DT result_NN shows_VVZ that_IN/that both_DT dispositionaland_JJ contextual_JJ forces_NNS interweave_VVP to_TO shape_VV individuals�_NN self-views_NNS and_CC experiences_NNS over_IN time_NN ._SENT 
Core_JJ self-evaluations_NNS (_( CSE_NN )_) are_VBP individuals_NNS '_POS appraisals_NNS of_IN their_PP$ own_JJ worthiness_NN ,_, effectiveness_NN ,_, and_CC capability_NN (_( Judge_NP ,_, Erez_NP ,_, Bono_NP ,_, &_CC Thoresen_NP ,_, 2003_CD )_) ._SENT CSE_NNS was_VBD proposed_VVN as_IN a_DT latent_JJ personality_NN trait_NN indicated_VVN by_IN several_JJ more_JJR specific_JJ traits_NNS that_WDT were_VBD (_( a_DT )_) evaluation_NN focused_VVD ,_, (_( b_LS )_) fundamental_JJ and_CC basic_JJ ,_, and_CC (_( c_LS )_) broad_JJ and_CC encompassing_VVG (_( Judge_NP ,_, Locke_NP ,_, &_CC Durham_NP ,_, 1997_CD )_) ._SENT Four_CD traits_NNS ,_, including_VVG self-esteem_NN ,_, generalized_VVN self-efficacy_NN ,_, locus_NN of_IN control_NN ,_, and_CC emotional_JJ stability_NN (_( or_CC neuroticism_NN )_) ,_, were_VBD proposed_VVN to_TO meet_VV these_DT criteria_NNS (_( Judge_NP et_NP al_NP ._SENT ,_, 1997_CD ;_: Judge_NP ,_, Locke_NP ,_, Durham_NP ,_, &_CC Kluger_NP ,_, 1998_CD )_) ._SENT 
Consistent_JJ with_IN the_DT dispositional_JJ perspective_NN of_IN personality_NN traits_NNS (_( McCrae_NP et_NP al_NP ._SENT ,_, 2000_CD )_) ,_, most_JJS studies_NNS have_VHP assumed_VVN that_IN/that CSE_NP is_VBZ a_DT relatively_RB stable_JJ personality_NN construct_NN that_WDT influences_VVZ more_JJR variable_JJ outcomes_NNS such_JJ as_IN job_NN satisfaction_NN ,_, job_NN performance_NN ,_, career_NN success_NN ,_, and_CC work_NN stress_NN (_( see_VV Judge_NP ,_, 2009_CD ,_, for_IN a_DT review_NN )_) ._SENT These_DT studies_NNS often_RB emphasize_VVP the_DT role_NN of_IN a_DT self-verification_NN motivation_NN (_( Swann_NP ,_, Rentfrow_NP ,_, &_CC Guinn_NP ,_, 2003_CD )_) to_TO explain_VV why_WRB CSE_NN influences_VVZ work_NN experiences_NNS ._SENT In_IN brief_JJ ,_, CSE_JJ guides_NNS an_DT individual_NN to_TO seek_VV self-consistent_JJ information_NN and_CC employ_VV self-consistent_JJ interpretations_NNS ._SENT People_NNS with_IN higher_JJR CSE_NNS are_VBP more_RBR likely_JJ to_TO attend_VV to_TO and_CC seek_VV positive_JJ feedback_NN from_IN the_DT environment_NN and_CC strive_VVP to_TO obtain_VV positive_JJ experiences_NNS ,_, such_JJ as_IN better_JJR job_NN performance_NN or_CC higher_JJR job_NN satisfaction_NN ,_, to_TO confirm_VV their_PP$ positive_JJ self-view_NN ._SENT 
Although_IN CSE_NN is_VBZ typically_RB conceptualized_VVN as_IN a_DT stable_JJ trait_NN by_IN organizational_JJ researchers_NNS ,_, personality_NN theorists_NNS are_VBP paying_VVG increasing_VVG attention_NN to_TO the_DT way_NN that_IN/that traits_NNS change_VVP over_IN time_NN through_IN interaction_NN with_IN life_NN experiences_NNS (_( Roberts_NP &_CC Mroczek_NP ,_, 2008_CD ;_: Roberts_NP ,_, Walton_NP ,_, &_CC Viechtbauer_NP ,_, 2006_CD )_) ._SENT Evidence_NN for_IN interaction_NN between_IN traits_NNS and_CC contexts_NNS suggests_VVZ that_IN/that work_NN experiences_NNS might_MD also_RB influence_VV CSE_JJ so_RB that_IN/that CSE_NP is_VBZ �subject_VVG to_TO prediction_NN and_CC variation_NN over_IN time�_NN (_( Judge_NP ,_, Hurst_NP ,_, &_CC Simon_NP ,_, 2009_CD ,_, p_NN ._SENT 744_LS )_) ._SENT Beyond_IN the_DT dispositional_JJ perspective_NN ,_, we_PP investigate_VVP two_CD alternative_JJ views_NNS of_IN change_NN in_IN CSE_JJ traits_NNS :_: the_DT contextual_JJ perspective_NN and_CC the_DT corresponsive_JJ perspective_NN ._SENT 
First_RB ,_, the_DT contextual_JJ perspective_NN (_( e.g._FW ,_, Lewis_NP ,_, 2001_CD )_) proposes_VVZ that_IN/that personality_NN is_VBZ shaped_VVN by_IN life_NN and_CC work_NN experiences_NNS ._SENT Supporting_VVG this_DT perspective_NN ,_, Schinkel_NP ,_, van_NP Dierendonck_NP ,_, and_CC Anderson_NP (_( 2004_CD )_) found_VVD that_IN/that CSE_NP declined_VVD for_IN job_NN applicants_NNS who_WP were_VBD rejected_VVN and_CC received_VVN performance_NN feedback_NN to_TO explain_VV the_DT decision_NN ._SENT Studies_NNS from_IN this_DT perspective_NN are_VBP more_RBR likely_JJ to_TO focus_VV on_IN the_DT role_NN of_IN self-enhancement_NN motivation_NN (_( Sedikides_NP &_CC Strube_NP ,_, 1995_CD )_) to_TO explain_VV how_WRB work_NN experiences_NNS provide_VVP a_DT basis_NN for_IN changes_NNS in_IN self-worth_NN ._SENT Specifically_RB ,_, because_IN individuals_NNS tend_VVP to_TO make_VV internal_JJ attribution_NN for_IN success_NN (_( Sedikides_NP &_CC Gregg_NP ,_, 2008_CD )_) ,_, positive_JJ work_NN experiences_NNS ,_, such_JJ as_IN better_JJR job_NN performance_NN or_CC higher_JJR job_NN satisfaction_NN ,_, are_VBP more_RBR likely_JJ to_TO boost_VV their_PP$ positivity_NN about_IN the_DT self_NN and_CC lead_VV to_TO higher_JJR self-evaluations_NNS ._SENT 
Second_RB ,_, the_DT corresponsive_JJ perspective_NN (_( Caspi_NP ,_, Roberts_NP ,_, &_CC Shiner_NP ,_, 2005_CD )_) suggests_VVZ that_IN/that CSE_NN and_CC work_NN experiences_NNS mutually_RB influence_VVP each_DT other_JJ over_IN time_NN ._SENT The_DT dynamic_JJ relationship_NN proposed_VVN in_IN the_DT corresponsive_JJ perspective_NN of_IN personality_NN development_NN suggests_VVZ that_IN/that CSE_NN and_CC work_NN experiences_NNS can_MD mutually_RB influence_VV each_DT other_JJ (_( Caspi_NP et_NP al_NP ._SENT ,_, 2005_CD )_) ._SENT Therefore_RB ,_, both_DT dispositional_JJ and_CC contextual_JJ forces_NNS can_MD work_VV together_RB to_TO shape_VV individuals_NNS '_POS personality_NN and_CC experiences_NNS over_IN time_NN (_( Bandura_NP ,_, 1999_CD )_) ._SENT The_DT corresponsive_JJ perspective_NN implies_VVZ that_IN/that CSE_NNS can_MD be_VB changed_VVN to_TO a_DT moderate_JJ degree_NN through_IN a_DT combination_NN of_IN self-verification_NN and_CC self-enhancement_NN motivations_NNS ._SENT This_DT process_NN is_VBZ in_IN line_NN with_IN the_DT attributional_JJ model_NN of_IN CSE_NN and_CC work_NN experience_NN (_( Judge_NP &_CC Kammeyer-Mueller_NP ,_, 2004_CD )_) and_CC the_DT self-concept_NN enhancing_VVG tactician_NN (_( SCENT_JJ )_) model_NN (_( Sedikides_NP &_CC Strube_NP ,_, 1997_CD )_) of_IN self-evaluation_NN motivations_NNS ._SENT 
This_DT study_NN examined_VVD the_DT relationship_NN between_IN CSE_NN and_CC work_NN experiences_NNS from_IN dispositional_JJ ,_, contextual_JJ ,_, and_CC corresponsive_JJ perspectives_NNS with_IN a_DT longitudinal_JJ design_NN spanning_VVG 10_CD years_NNS ._SENT We_PP used_VVD job_NN satisfaction_NN and_CC its_PP$ growth_NN as_IN indicators_NNS of_IN work_NN experiences_NNS ._SENT Job_NN satisfaction_NN represents_VVZ an_DT overall_JJ appraisal_NN of_IN work_NN experiences_NNS (_( Locke_NP ,_, 1976_CD )_) that_WDT conveys_VVZ the_DT meaning_NN of_IN work_NN success_NN in_IN general_NN for_IN an_DT individual_JJ (_( Erdogan_NP &_CC Bauer_NP ,_, 2005_CD ;_: Judge_NP &_CC Hurst_NP ,_, 2008_CD )_) ._SENT We_PP also_RB included_VVD the_DT growth_NN of_IN job_NN satisfaction_NN because_IN the_DT within-person_NN change_NN in_IN job_NN satisfaction_NN over_IN time_NN indicates_VVZ the_DT progress_NN of_IN work_NN success_NN for_IN an_DT individual_NN and_CC has_VHZ been_VBN used_VVN to_TO indicate_VV an_DT individual_NN 's_POS career_NN success_NN (_( Judge_NP &_CC Hurst_NP ,_, 2008_CD )_) ._SENT 
Our_PP$ study_NN contributes_VVZ to_TO a_DT better_JJR understanding_NN of_IN the_DT malleability_NN of_IN CSE_NN over_IN long_JJ periods_NNS ._SENT Because_IN most_JJS studies_NNS draw_VVP on_IN a_DT dispositional_JJ perspective_NN to_TO investigate_VV CSE_NN ,_, little_RB is_VBZ known_VVN about_IN the_DT relative_JJ importance_NN of_IN the_DT different_JJ causal_JJ processes_NNS that_IN/that we_PP identify_VVP ._SENT For_IN example_NN ,_, of_IN the_DT 76_CD empirical_JJ articles_NNS on_IN CSE_NNS located_VVN in_IN the_DT PsycINFO_NP database_NN with_IN core_JJ self-evaluations_NNS as_IN the_DT search_NN term_NN ,_, 74_CD articles_NNS (_( 97_CD %_NN )_) treated_VVN CSE_NN as_IN a_DT static_JJ trait_NN from_IN the_DT dispositional_JJ perspective_NN ,_, two_CD articles_NNS recognized_VVD the_DT contextual_JJ force_NN in_IN shaping_VVG CSE_NP (_( Judge_NP et_NP al_NP ._SENT ,_, 2009_CD ;_: Schinkel_NP et_NP al_NP ._SENT ,_, 2004_CD )_) ,_, and_CC no_DT empirical_JJ study_NN examined_VVD the_DT reciprocal_JJ relationship_NN between_IN CSE_NN and_CC work_NN experiences_NNS ._SENT To_TO provide_VV a_DT better_JJR understanding_NN of_IN the_DT alternative_NN processes_VVZ ,_, we_PP review_VVP each_DT perspective_NN below_IN and_CC develop_VV hypotheses_NNS about_IN the_DT potential_JJ links_NNS between_IN job_NN satisfaction_NN and_CC CSE_NN over_IN time_NN ._SENT 
Effects_NNS of_IN CSE_NN on_IN Job_NP Satisfaction_NP and_CC Its_PP$ Growth_NN :_: The_DT Dispositional_NP Perspective_NP 
Adopting_VVG the_DT dispositional_JJ perspective_NN ,_, Judge_NP et_CC al_NP ._SENT (_( 1997_LS )_) proposed_VVN CSE_NN as_IN a_DT stable_JJ and_CC fundamental_JJ personality_NN trait_NN that_WDT has_VHZ a_DT profound_JJ impact_NN on_IN job_NN satisfaction_NN ._SENT This_DT dispositional_JJ perspective_NN is_VBZ rooted_VVN in_IN the_DT tradition_NN of_IN trait_NN theories_NNS of_IN personality_NN ,_, which_WDT suggest_VVP that_IN/that personality_NN traits_NNS are_VBP stable_JJ and_CC not_RB influenced_VVN by_IN environments_NNS (_( McCrae_NP et_NP al_NP ._SENT ,_, 2000_CD )_) ._SENT Most_RBS subsequent_JJ studies_NNS of_IN the_DT relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN have_VHP treated_VVN CSE_NN as_IN a_DT dispositional_JJ construct_NN that_WDT is_VBZ assumed_VVN to_TO influence_VV job_NN satisfaction_NN in_IN a_DT unidirectional_JJ way_NN ._SENT Several_JJ studies_NNS support_VVP this_DT argument_NN by_IN showing_VVG that_IN/that CSE_NP can_MD strongly_RB and_CC positively_RB predict_VVP job_NN satisfaction_NN (_( e.g._FW ,_, Dormann_NP ,_, Fay_NP ,_, Zapf_NP ,_, &_CC Frese_NP ,_, 2006_CD ;_: Judge_NP ,_, Bono_NP ,_, &_CC Locke_NP ,_, 2000_CD ;_: Judge_NP et_NP al_NP ._SENT ,_, 1998_CD )_) ,_, including_VVG when_WRB the_DT effects_NNS of_IN Big_NP Five_CD personality_NN and_CC positive_JJ and_CC negative_JJ affectivity_NN are_VBP controlled_VVN (_( Judge_NP ,_, Heller_NP ,_, &_CC Klinger_NP ,_, 2008_CD )_) ._SENT 
Studies_NNS from_IN the_DT dispositional_JJ perspective_NN often_RB stress_VVP how_WRB individuals_NNS with_IN higher_JJR CSE_NNS are_VBP motivated_VVN to_TO sustain_VV their_PP$ level_NN of_IN CSE_NN through_IN a_DT self-verification_NN motivation_NN (_( Swann_NP et_NP al_NP ._SENT ,_, 2003_CD )_) ._SENT Evidence_NN supporting_VVG this_DT striving_VVG for_IN self-verification_NN has_VHZ been_VBN identified_VVN in_IN perception_NN (_( Best_NP ,_, Stapleton_NP ,_, &_CC Downey_NP ,_, 2005_CD ;_: Judge_NP et_NP al_NP ._SENT ,_, 2000_CD ,_, Judge_NP et_CC al_NP ._SENT ,_, 1998_CD )_) and_CC goal_NN processes_NNS (_( Judge_NP ,_, Bono_NP ,_, Erez_NP ,_, &_CC Locke_NP ,_, 2005_CD )_) ,_, which_WDT link_VVP CSE_NN and_CC job_NN satisfaction_NN ._SENT Specifically_RB ,_, people_NNS with_IN higher_JJR CSE_NNS tend_VVP to_TO seek_VV positive_JJ feedback_NN and_CC maintain_VV their_PP$ positive_JJ self-concept_NN by_IN focusing_VVG on_IN the_DT positive_JJ features_NNS of_IN the_DT work_NN environment_NN (_( Swann_NP et_NP al_NP ._SENT ,_, 2003_CD )_) ,_, and_CC perceive_VVP more_RBR intrinsic_JJ job_NN characteristics_NNS (_( i.e._FW ,_, autonomy_NN ,_, task_NN identity_NN ,_, skill_NN variety_NN ,_, task_NN significance_NN ,_, and_CC task_NN feedback_NN ;_: Judge_NP et_NP al_NP ._SENT ,_, 2000_CD ,_, Judge_NP et_CC al_NP ._SENT ,_, 1998_CD )_) and_CC fewer_JJR organizational_JJ constraints_NNS (_( Best_NP et_NP al_NP ._SENT ,_, 2005_CD )_) ._SENT These_DT factors_NNS contribute_VVP to_TO higher_JJR job_NN satisfaction_NN by_IN promoting_VVG intrinsic_JJ motivation_NN at_IN work_NN (_( Hackman_NP &_CC Oldham_NP ,_, 1976_CD )_) and_CC by_IN preventing_VVG stress_NN outcomes_NNS such_JJ as_IN burnout_NNS (_( Hobfoll_NP ,_, 1989_CD )_) ._SENT 
Empirically_RB ,_, past_JJ studies_NNS show_VVP that_IN/that CSE_NP predicts_VVZ job_NN satisfaction_NN when_WRB using_VVG a_DT cross-sectional_JJ design_NN (_( Best_NP et_NP al_NP ._SENT ,_, 2005_CD ;_: Judge_NP et_NP al_NP ._SENT ,_, 1998_CD )_) or_CC an_DT asymmetric_JJ time-lagged_JJ design_NN (_( i.e._FW ,_, CSE_NNS was_VBD measured_VVN at_IN Time_NP 1_CD and_CC job_NN satisfaction_NN was_VBD measured_VVN at_IN Time_NP 2_CD ;_: Judge_NP et_NP al_NP ._SENT ,_, 2005_CD ,_, Judge_NP et_CC al_NP ._SENT ,_, 2000_CD )_) ,_, which_WDT are_VBP limited_VVN in_IN the_DT extent_NN to_TO which_WDT they_PP can_MD test_VV the_DT directional_JJ impact_NN of_IN CSE_NN on_IN job_NN satisfaction_NN ._SENT To_TO provide_VV a_DT stronger_JJR test_NN ,_, our_PP$ first_JJ hypothesis_NN replicates_VVZ previous_JJ research_NN in_IN a_DT symmetric_JJ time-lagged_JJ design_NN (_( i.e._FW ,_, CSE_NN and_CC job_NN satisfaction_NN were_VBD both_DT measured_VVN over_IN time_NN )_) :_: 
Hypothesis_NN 1_CD :_: CSE_NNS will_MD positively_RB predict_VV later_JJR job_NN satisfaction_NN ._SENT 
In_IN addition_NN to_TO the_DT level_NN of_IN job_NN satisfaction_NN ,_, CSE_NP contributes_VVZ to_TO a_DT greater_JJR positive_JJ growth_NN of_IN job_NN satisfaction_NN ._SENT First_RB ,_, people_NNS with_IN higher_JJR CSE_NNS should_MD have_VH more_JJR opportunities_NNS to_TO advance_VV in_IN their_PP$ careers_NNS through_IN better_JJR job_NN performance_NN and_CC stronger_JJR motivation_NN for_IN goal_NN attainment_NN (_( Erez_NP &_CC Judge_NP ,_, 2001_CD )_) ._SENT Also_RB ,_, a_DT self-verification_NN process_NN might_MD encourage_VV people_NNS with_IN higher_JJR CSE_NNS to_TO achieve_VV further_JJR success_NN by_IN selecting_VVG jobs_NNS with_IN more_JJR positive_JJ job_NN characteristics_NNS (_( Judge_NP et_NP al_NP ._SENT ,_, 2000_CD )_) ,_, whereas_IN those_DT lower_JJR in_IN CSE_NP might_MD verify_VV their_PP$ negative_JJ self-views_NNS by_IN staying_VVG in_IN undesirable_JJ job_NN conditions_NNS ._SENT Finally_RB ,_, people_NNS with_IN higher_JJR CSE_NNS tend_VVP to_TO capitalize_VV on_IN their_PP$ advantages_NNS ,_, such_JJ as_IN family_NN socioeconomic_JJ status_NN and_CC academic_JJ achievement_NN ,_, to_TO accelerate_VV the_DT increase_NN of_IN their_PP$ incomes_NNS (_( Judge_NP &_CC Hurst_NP ,_, 2007_CD )_) ._SENT On_IN the_DT basis_NN of_IN these_DT reasons_NNS ,_, people_NNS higher_JJR in_IN CSE_NNS are_VBP more_RBR likely_JJ to_TO accelerate_VV their_PP$ job_NN satisfaction_NN than_IN those_DT lower_JJR in_IN CSE_NP ._SENT In_IN line_NN with_IN these_DT reasons_NNS ,_, Judge_NP and_CC Hurst_NP (_( 2008_CD )_) found_VVD that_IN/that people_NNS with_IN higher_JJR CSE_NNS showed_VVD greater_JJR increase_NN in_IN job_NN satisfaction_NN ,_, pay_VV ,_, and_CC occupational_JJ status_NN over_IN 26_CD years_NNS than_IN those_DT with_IN lower_JJR CSE_NNS ._SENT Hence_RB ,_, we_PP propose_VVP the_DT following_VVG hypothesis_NN to_TO replicate_VV the_DT previous_JJ finding_NN :_: 
Hypothesis_NN 2_CD :_: CSE_NNS will_MD positively_RB predict_VV growth_NN of_IN job_NN satisfaction_NN in_IN later_JJR years_NNS ._SENT 
Effect_NN of_IN Job_NP Satisfaction_NP and_CC Its_PP$ Growth_NN on_IN CSE_NN :_: The_DT Contextual_NP Perspective_NP 
In_IN this_DT section_NN ,_, we_PP adopt_VVP a_DT contextual_JJ perspective_NN to_TO propose_VV that_DT job_NN satisfaction_NN and_CC its_PP$ growth_NN can_MD also_RB influence_VV CSE_NN ._SENT From_IN this_DT perspective_NN ,_, �evaluations_NNS of_IN our_PP$ self-concept_NN are_VBP intimately_RB tied_VVN to_TO our_PP$ environment�_NN (_( Judge_NP ,_, 2009_CD ,_, p_NN ._SENT 61_LS )_) ,_, and_CC over_IN time_NN ,_, work_NN experiences_NNS can_MD influence_VV the_DT development_NN of_IN personality_NN traits_NNS ._SENT Some_DT previous_JJ studies_NNS provide_VVP support_NN for_IN this_DT process_NN in_IN relation_NN to_TO overall_JJ CSE_NNS and_CC the_DT constituent_JJ elements_NNS of_IN CSE_NP ._SENT Gilliland_NP (_( 1994_CD )_) found_VVD that_IN/that individuals_NNS '_POS situational_JJ self-efficacy_NN increased_VVD when_WRB they_PP were_VBD successfully_RB selected_VVN for_IN a_DT job_NN with_IN higher_JJR perceived_VVN procedural_JJ fairness_NN ._SENT Schinkel_NP et_FW al_NP ._SENT (_( 2004_CD )_) found_VVD that_IN/that individuals_NNS '_POS CSE_NNS decreased_VVD when_WRB they_PP were_VBD rejected_VVN for_IN a_DT job_NN but_CC increased_VVD when_WRB they_PP could_MD attribute_VV the_DT rejection_NN to_TO an_DT unfair_JJ selection_NN procedure_NN ._SENT In_IN addition_NN to_TO the_DT evidence_NN for_IN short-term_JJ change_NN in_IN CSE_NP ,_, other_JJ studies_NNS show_VVP that_IN/that life_NN experiences_NNS can_MD influence_VV CSE_NN over_IN a_DT longer_JJR time_NN frame_NN ._SENT Costa_NP ,_, Herbst_NP ,_, McCrae_NP ,_, and_CC Siegler_NP (_( 2000_CD )_) found_VVD that_IN/that people_NNS who_WP reported_VVD their_PP$ work_NN lives_NNS were_VBD getting_VVG worse_JJR over_IN 6_CD years_NNS and_CC people_NNS who_WP were_VBD fired_VVN showed_VVD increased_VVN levels_NNS of_IN neuroticism_NN ,_, an_DT indicator_NN of_IN CSE_NP ._SENT These_DT findings_NNS suggest_VVP that_IN/that CSE_NNS can_MD be_VB influenced_VVN by_IN work_NN experiences_NNS over_IN both_DT short-_NN and_CC long-term_JJ periods_NNS ._SENT In_IN addition_NN ,_, these_DT studies_NNS suggest_VVP a_DT self-enhancement_NN motivation_NN behind_IN the_DT change_NN of_IN CSE_NN (_( e.g._FW ,_, Schinkel_NP et_CC al_NP ._SENT ,_, 2004_CD )_) by_IN proposing_VVG that_IN/that individuals_NNS strive_VVP to_TO enhance_VV their_PP$ self-value_NN and_CC worth_JJ (_( Sedikides_NP &_CC Strube_NP ,_, 1995_CD )_) ._SENT For_IN example_NN ,_, Gilliland_NP and_CC Schinkel_NP et_FW al._FW 's_POS findings_NNS suggest_VVP that_IN/that self-enhancement_NN motivation_NN leads_VVZ people_NNS to_TO attribute_VV positive_JJ experiences_NNS to_TO individual_JJ causes_NNS and_CC attribute_VVP negative_JJ experiences_NNS to_TO external_JJ causes_NNS to_TO increase_VV and_CC protect_VV their_PP$ self-worth_NN ._SENT 
In_IN accordance_NN with_IN the_DT contextual_JJ perspective_NN ,_, we_PP propose_VVP that_IN/that job_NN satisfaction_NN and_CC the_DT growth_NN of_IN job_NN satisfaction_NN can_MD contribute_VV to_TO higher_JJR CSE_NN ._SENT We_PP base_VVP our_PP$ reasoning_NN on_IN a_DT discrepancy-reduction_NN mechanism_NN (_( e.g._FW ,_, Bandura_NP ,_, 1989_CD ;_: Carver_NP &_CC Scheier_NP ,_, 2000_CD ;_: Higgins_NP ,_, 1987_CD )_) ._SENT Specifically_RB ,_, motivational_JJ theories_NNS of_IN discrepancy_NN reduction_NN propose_VVP that_IN/that self-evaluations_NNS are_VBP constructed_VVN by_IN comparing_VVG one_PP 's_POS actual_JJ status_NN with_IN a_DT certain_JJ standard_NN (_( e.g._FW ,_, Bandura_NP ,_, 1989_CD ;_: Carver_NP &_CC Scheier_NP ,_, 2000_CD ;_: Higgins_NP ,_, 1987_CD )_) ._SENT Because_IN job_NN satisfaction_NN denotes_VVZ a_DT discrepancy_NN between_IN current_JJ status_NN and_CC a_DT wanted_JJ status_NN at_IN work_NN (_( e.g._FW ,_, Locke_NP ,_, 1969_CD )_) ,_, individuals_NNS with_IN higher_JJR job_NN satisfaction_NN tend_VVP to_TO perceive_VV a_DT smaller_JJR discrepancy_NN with_IN the_DT self-standard_NN and_CC ,_, thus_RB ,_, greater_JJR enhancement_NN at_IN work_NN compared_VVN with_IN their_PP$ counterparts_NNS ._SENT This_DT discrepancy-reduction_NN mechanism_NN is_VBZ more_RBR likely_RB driven_VVN by_IN a_DT self-enhancement_NN motive_NN ._SENT According_VVG to_TO social_JJ cognitive_JJ theory_NN (_( Bandura_NP ,_, 1989_CD ;_: Wood_NP &_CC Bandura_NP ,_, 1989_CD )_) ,_, individuals_NNS regulate_VV themselves_PP by_IN setting_VVG goals_NNS that_WDT create_VVP a_DT discrepancy_NN from_IN the_DT current_JJ status_NN and_CC then_RB devote_VV their_PP$ effort_NN to_TO accomplish_VV these_DT goals_NNS ._SENT This_DT discrepancy_NN production_NN and_CC reduction_NN process_NN contributes_VVZ to_TO higher_JJR self-value_NN and_CC worth_JJ because_IN it_PP can_MD �increase_VV one_CD 's_POS belief_NN in_IN his_PP$ or_CC her_PP$ capabilities_NNS to_TO perform_VV at_IN yet_RB higher_JJR levels�_NN (_( Tolli_NP &_CC Schmidt_NP ,_, 2008_CD ,_, p_NN ._SENT 693_LS )_) ._SENT 
On_IN the_DT basis_NN of_IN the_DT same_JJ notion_NN ,_, we_PP also_RB hypothesize_VVP that_IN/that stronger_JJR growth_NN of_IN job_NN satisfaction_NN over_IN time_NN will_MD contribute_VV to_TO higher_JJR CSE_NN ._SENT Because_IN a_DT prior_JJ satisfaction_NN level_NN provides_VVZ a_DT reference_NN point_NN for_IN interpreting_VVG the_DT meaning_NN of_IN later_JJR satisfaction_NN level_NN (_( Chen_NP ,_, Ployhart_NP ,_, Thomas_NP ,_, Anderson_NP ,_, &_CC Bliese_NP ,_, 2011_CD )_) ,_, stronger_JJR growth_NN of_IN job_NN satisfaction_NN reflects_VVZ a_DT greater_JJR rate_NN in_IN reducing_VVG the_DT have�want_NP discrepancy_NN and_CC thus_RB a_DT greater_JJR rate_NN in_IN self-enhancement_NN at_IN work_NN (_( Carver_NP &_CC Scheier_NP ,_, 2000_CD )_) ,_, which_WDT then_RB contributes_VVZ to_TO higher_JJR CSE_NN ._SENT To_TO examine_VV the_DT hypothesized_VVN directional_JJ impact_NN of_IN job_NN satisfaction_NN and_CC its_PP$ growth_NN on_IN CSE_NN ,_, we_PP express_VVP our_PP$ hypothesis_NN in_IN a_DT time-lagged_JJ context_NN as_RB follows_VVZ :_: 
Hypothesis_NN 3_CD :_: Job_NN satisfaction_NN will_MD positively_RB predict_VV later_RBR CSE_JJ ._SENT 
Hypothesis_NN 4_CD :_: Growth_NN of_IN job_NN satisfaction_NN will_MD positively_RB predict_VV later_RBR CSE_JJ ._SENT 
Longitudinal_JJ Reciprocal_JJ Relationship_NN Between_IN CSE_NP and_CC Job_NP Satisfaction_NP and_CC Its_PP$ Growth_NN 
We_PP have_VHP elaborated_VVN the_DT relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN and_CC its_PP$ growth_NN from_IN dispositional_JJ and_CC contextual_JJ perspectives_NNS that_WDT together_RB imply_VVP a_DT process_NN of_IN longitudinal_JJ reciprocal_JJ influence_NN ._SENT According_VVG to_TO the_DT idea_NN of_IN triadic_JJ reciprocal_JJ causation_NN (_( Bandura_NP ,_, 1999_CD )_) ,_, it_PP is_VBZ possible_JJ that_IN/that both_DT dispositional_JJ and_CC contextual_JJ forces_NNS dynamically_RB work_VVP together_RB to_TO shape_VV one_CD 's_POS CSE_NN and_CC job_NN satisfaction_NN and_CC its_PP$ growth_NN over_IN time_NN ._SENT This_DT dynamic_NN can_MD be_VB understood_VVN by_IN the_DT corresponsive_JJ principle_NN of_IN personality_NN development_NN ,_, which_WDT suggests_VVZ that_IN/that life_NN experiences_NNS influence_VVP the_DT personality_NN traits_NNS that_WDT lead_VVP people_NNS to_TO find_VV these_DT experiences_NNS in_IN the_DT first_JJ place_NN (_( Caspi_NP et_NP al_NP ._SENT ,_, 2005_CD )_) ._SENT Accordingly_RB ,_, it_PP is_VBZ possible_JJ that_IN/that CSE_NN and_CC job_NN satisfaction_NN will_MD have_VH a_DT longitudinal_JJ reciprocal_JJ relationship_NN ,_, such_PDT that_DT job_NN satisfaction_NN and_CC its_PP$ growth_NN influence_VV the_DT CSE_NNS that_WDT lead_VVP people_NNS to_TO have_VH higher_JJR job_NN satisfaction_NN in_IN the_DT first_JJ place_NN ._SENT 
A_DT reciprocal_JJ relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN and_CC its_PP$ growth_NN is_VBZ implied_VVN in_IN the_DT theoretical_JJ mechanisms_NNS that_WDT are_VBP currently_RB proposed_VVN to_TO link_VV CSE_NN and_CC work_NN experiences_NNS ._SENT In_IN their_PP$ attributional_JJ model_NN ,_, Judge_NP and_CC Kammeyer-Mueller_NP (_( 2004_CD )_) proposed_VVD that_IN/that CSE_NP leads_VVZ to_TO better_JJR performance_NN and_CC success_NN because_IN ,_, at_IN first_JJ ,_, people_NNS with_IN higher_JJR CSE_NNS tend_VVP to_TO construct_VV and_CC embrace_VV favorable_JJ work_NN environments_NNS (_( e.g._FW ,_, Best_JJS et_NP al_NP ._SENT ,_, 2005_CD ;_: Judge_NP et_NP al_NP ._SENT ,_, 1998_CD )_) ,_, set_VVN self-concordant_JJ goals_NNS (_( Judge_NP et_NP al_NP ._SENT ,_, 2005_CD )_) with_IN higher_JJR standards_NNS ,_, and_CC strive_VVP to_TO achieve_VV goals_NNS that_WDT lead_VVP to_TO higher_JJR performance_NN (_( Erez_NP &_CC Judge_NP ,_, 2001_CD )_) ._SENT The_DT experience_NN of_IN success_NN subsequently_RB results_VVZ in_IN higher_JJR CSE_NNS because_IN individuals_NNS with_IN a_DT self-enhancement_NN motivation_NN tend_VVP to_TO make_VV internal_JJ attributions_NNS for_IN their_PP$ successes_NNS (_( Sedikides_NP &_CC Strube_NP ,_, 1997_CD )_) ._SENT Because_IN job_NN satisfaction_NN conveys_VVZ the_DT meaning_NN of_IN work_NN success_NN in_IN general_JJ (_( Erdogan_NP &_CC Bauer_NP ,_, 2005_CD ;_: Judge_NP &_CC Hurst_NP ,_, 2008_CD )_) ,_, this_DT process_NN implies_VVZ a_DT longitudinal_JJ reciprocal_JJ relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN and_CC its_PP$ growth_NN ,_, consistent_JJ with_IN the_DT corresponsive_JJ principle_NN of_IN personality_NN development_NN ._SENT 
The_DT reciprocal_JJ relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN is_VBZ also_RB implied_VVN in_IN the_DT SCENT_JJ model_NN (_( Sedikides_NP &_CC Strube_NP ,_, 1997_CD )_) for_IN self-evaluation_NN motivations_NNS ._SENT Sedikides_NP and_CC Strube_NP (_( 1997_CD )_) proposed_VVD that_IN/that self-evaluation_NN is_VBZ an_DT adaptive_JJ process_NN in_IN which_WDT confirming_VVG a_DT positive_JJ self-concept_NN (_( i.e._FW ,_, self-verification_NN )_) is_VBZ a_DT tactical_JJ step_NN in_IN achieving_VVG self-enhancement_NN ._SENT Accordingly_RB ,_, self-verification_NN and_CC self-enhancement_NN motivations_NNS are_VBP not_RB independent_JJ ._SENT For_IN example_NN ,_, people_NNS with_IN higher_JJR CSE_NNS are_VBP more_RBR likely_JJ to_TO focus_VV on_IN the_DT positivity_NN of_IN the_DT environment_NN ,_, such_JJ as_IN more_JJR positive_JJ job_NN characteristics_NNS (_( Judge_NP et_NP al_NP ._SENT ,_, 1998_CD )_) ,_, which_WDT in_IN turn_NN helps_VVZ them_PP to_TO react_VV to_TO environmental_JJ events_NNS in_IN ways_NNS that_WDT can_MD maximize_VV a_DT positive_JJ view_NN of_IN the_DT self_NN ,_, such_JJ as_IN better_JJR job_NN performance_NN or_CC higher_JJR job_NN satisfaction_NN ._SENT 
In_IN sum_NN ,_, perspectives_NNS from_IN the_DT corresponsive_JJ principle_NN of_IN personality_NN development_NN (_( Caspi_NP et_NP al_NP ._SENT ,_, 2005_CD )_) ,_, the_DT attributional_JJ model_NN of_IN CSE_NN and_CC work_NN experiences_NNS (_( Judge_NP &_CC Kammeyer-Mueller_NP ,_, 2004_CD )_) ,_, and_CC the_DT SCENT_JJ model_NN of_IN self-evaluation_NN motivations_NNS (_( Sedikides_NP &_CC Strube_NP ,_, 1997_CD )_) all_DT support_NN the_DT longitudinal_JJ reciprocal_JJ relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN and_CC its_PP$ growth_NN ._SENT The_DT two_CD corresponding_JJ hypotheses_NNS are_VBP 
Hypothesis_NN 5_CD :_: There_EX will_MD be_VB a_DT longitudinal_JJ reciprocal_JJ relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN ,_, such_PDT that_DT Hypothesis_NP 1_CD and_CC Hypothesis_NP 3_CD will_MD both_CC be_VB supported_VVN ._SENT 
Hypothesis_NN 6_CD :_: There_EX will_MD be_VB a_DT longitudinal_JJ reciprocal_JJ relationship_NN between_IN CSE_NN and_CC growth_NN of_IN job_NN satisfaction_NN ,_, such_PDT that_DT Hypothesis_NP 2_CD and_CC Hypothesis_NP 4_CD will_MD both_CC be_VB supported_VVN ._SENT 
Hypothesized_VVN Model_NP 
In_IN this_DT study_NN ,_, we_PP aimed_VVD to_TO examine_VV the_DT longitudinal_JJ relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN and_CC its_PP$ growth_NN in_IN order_NN to_TO evaluate_VV the_DT assumptions_NNS of_IN the_DT dispositional_JJ ,_, contextual_JJ ,_, and_CC corresponsive_JJ perspectives_NNS ._SENT Because_IN our_PP$ research_NN paradigm_NN is_VBZ complex_JJ in_IN terms_NNS of_IN analysis_NN ,_, we_PP briefly_RB introduce_VV the_DT conceptual_JJ research_NN model_NN in_IN relation_NN to_TO the_DT study_NN design_NN (_( see_VV Figure_NP 1_CD )_) to_TO illustrate_VV how_WRB we_PP tested_VVD the_DT hypotheses_NNS ._SENT 
We_PP used_VVD a_DT 10-year_JJ longitudinal_JJ design_NN in_IN which_WDT job_NN satisfaction_NN was_VBD assessed_VVN annually_RB for_IN 10_CD years_NNS and_CC CSE_NNS was_VBD assessed_VVN in_IN Years_NNS 5_CD and_CC 10._CD We_PP divided_VVD our_PP$ data_NNS into_IN four_CD stages�_NN (_( a_DT )_) Year_NN 1_CD to_TO Year_NP 4_CD ,_, (_( b_LS )_) Year_NP 5_CD ,_, (_( c_LS )_) Year_NN 6_CD to_TO Year_NP 9_CD ,_, and_CC (_( d_LS )_) Year_NP 10�so_NP that_IN/that we_PP could_MD estimate_VV links_NNS between_IN the_DT two_CD measures_NNS of_IN CSE_NN and_CC the_DT levels_NNS of_IN and_CC growth_NN in_IN job_NN satisfaction_NN across_IN the_DT 10-year_JJ period_NN ._SENT For_IN job_NN satisfaction_NN ,_, we_PP estimated_VVD an_DT intercept_NN factor_NN (_( i.e._FW ,_, job_NN satisfaction_NN at_IN Year_NP 1_CD )_) and_CC a_DT growth_NN slope_NN factor_NN (_( i.e._FW ,_, growth_NN rate_NN in_IN job_NN satisfaction_NN from_IN Year_NN 1_CD to_TO Year_NP 4_CD )_) at_IN the_DT first_JJ stage_NN and_CC another_DT intercept_NN factor_NN (_( i.e._FW ,_, job_NN satisfaction_NN at_IN Year_NP 6_CD )_) and_CC growth_NN slope_NN factor_NN (_( i.e._FW ,_, growth_NN rate_NN in_IN job_NN satisfaction_NN from_IN Year_NN 6_CD to_TO Year_NP 9_CD )_) at_IN the_DT third_JJ stage_NN ._SENT For_IN CSE_NN ,_, we_PP assessed_VVD CSE_JJ level_NN at_IN Years_NNS 5_CD and_CC 10._CD The_DT full_JJ design_NN shown_VVN in_IN Figure_NP 1_CD involves_VVZ eight_CD latent_JJ constructs_NNS that_WDT allowed_VVD us_PP to_TO test_VV longitudinal_JJ relationships_NNS ._SENT Paths_NNS involving_VVG different_JJ hypotheses_NNS are_VBP indicated_VVN in_IN the_DT figure_NN ._SENT 
Method_NN 
Participants_NNS and_CC Procedure_NN 
Data_NNS from_IN the_DT British_NP Household_NP Panel_NP Survey_NP (_( BHPS_NP ;_: Taylor_NP ,_, 2010_CD )_) were_VBD used_VVN in_IN the_DT current_JJ study_NN ._SENT The_DT BHPS_NP is_VBZ an_DT annual_JJ survey_NN with_IN a_DT nationally_RB representative_JJ sample_NN recruited_VVN in_IN 1991._CD To_TO date_VV ,_, it_PP is_VBZ composed_VVN of_IN 18_CD waves_NNS that_WDT have_VHP occurred_VVN between_IN 1991_CD and_CC 2008._CD Job_NP satisfaction_NN has_VHZ been_VBN assessed_VVN at_IN each_DT wave_NN ,_, and_CC CSE_NN measures_NNS were_VBD extracted_VVN from_IN data_NNS in_IN 2001_CD and_CC 2006._CD To_TO investigate_VV the_DT relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN ,_, we_PP used_VVD data_NNS from_IN 1997_CD to_TO 2006_CD (_( denoted_VVN as_IN Year_NN 1_CD to_TO Year_NP 10_CD in_IN the_DT following_VVG report_NN )_) ,_, which_WDT resulted_VVD in_IN a_DT four-stage_NN data_NN structure_NN :_: (_( a_DT )_) Year_NN 1_CD to_TO Year_NP 4_CD (_( 1997�2000_NP )_) ,_, (_( b_LS )_) Year_NN 5_CD (_( 2001_CD )_) ,_, (_( c_LS )_) Year_NN 6_CD to_TO Year_NP 9_CD (_( 2002�2005_JJ )_) ,_, and_CC (_( d_LS )_) Year_NN 10_CD (_( 2006_CD )_) ._SENT 
Participants_NNS were_VBD selected_VVN according_VVG to_TO the_DT following_VVG criteria_NNS :_: (_( a_DT )_) participants_NNS are_VBP all_DT employees_NNS (_( self-employed_JJ participants_NNS are_VBP not_RB included_VVN )_) ,_, (_( b_LS )_) participants_NNS had_VHD at_IN least_JJS two_CD data_NNS points_NNS in_IN each_DT 5-year_JJ period_NN in_IN order_NN for_IN us_PP to_TO construct_VV the_DT growth_NN effect_NN of_IN job_NN satisfaction_NN in_IN that_DT period_NN ,_, and_CC (_( c_LS )_) participants_NNS had_VHD completed_VVN demographic_JJ data_NNS on_IN sex_NN ,_, age_NN ,_, and_CC averaged_VVD annual_JJ income_NN in_IN order_NN for_IN us_PP to_TO control_VV the_DT demographic_JJ variables_NNS in_IN our_PP$ model_NN ._SENT On_IN the_DT basis_NN of_IN these_DT three_CD criteria_NNS ,_, 5_CD ,_, 827_CD participants_NNS were_VBD selected_VVN and_CC used_VVN in_IN the_DT following_VVG analysis_NN ._SENT In_IN this_DT sample_NN ,_, 2_CD ,_, 831_CD were_VBD male_JJ (_( 48.6_CD %_NN )_) and_CC 2_CD ,_, 996_CD were_VBD female_JJ (_( 51.4_CD %_NN )_) ._SENT Their_PP$ ages_NNS were_VBD from_IN 13_CD to_TO 73_CD years_NNS at_IN 1997_CD ,_, with_IN a_DT mean_NN of_IN 34.61_CD and_CC a_DT standard_JJ deviation_NN of_IN 11.70_CD ._SENT There_EX were_VBD 443_CD (_( 7.6_CD %_NN )_) participants_NNS under_IN the_DT age_NN of_IN 18_CD (_( 13�17_JJ )_) in_IN 1997._CD We_PP included_VVD these_DT participants_NNS because_IN they_PP began_VVD to_TO provide_VV job_NN satisfaction_NN data_NNS after_IN they_PP were_VBD 16-year-olds_JJ and_CC had_VHD a_DT paid_VVN job_NN experience_NN ._SENT Only_RB 18_CD participants_NNS in_IN the_DT sample_NN were_VBD older_JJR than_IN 65_CD in_IN 1997._CD 
Measures_NNS 
Job_NN satisfaction_NN 
Job_NN satisfaction_NN was_VBD measured_VVN by_IN five_CD available_JJ items_NNS across_IN 10_CD years_NNS in_IN the_DT BHPS_NP database_NN from_IN 1997_CD to_TO 2006_CD (_( Year_NN 1_CD to_TO Year_NP 10_CD )_) ._SENT The_DT five_CD items_NNS were_VBD overall_JJ job_NN satisfaction_NN ,_, satisfaction_NN with_IN total_JJ pay_NN ,_, satisfaction_NN with_IN security_NN ,_, satisfaction_NN with_IN work_NN itself_PP ,_, and_CC satisfaction_NN with_IN work_NN hours_NNS ._SENT The_DT mean_NN of_IN the_DT five_CD items_NNS was_VBD used_VVN to_TO indicate_VV the_DT level_NN of_IN job_NN satisfaction_NN in_IN each_DT year_NN ._SENT Participants_NNS used_VVD 7-point_JJ scales_NNS from_IN 1_CD (_( not_RB satisfied_VVN at_IN all_DT )_) to_TO 7_CD (_( completely_RB satisfied_VVN )_) to_TO rate_VV these_DT items_NNS ._SENT Cronbach_NP 's_POS alpha_NN coefficients_NNS for_IN these_DT five_CD items_NNS were_VBD higher_JJR than_IN .75_CD for_IN each_DT year_NN (_( see_VV Table_NP 1_CD )_) ._SENT 
Core_NN self-evaluations_NNS 
Because_IN there_EX is_VBZ no_DT standard_JJ measure_NN of_IN CSE_NN in_IN the_DT BHPS_NP database_NN ,_, following_VVG the_DT work_NN of_IN Judge_NP and_CC Hurst_NP (_( 2008_CD )_) ,_, we_PP selected_VVD items_NNS that_IN/that best_JJS represent_VVP the_DT construct_NN of_IN CSE_NP ._SENT The_DT same_JJ six_CD items_NNS were_VBD selected_VVN to_TO measure_VV CSE_NN in_IN Year_NP 5_CD (_( 2001_CD )_) and_CC Year_NP 10_CD (_( 2006_CD )_) ._SENT Four_CD items_NNS were_VBD selected_VVN from_IN the_DT General_NP Health_NP Questionnaire_NP (_( GHQ-12_NP ;_: Goldberg_NP &_CC Williams_NP ,_, 1988_CD )_) ,_, and_CC two_CD items_NNS were_VBD selected_VVN from_IN the_DT CASP_NP quality-of-life_NN measure_NN (_( CASP-19_NP ;_: Wiggins_NP ,_, Netuveli_NP ,_, Hyde_NP ,_, Higgs_NP ,_, &_CC Blane_NP ,_, 2008_CD )_) ._SENT The_DT four_CD items_NNS from_IN the_DT GHQ-12_NP were_VBD �Have_JJ you_PP recently_RB been_VBN able_JJ to_TO face_VV up_RP to_TO problems_NNS ?_SENT �_NN (_( for_IN self-efficacy_NN )_) ,_, �Have_VVD you_PP recently_RB been_VBN feeling_VVG unhappy_JJ or_CC depressed_JJ ?_SENT �_NN (_( for_IN neuroticism_NN )_) ,_, �Have_VVD you_PP recently_RB been_VBN losing_VVG confidence_NN in_IN yourself_PP ?_SENT �_NN (_( for_IN self-efficacy_NN )_) ,_, and_CC �Have_VV you_PP recently_RB been_VBN thinking_VVG of_IN yourself_PP as_IN a_DT worthless_JJ person_NN ?_SENT �_NN (_( for_IN self-esteem_NN )_) ._SENT Participants_NNS used_VVD 4-point_JJ scales_NNS with_IN different_JJ descriptors_NNS to_TO indicate_VV their_PP$ answers_NNS to_TO these_DT questions_NNS ._SENT Because_IN these_DT four_CD items_NNS have_VHP momentary_JJ wording_NN (_( i.e._FW ,_, �Have_VV you_PP recently��_JJ )_) ,_, the_DT momentary_JJ effect_NN will_MD be_VB controlled_VVN in_IN analysis_NN ._SENT The_DT two_CD items_NNS from_IN the_DT CASP-19_NP were_VBD �I_NNS feel_VVP that_IN/that what_WP happens_VVZ to_TO me_PP is_VBZ out_RB of_IN my_PP$ control�_NN (_( for_IN sense_NN of_IN control_NN )_) and_CC �I_NNS feel_VVP left_VVN out_RP of_IN things�_NN (_( also_RB for_IN sense_NN of_IN control_NN )_) ._SENT Participants_NNS used_VVD 4-point_JJ scales_NNS from_IN 1_CD (_( often_RB )_) to_TO 4_CD (_( never_RB )_) to_TO rate_VV themselves_PP on_IN these_DT two_CD items_NNS ._SENT All_PDT the_DT used_VVN items_NNS were_VBD similar_JJ to_TO the_DT existing_VVG instruments_NNS for_IN assessing_VVG CSE_NP (_( Judge_NP et_NP al_NP ._SENT ,_, 2003_CD ;_: Judge_NP &_CC Hurst_NP ,_, 2008_CD )_) ._SENT Cronbach_NP 's_POS alpha_NN coefficients_NNS for_IN these_DT six_CD items_NNS were_VBD .78_CD and_CC .80_CD for_IN each_DT year_NN (_( see_VV Table_NP 1_CD )_) ._SENT To_TO ensure_VV the_DT validity_NN of_IN our_PP$ measure_NN ,_, we_PP examined_VVD the_DT correlation_NN between_IN the_DT six-item_NN measure_NN used_VVN in_IN our_PP$ study_NN and_CC the_DT CSE_JJ measure_NN developed_VVN by_IN Judge_NP et_CC al_NP ._SENT (_( 2003_CD )_) in_IN a_DT sample_NN of_IN 310_CD undergraduates_NNS in_IN Taiwan_NP ._SENT The_DT correlation_NN between_IN the_DT two_CD CSE_JJ measures_NNS was_VBD .70_CD ,_, supporting_VVG the_DT criterion_NN validity_NN of_IN our_PP$ used_VVN CSE_NN items_NNS ._SENT 
Data_NP Analysis_NP 
We_PP tested_VVD our_PP$ hypotheses_NNS with_IN a_DT series_NN of_IN structural_JJ equation_NN models_NNS ._SENT Specification_NN for_IN each_DT model_NN is_VBZ provided_VVN in_IN the_DT Results_NNS section_NN ._SENT All_DT models_NNS were_VBD estimated_VVN with_IN Mplus_NP (_( Muth�n_NP &_CC Muth�n_NP ,_, 2007_CD )_) ._SENT To_TO take_VV into_IN account_NN nonnormality_JJ of_IN data_NNS and_CC missing_VVG data_NNS ,_, we_PP used_VVD a_DT maximum_JJ likelihood_NN estimator_NN with_IN robust_JJ standard_JJ errors_NNS using_VVG a_DT numerical_JJ integration_NN algorithm_NN (_( i.e._FW ,_, the_DT MLR_NP estimator_NN in_IN Mplus_NP )_) ._SENT This_DT estimator_NN generates_VVZ robust_JJ estimation_NN to_TO nonnormality_JJ and_CC nonindependence_NN of_IN data_NNS and_CC can_MD also_RB deal_VV with_IN missing_VVG data_NNS in_IN estimation_NN that_WDT is_VBZ based_VVN on_IN the_DT missing-at-random_NN assumption_NN (_( Muth�n_NP &_CC Muth�n_NP ,_, 2007_CD )_) ._SENT In_IN all_DT models_NNS ,_, the_DT first_JJ loading_NN of_IN each_DT latent_JJ factor_NN was_VBD set_VVN as_IN 1_CD to_TO fix_VV the_DT latent_JJ factor_NN scale_NN ._SENT We_PP relied_VVD on_IN four_CD fit_NN indices�comparative_JJ fit_NN index_NN (_( CFI_NP )_) ,_, Tucker�Lewis_NP index_NN (_( TLI_NP )_) ,_, root-mean-square_NN error_NN of_IN approximation_NN (_( RMSEA_NP )_) ,_, and_CC standardized_JJ root-mean-square_NN residual_NN (_( SRMR_NP )_) �as_NNS suggested_VVN by_IN Hu_NP and_CC Bentler_NP (_( 1999_CD )_) ,_, to_TO evaluate_VV our_PP$ models_NNS ._SENT Finally_RB ,_, in_IN all_DT models_NNS ,_, sex_NN ,_, age_NN ,_, and_CC average_JJ annual_JJ income_NN were_VBD included_VVN in_IN the_DT model_NN as_IN time-invariant_JJ control_NN variables_NNS ._SENT We_PP did_VVD not_RB treat_VV income_NN as_IN a_DT time-varied_JJ variable_NN because_IN annual_JJ income_NN had_VHD nonsignificant_JJ or_CC small_JJ effects_NNS on_IN job_NN satisfaction_NN in_IN a_DT preliminary_JJ analysis_NN when_WRB it_PP was_VBD treated_VVN as_IN a_DT time-varied_JJ variable_NN ,_, and_CC treating_VVG income_NN as_IN a_DT time-invariant_JJ variable_NN also_RB reduced_VVD the_DT number_NN of_IN parameters_NNS in_IN the_DT models_NNS ._SENT Given_VVN that_DT effects_NNS of_IN sex_NN ,_, age_NN ,_, and_CC average_JJ annual_JJ income_NN were_VBD small_JJ and_CC not_RB main_JJ concerns_NNS in_IN our_PP$ study_NN ,_, we_PP have_VHP not_RB reported_VVN their_PP$ specific_JJ effects_NNS in_IN the_DT Results_NNS section_NN ._SENT Results_NNS of_IN these_DT effects_NNS are_VBP available_JJ upon_IN request_NN ._SENT 
Results_NNS 
Descriptive_JJ Analysis_NN 
Table_NN 1_CD presents_VVZ means_NNS ,_, standard_JJ deviations_NNS ,_, and_CC correlations_NNS among_IN the_DT variables_NNS ,_, including_VVG job_NN satisfaction_NN for_IN each_DT of_IN the_DT 10_CD years_NNS and_CC CSE_NNS assessed_VVN at_IN Year_NP 5_CD and_CC Year_NP 10._CD 
Preliminary_JJ Analysis_NN 
Before_IN examining_VVG our_PP$ hypotheses_NNS ,_, we_PP conducted_VVD several_JJ preliminary_JJ tests_NNS ._SENT First_RB ,_, factor_NN invariance_NN of_IN the_DT measures_NNS is_VBZ essential_JJ to_TO ensure_VV that_IN/that the_DT same_JJ construct_NN over_IN time_NN is_VBZ assessed_VVN (_( Golembiewski_NP ,_, Billingsley_NP ,_, &_CC Yeager_NP ,_, 1976_CD )_) ._SENT To_TO test_VV invariance_NN ,_, we_PP examined_VVD the_DT factor_NN structure_NN of_IN CSE_NN and_CC job_NN satisfaction_NN measurements_NNS and_CC tested_VVD the_DT invariance_NN of_IN factor_NN loadings_NNS and_CC item_NN intercepts_NNS within_IN the_DT same_JJ constructs_NNS over_IN time_NN ._SENT These_DT tests_NNS are_VBP essential_JJ to_TO ensure_VV that_IN/that the_DT change_NN phenomena_NNS upon_IN which_WDT we_PP rely_VVP in_IN the_DT following_VVG longitudinal_JJ analysis_NN are_VBP about_RB the_DT changes_NNS in_IN target_NN construct_NN (_( true_JJ change_NN )_) ,_, rather_RB than_IN the_DT changes_NNS resulting_VVG from_IN scale_NN recalibration_NN (_( i.e._FW ,_, beta_JJ change_NN )_) and_CC construct_NN reconceptualization_NN (_( i.e._FW ,_, gamma_NN change_NN ;_: Golembiewski_NP et_NP al_NP ._SENT ,_, 1976_CD ;_: Sprangers_NP &_CC Schwartz_NP ,_, 1999_CD )_) ._SENT Second_RB ,_, because_IN we_PP included_VVD the_DT rate_NN of_IN change_NN in_IN job_NN satisfaction_NN as_IN a_DT construct_NN in_IN our_PP$ main_JJ analysis_NN ,_, we_PP also_RB tested_VVD the_DT size_NN and_CC shape_NN of_IN the_DT growth_NN effect_NN of_IN job_NN satisfaction_NN in_IN our_PP$ sample_NN ._SENT We_PP first_RB report_VVP factor_NN analysis_NN results_NNS for_IN CSE_NN and_CC job_NN satisfaction_NN and_CC then_RB report_VV the_DT results_NNS of_IN the_DT job_NN satisfaction_NN growth_NN structure_NN ._SENT Model_NN fit_NN of_IN all_DT models_NNS is_VBZ summarized_VVN in_IN Table_NP 2._CD 
Measurement_NN model_NN of_IN CSE_NP 
In_IN the_DT measurement_NN model_NN of_IN CSE_NN ,_, we_PP aimed_VVD to_TO extract_VV a_DT latent_JJ CSE_NN factor_NN to_TO represent_VV a_DT general_JJ ,_, trait-like_JJ construct_NN ._SENT Because_IN four_CD items_NNS for_IN CSE_NP were_VBD assessed_VVN with_IN a_DT momentary_JJ wording_NN (_( i.e._FW ,_, �Have_VV you_PP recently��_JJ )_) ,_, we_PP used_VVD the_DT correlated_VVN trait�correlated_JJ method_NN minus_CC one_CD (_( CT-C_NP (_( M_NP -_: 1_LS )_) )_) model_NN (_( Eid_NP ,_, 2000_CD ;_: Eid_NP ,_, Lischetzke_NP ,_, Nussbeck_NP ,_, &_CC Trierweiler_NP ,_, 2003_CD )_) to_TO control_VV the_DT momentary_JJ effect_NN when_WRB extracting_VVG the_DT CSE_JJ factor_NN ._SENT 
The_DT CT-C_NP (_( M_NP -_: 1_LS )_) model_VV (_( Eid_NP ,_, 2000_CD )_) was_VBD developed_VVN to_TO analyze_VV the_DT multitrait�multimethod_JJ data_NNS (_( Campbell_NP &_CC Fiske_NP ,_, 1959_CD )_) ._SENT Instead_RB of_IN estimating_VVG method_NN factors_NNS for_IN all_DT methods_NNS as_IN in_IN the_DT correlated_VVN trait�correlated_JJ method_NN model_NN (_( Marsh_NP ,_, 1989_CD ;_: Widaman_NP ,_, 1985_CD )_) ,_, the_DT CT-C_NP (_( M_NP -_: 1_LS )_) model_NN uses_VVZ one_CD method_NN as_IN the_DT reference_NN method_NN and_CC thus_RB has_VHZ one_CD method_NN factor_NN fewer_JJR than_IN the_DT number_NN of_IN methods_NNS ._SENT For_IN example_NN ,_, if_IN employees_NNS '_POS job_NN performance_NN was_VBD assessed_VVN by_IN self-report_NN ,_, peer_NN report_NN ,_, and_CC supervisor_NN 's_POS report_NN (_( i.e._FW ,_, three_CD methods_NNS )_) ,_, one_CD method_NN is_VBZ chosen_VVN as_IN the_DT reference_NN method_NN ,_, and_CC method_NN factors_NNS for_IN the_DT other_JJ two_CD methods_NNS will_MD be_VB extracted_VVN ._SENT In_IN this_DT specification_NN ,_, a_DT trait_NN factor_NN is_VBZ thus_RB defined_VVN as_IN the_DT true_JJ score_NN of_IN a_DT construct_NN measured_VVN by_IN the_DT reference_NN method_NN (_( for_IN more_JJR elaboration_NN and_CC applications_NNS ,_, see_VVP Eid_NP et_NP al_NP ._SENT ,_, 2003_CD ,_, Eid_NP et_CC al_NP ._SENT ,_, 2008_CD ;_: Nussbeck_NP ,_, Eid_NP ,_, Geiser_NP ,_, Courvoisier_JJ ,_, &_CC Lischetzke_NP ,_, 2009_CD )_) ._SENT 
In_IN our_PP$ study_NN ,_, the_DT six_CD CSE_JJ items_NNS can_MD be_VB viewed_VVN as_IN assessing_VVG one_CD trait_NN (_( i.e._FW ,_, CSE_NP )_) with_IN two_CD methods_NNS (_( i.e._FW ,_, momentary_JJ and_CC general_JJ )_) ._SENT Because_IN we_PP aimed_VVD to_TO focus_VV CSE_NN on_IN the_DT trait_NN level_NN ,_, using_VVG the_DT CT-C_NP (_( M_NP -_: 1_LS )_) model_NN ,_, we_PP treated_VVD the_DT general_JJ statements_NNS as_IN the_DT reference_NN method_NN and_CC then_RB constructed_VVN a_DT method_NN factor_NN for_IN the_DT momentary_JJ meaning_NN of_IN CSE_NP ._SENT Therefore_RB ,_, at_IN each_DT time_NN point_NN ,_, we_PP estimated_VVD a_DT latent_JJ CSE_NN factor_NN influencing_VVG all_PDT the_DT six_CD items_NNS to_TO represent_VV the_DT trait-like_JJ CSE_NN (_( because_IN it_PP was_VBD measured_VVN by_IN the_DT general_JJ statement_NN )_) and_CC a_DT method_NN factor_NN influencing_VVG only_RB the_DT four_CD items_NNS with_IN momentary_JJ wordings_NNS to_TO represent_VV the_DT momentary_JJ method_NN in_IN assessing_VVG CSE_NNS ._SENT In_IN this_DT way_NN ,_, the_DT CT-C_NP (_( M_NP -_: 1_LS )_) model_NN uses_VVZ information_NN from_IN both_CC the_DT general_JJ and_CC momentary_JJ statements_NNS to_TO better_RBR identify_VV the_DT trait-like_JJ properties_NNS of_IN the_DT CSE_JJ items_NNS ._SENT Across_IN the_DT two_CD time_NN points_NNS ,_, the_DT two_CD latent_JJ CSE_NN factors_NNS were_VBD allowed_VVN to_TO correlate_VV ,_, and_CC the_DT two_CD method_NN factors_NNS were_VBD allowed_VVN to_TO correlate_VV ._SENT However_RB ,_, CSE_JJ factors_NNS were_VBD not_RB correlated_VVN with_IN the_DT two_CD method_NN factors_NNS ._SENT Finally_RB ,_, errors_NNS of_IN the_DT same_JJ items_NNS were_VBD correlated_VVN across_IN time_NN to_TO capture_VV reliable_JJ item-specific_NN variance_NN across_IN time_NN ._SENT 
The_DT measurement_NN model_NN of_IN CSE_NP (_( Model_NP 1A_NP )_) was_VBD acceptable_JJ ,_, ?_SENT 2_LS (_( 38_LS )_) =_SYM 303.47_CD ,_, CFI_NP =_SYM .99_CD ,_, TLI_NP =_SYM .98_CD ,_, RMSEA_NP =_SYM .035_CD ,_, SRMR_NP =_SYM .022_CD ._SENT All_DT factor_NN loadings_NNS were_VBD significant_JJ at_IN p_NN <_SYM .01_CD ._SENT On_IN the_DT basis_NN of_IN this_DT model_NN ,_, we_PP further_RBR tested_VVD the_DT longitudinal_JJ invariance_NN of_IN the_DT items_NNS ._SENT We_PP first_RB constrained_VVD the_DT equality_NN of_IN factor_NN loadings_NNS for_IN the_DT same_JJ items_NNS across_IN time_NN ._SENT The_DT model_NN with_IN equality_NN of_IN factor_NN loadings_NNS (_( Model_NP 1B_NP )_) had_VHD a_DT similar_JJ model_NN fit_NN ,_, ?_SENT 2_LS (_( 43_LS )_) =_SYM 311.56_CD ,_, CFI_NP =_SYM .99_CD ,_, TLI_NP =_SYM .98_CD ,_, RMSEA_NP =_SYM .033_CD ,_, SRMR_NP =_SYM .023_CD ,_, denoting_VVG a_DT weak_JJ invariance_NN property_NN (_( i.e._FW ,_, equality_NN of_IN factor_NN loadings_NNS )_) ._SENT Next_RB ,_, we_PP imposed_VVD equality_NN constraints_NNS for_IN item_NN intercepts_NNS of_IN the_DT same_JJ items_NNS over_IN time_NN ._SENT The_DT model_NN with_IN equality_NN of_IN item_NN intercepts_NNS (_( Model_NP 1C_NP )_) also_RB had_VHD a_DT good_JJ model_NN fit_NN ,_, ?_SENT 2_LS (_( 49_LS )_) =_SYM 347.98_CD ,_, CFI_NP =_SYM .98_CD ,_, TLI_NP =_SYM .98_CD ,_, RMSEA_NP =_SYM .033_CD ,_, SRMR_NP =_SYM .026_CD ,_, revealing_VVG that_IN/that these_DT items_NNS capture_VV the_DT same_JJ meaning_NN of_IN the_DT latent_JJ CSE_NN construct_NN and_CC also_RB have_VH the_DT same_JJ intercepts_NNS over_IN time_NN ._SENT In_IN this_DT strong_JJ invariance_NN model_NN (_( i.e._FW ,_, equality_NN of_IN factor_NN loadings_NNS and_CC item_NN intercepts_NNS )_) ,_, the_DT correlation_NN between_IN the_DT two_CD latent_JJ CSE_NN factors_NNS was_VBD .63_JJ ._SENT 
Measurement_NN model_NN of_IN job_NN satisfaction_NN 
Five_CD items_NNS for_IN job_NN satisfaction_NN assessed_VVN in_IN the_DT same_JJ year_NN were_VBD influenced_VVN by_IN a_DT latent_JJ factor_NN for_IN that_DT year_NN ,_, which_WDT resulted_VVD in_IN 10_CD factors_NNS for_IN the_DT 10-year_JJ data_NNS ._SENT The_DT 10_CD factors_NNS were_VBD allowed_VVN to_TO correlate_VV ._SENT Except_IN for_IN the_DT global_JJ job_NN satisfaction_NN item_NN ,_, 1_CD errors_NNS in_IN items_NNS were_VBD allowed_VVN to_TO be_VB correlated_VVN between_IN the_DT same_JJ items_NNS to_TO capture_VV the_DT reliable_JJ item-specific_NN variance_NN across_IN time_NN ._SENT 
Overall_JJ fit_NN indices_NN showed_VVD that_IN/that this_DT measurement_NN model_NN (_( Model_NP 2A_NP )_) was_VBD acceptable_JJ ,_, ?_SENT 2_LS (_( 950_LS )_) =_SYM 2469.10_CD ,_, CFI_NP =_SYM .98_CD ,_, TLI_NP =_SYM .98_CD ,_, RMSEA_NP =_SYM .017_CD ,_, SRMR_NP =_SYM .037_CD ._SENT All_DT factor_NN loadings_NNS were_VBD significant_JJ at_IN p_NN <_SYM .01_CD ._SENT We_PP further_RBR constrained_VVD the_DT equality_NN of_IN factor_NN loadings_NNS for_IN the_DT same_JJ items_NNS over_IN time_NN ._SENT The_DT model_NN with_IN equality_NN of_IN factor_NN loadings_NNS (_( Model_NP 2B_NP )_) had_VHD a_DT similar_JJ model_NN fit_NN ,_, ?_SENT 2_LS (_( 986_LS )_) =_SYM 2513.58_CD ,_, CFI_NP =_SYM .98_CD ,_, TLI_NP =_SYM .98_CD ,_, RMSEA_NP =_SYM .016_CD ,_, SRMR_NP =_SYM .037_CD ,_, denoting_VVG a_DT weak_JJ invariance_NN property_NN ._SENT Next_RB ,_, we_PP additionally_RB imposed_VVD the_DT equality_NN of_IN item_NN intercepts_NNS for_IN the_DT same_JJ items_NNS over_IN time_NN ._SENT The_DT model_NN with_IN equality_NN of_IN item_NN intercepts_NNS (_( Model_NP 2C_NP )_) had_VHD a_DT similar_JJ model_NN fit_NN ,_, ?_SENT 2_CD (_( 1_CD ,_, 031_CD )_) =_SYM 2960.42_CD ,_, CFI_NP =_SYM .98_CD ,_, TLI_NP =_SYM .97_CD ,_, RMSEA_NP =_SYM .018_CD ,_, SRMR_NP =_SYM .038_CD ,_, denoting_VVG a_DT strong_JJ invariance_NN property_NN ._SENT In_IN this_DT strong_JJ invariance_NN model_NN ,_, the_DT correlations_NNS between_IN job_NN satisfaction_NN factors_NNS ranged_VVD from_IN .25_CD to_TO .57_CD with_IN a_DT mean_NN of_IN .39_CD ._SENT 
Piecewise_JJ latent_JJ growth_NN curve_NN model_NN of_IN job_NN satisfaction_NN 
We_PP examined_VVD the_DT growth_NN of_IN job_NN satisfaction_NN using_VVG a_DT piecewise_JJ latent_JJ growth_NN curve_NN model_NN (_( e.g._FW ,_, Chou_NP ,_, Yang_NP ,_, Pentz_NP ,_, &_CC Hser_NP ,_, 2004_CD ;_: Duncan_NP &_CC Duncan_NP ,_, 2004_CD )_) that_WDT enabled_VVD us_PP to_TO examine_VV growth_NN effects_NNS in_IN multiple_JJ stages_NNS ._SENT As_RB noted_VVD above_IN ,_, we_PP divided_VVD our_PP$ 10-year_JJ data_NNS into_IN four_CD stages_NNS :_: (_( a_DT )_) Year_NN 1_CD to_TO Year_NP 4_CD ,_, (_( b_LS )_) Year_NP 5_CD ,_, (_( c_LS )_) Year_NN 6_CD to_TO Year_NP 9_CD ,_, and_CC (_( d_LS )_) Year_NP 10._CD With_IN this_DT framework_NN ,_, we_PP examined_VVD how_WRB the_DT growth_NN process_NN of_IN job_NN satisfaction_NN in_IN the_DT first_JJ 4_CD years_NNS influenced_VVD CSE_NN and_CC job_NN satisfaction_NN at_IN Year_NP 5_CD ;_: and_CC how_WRB CSE_NN and_CC job_NN satisfaction_NN at_IN Year_NP 5_CD ,_, in_IN turn_NN ,_, influenced_VVD the_DT growth_NN process_NN of_IN job_NN satisfaction_NN in_IN the_DT next_JJ 4_CD years_NNS (_( Year_NN 6_CD to_TO Year_NP 9_CD )_) ,_, which_WDT subsequently_RB influences_VVZ CSE_NN and_CC job_NN satisfaction_NN at_IN Year_NP 10._CD 
We_PP used_VVD a_DT composite_JJ score_NN for_IN each_DT year_NN to_TO build_VV the_DT piecewise_JJ latent_JJ growth_NN curve_NN model_NN ,_, because_IN a_DT convergent_JJ solution_NN could_MD not_RB be_VB obtained_VVN when_WRB the_DT model_NN was_VBD built_VVN using_VVG item_NN scores_NNS ._SENT This_DT approach_NN seemed_VVD reasonable_JJ ,_, given_VVN that_IN/that the_DT measures_NNS of_IN job_NN satisfaction_NN were_VBD found_VVN to_TO be_VB invariant_JJ over_IN the_DT 10_CD years_NNS of_IN the_DT study_NN ._SENT 
We_PP constructed_VVD two_CD intercept_NN factors_NNS and_CC two_CD slope_NN factors_NNS in_IN the_DT model_NN so_IN that_DT job_NN satisfaction_NN scores_NNS from_IN Year_NN 1_CD to_TO Year_NP 4_CD were_VBD influenced_VVN by_IN an_DT intercept_NN factor_NN and_CC a_DT slope_NN factor_NN ,_, and_CC job_NN satisfaction_NN scores_NNS from_IN Year_NN 6_CD to_TO Year_NP 9_CD were_VBD influenced_VVN by_IN another_DT intercept_NN factor_NN and_CC slope_NN factor_NN ._SENT Loadings_NNS on_IN intercept_NN factors_NNS were_VBD all_RB set_VVN at_IN 1_CD ,_, and_CC the_DT first_JJ and_CC last_JJ loadings_NNS on_IN each_DT slope_NN factor_NN were_VBD set_VVN as_IN 0_CD and_CC 1_CD ,_, respectively_RB ._SENT Other_JJ loadings_NNS on_IN slope_NN factors_NNS were_VBD freely_RB estimated_VVN ._SENT These_DT specifications_NNS define_VVP job_NN satisfaction_NN scores_NNS at_IN Year_NP 1_CD and_CC Year_NP 6_CD as_IN the_DT intercepts_NNS ,_, and_CC the_DT two_CD slope_NN factors_NNS represent_VVP the_DT growth_NN rate_NN of_IN job_NN satisfaction_NN in_IN each_DT 4-year_JJ period_NN ._SENT Job_NN satisfaction_NN scores_NNS at_IN Year_NP 5_CD and_CC Year_NP 10_CD were_VBD each_DT influenced_VVN by_IN a_DT single-indicator_NN factor_NN ._SENT All_DT factors_NNS were_VBD free_JJ to_TO correlate_VV ._SENT This_DT piecewise_JJ latent_JJ growth_NN curve_NN model_NN (_( Model_NP 3_CD )_) had_VHD a_DT better_RBR fit_JJ ,_, ?_SENT 2_LS (_( 26_LS )_) =_SYM 57.30_CD ,_, CFI_NP =_SYM 1.00_CD ,_, TLI_NP =_SYM .99_CD ,_, RMSEA_NP =_SYM .014_CD ,_, SRMR_NP =_SYM .010_CD ,_, than_IN other_JJ alternative_JJ models_NNS ._SENT 2_LS 
All_DT loadings_NNS in_IN the_DT piecewise_JJ growth_NN model_NN were_VBD significant_JJ (_( ps_NNS <_SYM .01_CD )_) ,_, except_IN for_IN the_DT loading_NN from_IN the_DT first_JJ slope_NN factor_NN to_TO the_DT job_NN satisfaction_NN score_NN at_IN Year_NP 2._CD The_DT means_NN of_IN the_DT two_CD intercepts_NNS were_VBD 5.25_CD and_CC 5.27_CD ,_, which_WDT respectively_RB represent_VVP the_DT means_NN of_IN job_NN satisfaction_NN at_IN Year_NP 1_CD and_CC Year_NP 6._CD The_DT means_NN of_IN the_DT two_CD factors_NNS for_IN job_NN satisfaction_NN scores_NNS at_IN Year_NP 5_CD and_CC Year_NP 10_CD were_VBD 5.30_CD and_CC 5.33_CD ,_, respectively_RB ._SENT Factor_NN loadings_NNS of_IN both_DT slope_NN factors_NNS revealed_VVD a_DT pattern_NN of_IN increasing_JJ values_NNS (_( 0.00_CD ,_, 0.00_CD ,_, 0.38_CD ,_, and_CC 1.00_CD for_IN the_DT first_JJ slope_NN factor_NN ;_: 0.00_CD ,_, 0.36_CD ,_, 0.66_CD ,_, and_CC 1.00_CD for_IN the_DT second_JJ slope_NN factor_NN )_) ._SENT The_DT mean_NN of_IN the_DT first_JJ slope_NN factor_NN was_VBD not_RB different_JJ from_IN 0_CD ,_, suggesting_VVG that_IN/that the_DT average_JJ growth_NN rate_NN of_IN the_DT whole_JJ sample_NN was_VBD zero_CD for_IN this_DT period_NN ._SENT The_DT mean_NN of_IN the_DT second_JJ slope_NN factor_NN was_VBD significant_JJ (_( unstandardized_JJ estimate_NN =_SYM .05_CD ,_, p_NN <_SYM .01_CD )_) ,_, suggesting_VVG that_IN/that the_DT average_JJ growth_NN rate_NN of_IN job_NN satisfaction_NN was_VBD slightly_RB positive_JJ for_IN this_DT period_NN ._SENT Variances_NNS of_IN factors_NNS were_VBD all_RB significant_JJ (_( ps_NNS <_SYM .01_CD )_) ,_, indicating_VVG that_IN/that participants_NNS had_VHD different_JJ scores_NNS in_IN job_NN satisfaction_NN at_IN Year_NP 1_CD ,_, Year_NP 5_CD ,_, Year_NP 6_CD ,_, and_CC Year_NP 10._CD They_PP also_RB had_VHD significantly_RB different_JJ rates_NNS in_IN growth_NN of_IN job_NN satisfaction_NN within_IN the_DT two_CD periods_NNS ._SENT These_DT variances_NNS permitted_VVD us_PP to_TO conduct_VV analyses_NNS to_TO examine_VV the_DT hypothesized_VVN relationships_NNS between_IN CSE_NN and_CC job_NN satisfaction_NN and_CC its_PP$ growth_NN ._SENT 
Hypothesis_NN Testing_NP 
We_PP now_RB turn_VVP to_TO the_DT examinations_NNS of_IN our_PP$ hypotheses_NNS using_VVG the_DT strong_JJ invariance_NN model_NN of_IN CSE_NP and_CC the_DT piecewise_JJ latent_JJ growth_NN curve_NN model_NN of_IN job_NN satisfaction_NN ._SENT There_EX were_VBD 10_CD factors_NNS in_IN this_DT model_NN :_: two_CD method_NN factors_NNS for_IN CSE_NP ,_, two_CD CSE_JJ factors_NNS (_( CSE_NP at_IN Year_NP 5_CD and_CC Year_NP 10_CD )_) ,_, two_CD intercept_NN factors_NNS (_( job_NN satisfaction_NN at_IN Year_NP 1_CD and_CC Year_NP 6_CD )_) ,_, two_CD slope_NN factors_NNS (_( the_DT first_JJ growth_NN rate_NN and_CC the_DT second_JJ growth_NN rate_NN )_) ,_, and_CC two_CD single-indicator_NN factors_NNS (_( job_NN satisfaction_NN at_IN Year_NP 5_CD and_CC Year_NP 10_CD ;_: the_DT simple_JJ terms_NNS were_VBD used_VVN to_TO simplify_VV our_PP$ following_VVG reports_NNS )_) ._SENT We_PP first_RB allowed_VVD all_DT factors_NNS to_TO correlate_VV except_IN for_IN the_DT two_CD method_NN factors_NNS for_IN CSE_NP ,_, which_WDT were_VBD correlated_VVN only_RB with_IN each_DT other_JJ ._SENT This_DT correlated-factor_NN model_NN (_( Model_NP 4A_NP )_) was_VBD acceptable_JJ ,_, ?_SENT 2_LS (_( 223_LS )_) =_SYM 866.26_CD ,_, CFI_NP =_SYM .98_CD ,_, TLI_NP =_SYM .98_CD ,_, RMSEA_NP =_SYM .022_CD ,_, SRMR_NP =_SYM .020_CD ,_, suggesting_VVG that_IN/that both_CC the_DT strong_JJ invariance_NN model_NN of_IN CSE_NP and_CC the_DT piecewise_JJ latent_JJ growth_NN curve_NN model_NN of_IN job_NN satisfaction_NN together_RB were_VBD a_DT good_JJ fit_NN to_TO the_DT data_NNS ._SENT 
We_PP next_RB specified_VVD directional_JJ relationships_NNS among_IN factors_NNS according_VVG to_TO the_DT time_NN sequence_NN as_IN shown_VVN in_IN Figure_NP 2._CD To_TO test_VV Hypothesis_NP 1_CD ,_, CSE_NN at_IN Year_NP 5_CD predicted_VVD job_NN satisfaction_NN at_IN Year_NN 6._CD To_TO test_VV Hypothesis_NP 2_CD ,_, CSE_NN at_IN Year_NP 5_CD predicted_VVD the_DT second_JJ growth_NN rate_NN ._SENT To_TO test_VV Hypothesis_NP 3_CD ,_, job_NN satisfaction_NN at_IN Year_NP 1_CD predicted_VVD CSE_NN at_IN Year_NP 5_CD and_CC job_NN satisfaction_NN at_IN Year_NP 6_CD predicted_VVD CSE_NN at_IN Year_NN 10._CD To_TO test_VV Hypothesis_NP 4_CD ,_, the_DT first_JJ growth_NN rate_NN predicted_VVD CSE_NN at_IN Year_NP 5_CD and_CC the_DT second_JJ growth_NN rate_NN predicted_VVD CSE_NN at_IN Year_NP 10._CD 
We_PP also_RB included_VVD other_JJ directional_JJ effects_NNS in_IN the_DT model_NN to_TO control_VV for_IN prior_JJ levels_NNS of_IN the_DT main_JJ constructs_NNS ._SENT First_RB ,_, CSE_NN at_IN Year_NP 5_CD was_VBD used_VVN to_TO predict_VV CSE_NN at_IN Year_NP 10_CD ,_, to_TO account_VV for_IN its_PP$ autoregressive_NN effect_NN ._SENT Similarly_RB ,_, previous_JJ job_NN satisfaction_NN was_VBD used_VVN to_TO predict_VV the_DT next_JJ period_NN of_IN job_NN satisfaction_NN ._SENT Next_JJ ,_, directional_JJ paths_NNS associated_VVN with_IN intercept_NN and_CC slope_NN factors_NNS of_IN job_NN satisfaction_NN were_VBD specified_VVN ._SENT Specifically_RB ,_, job_NN satisfaction_NN at_IN Year_NP 1_CD predicted_VVD the_DT first_JJ growth_NN rate_NN ,_, and_CC job_NN satisfaction_NN at_IN Year_NP 6_CD predicted_VVD the_DT second_JJ growth_NN rate_NN ._SENT The_DT first_JJ growth_NN rate_NN predicted_VVD job_NN satisfaction_NN at_IN Year_NP 5_CD ,_, and_CC the_DT second_JJ growth_NN rate_NN predicted_VVD job_NN satisfaction_NN at_IN Year_NP 10._CD Finally_RB ,_, CSE_NN and_CC job_NN satisfaction_NN were_VBD correlated_VVN in_IN the_DT same_JJ year_NN to_TO account_VV for_IN shared_VVN variance_NN within_IN measurement_NN occasions_NNS ._SENT 
Overall_JJ fit_NN indices_NN showed_VVD that_IN/that this_DT directional_JJ path_NN model_NN (_( Model_NP 4B_NP )_) was_VBD acceptable_JJ ,_, ?_SENT 2_LS (_( 235_LS )_) =_SYM 1553.36_CD ,_, CFI_NP =_SYM .96_CD ,_, TLI_NP =_SYM .95_CD ,_, RMSEA_NP =_SYM .031_CD ,_, SRMR_NP =_SYM .041_CD ._SENT Figure_NN 2_CD presents_VVZ the_DT results_NNS of_IN this_DT model_NN ._SENT Regarding_VVG effects_NNS associated_VVN with_IN the_DT dispositional_JJ perspective_NN ,_, CSE_NN at_IN Year_NP 5_CD positively_RB predicted_VVN job_NN satisfaction_NN at_IN Year_NP 6_CD (_( unstandardized_VVN ?_SENT =_SYM .86_CD ,_, standardized_JJ ?_SENT =_SYM .14_CD ,_, p_NN <_SYM .01_CD )_) ,_, supporting_VVG Hypothesis_NP 1_CD and_CC revealing_VVG that_IN/that individuals_NNS with_IN higher_JJR CSE_NN at_IN Year_NP 5_CD tended_VVD to_TO report_VV a_DT higher_JJR level_NN of_IN job_NN satisfaction_NN at_IN Year_NP 6_CD compared_VVN with_IN their_PP$ counterparts_NNS ._SENT CSE_NNS at_IN Year_NP 5_CD also_RB positively_RB predicted_VVD the_DT second_JJ growth_NN rate_NN factor_NN (_( unstandardized_JJ ?_SENT =_SYM .52_CD ,_, standardized_JJ ?_SENT =_SYM .10_CD ,_, p_NN <_SYM .01_CD )_) ,_, supporting_VVG Hypothesis_NP 2_CD and_CC revealing_VVG that_IN/that individuals_NNS with_IN higher_JJR CSE_NN at_IN Year_NP 5_CD tended_VVD to_TO show_VV a_DT greater_JJR increase_NN in_IN their_PP$ job_NN satisfaction_NN from_IN Year_NN 6_CD to_TO Year_NP 9_CD compared_VVN with_IN their_PP$ counterparts_NNS ._SENT 
Regarding_VVG effects_NNS associated_VVN with_IN the_DT contextual_JJ perspective_NN ,_, job_NN satisfaction_NN at_IN Year_NP 1_CD positively_RB predicted_VVN CSE_NN at_IN Year_NP 5_CD (_( unstandardized_VVN ?_SENT =_SYM .06_CD ,_, standardized_JJ ?_SENT =_SYM .36_CD ,_, p_NN <_SYM .01_CD )_) ,_, and_CC job_NN satisfaction_NN at_IN Year_NP 6_CD positively_RB predicted_VVN CSE_NN at_IN Year_NP 10_CD (_( unstandardized_VVN ?_SENT =_SYM .02_CD ,_, standardized_JJ ?_SENT =_SYM .14_CD ,_, p_NN <_SYM .01_CD )_) ,_, supporting_VVG Hypothesis_NP 3_CD and_CC revealing_VVG that_IN/that individuals_NNS with_IN higher_JJR job_NN satisfaction_NN tended_VVN to_TO have_VH a_DT higher_JJR CSE_NNS later_RBR compared_VVN with_IN their_PP$ counterparts_NNS ._SENT The_DT first_JJ growth_NN rate_NN factor_NN positively_RB predicted_VVD CSE_NN at_IN Year_NP 5_CD (_( unstandardized_VVN ?_SENT =_SYM .03_CD ,_, standardized_JJ ?_SENT =_SYM .15_CD ,_, p_NN <_SYM .01_CD )_) ,_, and_CC the_DT second_JJ growth_NN rate_NN factor_NN positively_RB predicted_VVD CSE_NN at_IN Year_NP 10_CD (_( unstandardized_VVN ?_SENT =_SYM .03_CD ,_, standardized_JJ ?_SENT =_SYM .18_CD ,_, p_NN <_SYM .01_CD )_) ,_, supporting_VVG Hypothesis_NP 4_CD and_CC revealing_VVG that_IN/that individuals_NNS with_IN greater_JJR increase_NN in_IN their_PP$ job_NN satisfaction_NN tended_VVN to_TO have_VH a_DT higher_JJR CSE_NNS later_RBR compared_VVN with_IN their_PP$ counterparts_NNS ._SENT 
Because_IN all_DT proposed_VVN effects_NNS associated_VVN with_IN the_DT longitudinal_JJ reciprocal_JJ relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN were_VBD significant_JJ ,_, Hypothesis_NP 5_CD was_VBD supported_VVN ._SENT Similarly_RB ,_, all_DT effects_NNS associated_VVN with_IN the_DT longitudinal_JJ reciprocal_JJ relationship_NN between_IN CSE_NN and_CC growth_NN of_IN job_NN satisfaction_NN were_VBD significant_JJ ,_, and_CC thus_RB Hypothesis_NP 6_CD was_VBD supported_VVN ._SENT 
Discussion_NN 
Our_PP$ comparison_NN of_IN alternative_JJ causal_JJ relationships_NNS between_IN CSE_NN and_CC job_NN satisfaction_NN contributes_VVZ to_TO a_DT better_JJR understanding_NN of_IN the_DT malleability_NN of_IN CSE_NP ,_, the_DT link_NN between_IN CSE_NN and_CC work_NN experiences_NNS ,_, and_CC the_DT self-motives_NNS that_WDT shape_VVP CSE_NN ._SENT 
First_RB ,_, on_IN the_DT basis_NN of_IN recent_JJ research_NN on_IN personality_NN development_NN (_( e.g._FW ,_, Roberts_NP &_CC DelVecchio_NP ,_, 2000_CD ;_: Roberts_NP et_NP al_NP ._SENT ,_, 2006_CD )_) ,_, we_PP expected_VVD that_IN/that CSE_NP ,_, as_IN a_DT personality_NN trait_NN ,_, could_MD change_VV systematically_RB over_IN time_NN ._SENT Although_IN researchers_NNS have_VHP noted_VVN the_DT possibility_NN that_IN/that CSE_NP is_VBZ malleable_JJ ,_, long-term_JJ changes_NNS in_IN the_DT CSE_JJ trait_NN have_VHP not_RB been_VBN examined_VVN ._SENT Schinkel_NP et_FW al_NP ._SENT (_( 2004_CD )_) found_VVD that_IN/that CSE_NP was_VBD influenced_VVN by_IN experiences_NNS ,_, but_CC their_PP$ examination_NN focused_VVN on_IN a_DT state-like_JJ change_NN of_IN CSE_NN with_IN a_DT short-term_JJ experimental_JJ manipulation_NN ._SENT In_IN this_DT study_NN ,_, we_PP focused_VVD on_IN trait-like_JJ change_NN of_IN CSE_NN with_IN a_DT long-term_JJ longitudinal_JJ analysis_NN ._SENT We_PP found_VVD that_IN/that although_IN there_EX was_VBD a_DT high_JJ degree_NN of_IN stability_NN in_IN CSE_NP (_( test�retest_NN reliability_NN of_IN CSE_NP =_SYM .63_CD )_) ,_, there_EX was_VBD also_RB change_NN in_IN the_DT rank_JJ order_NN of_IN participants_NNS (_( Roberts_NP &_CC DelVecchio_NP ,_, 2000_CD )_) ,_, which_WDT implies_VVZ that_IN/that individuals_NNS can_MD change_VV their_PP$ CSE_JJ levels_NNS over_IN time_NN ._SENT The_DT malleability_NN of_IN CSE_NP could_MD play_VV a_DT functional_JJ role_NN in_IN goal_NN achievement_NN ,_, because_IN increased_VVN CSE_NN that_WDT arises_VVZ from_IN positive_JJ experiences_NNS gives_VVZ an_DT individual_JJ confidence_NN to_TO take_VV on_IN challenges_NNS and_CC more_JJR advanced_JJ goals_NNS ._SENT This_DT notion_NN is_VBZ indirectly_RB supported_VVN by_IN the_DT findings_NNS of_IN Tolli_NP and_CC Schmidt_NP (_( 2008_CD )_) and_CC Seo_NP and_CC Ilies_NNS (_( 2009_CD )_) regarding_VVG task-specific_JJ self-efficacy_NN ._SENT They_PP found_VVD that_IN/that feedback_NN or_CC past_JJ performance_NN determined_VVD the_DT level_NN of_IN task-specific_JJ self-efficacy_NN ,_, which_WDT ,_, in_IN turn_NN ,_, influenced_VVN goal_NN revision_NN ,_, suggesting_VVG that_IN/that self-evaluations_NNS can_MD mediate_VV the_DT relationship_NN between_IN previous_JJ experiences_NNS and_CC subsequent_JJ actions_NNS ._SENT 
Second_RB ,_, our_PP$ results_NNS showed_VVD that_IN/that systematic_JJ changes_NNS in_IN CSE_NP were_VBD related_VVN to_TO job_NN satisfaction_NN from_IN both_CC a_DT dispositional_JJ and_CC a_DT contextual_JJ perspective_NN ._SENT We_PP replicated_VVD previous_JJ findings_NNS that_IN/that dispositional_JJ CSE_NN influences_VVZ subsequent_JJ job_NN satisfaction_NN ._SENT Central_JJ to_TO our_PP$ study_NN 's_POS contribution_NN ,_, we_PP also_RB demonstrated_VVD a_DT contextual_JJ effect_NN of_IN job_NN satisfaction_NN on_IN subsequent_JJ CSE_NNS ._SENT Thus_RB ,_, overall_RB ,_, we_PP found_VVD a_DT corresponsive_JJ relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN over_IN time_NN that_WDT is_VBZ consistent_JJ with_IN a_DT dynamic_JJ process_NN through_IN which_WDT individuals_NNS shape_VVP and_CC are_VBP shaped_VVN by_IN experiences_NNS in_IN their_PP$ environment_NN (_( Caspi_NP et_NP al_NP ._SENT ,_, 2005_CD )_) ._SENT Our_PP$ findings_NNS suggest_VVP that_IN/that the_DT dispositional_JJ and_CC contextual_JJ effects_NNS are_VBP comparable_JJ in_IN size_NN ._SENT With_IN standardized_JJ path_NN coefficient_NN as_IN an_DT estimate_NN of_IN effect_NN size_NN and_CC Cohen_NP 's_POS (_( 1998_CD )_) effect_NN size_NN criteria_NNS for_IN correlation_NN coefficients_NNS (_( see_VV Kline_NP ,_, 2005_CD )_) ,_, effects_NNS of_IN prior_JJ CSE_NN on_IN later_JJR job_NN satisfaction_NN (_( .14_CD )_) and_CC growth_NN of_IN job_NN satisfaction_NN (_( .10_CD )_) and_CC effects_NNS of_IN prior_JJ job_NN satisfaction_NN (_( .36_CD and_CC .15_CD )_) and_CC growth_NN of_IN job_NN satisfaction_NN (_( .14_CD and_CC .18_CD )_) on_IN later_JJR CSE_NNS are_VBP small_JJ to_TO medium_NN ._SENT These_DT results_NNS suggest_VVP that_IN/that dispositional_JJ and_CC contextual_JJ forces_NNS had_VHD similar_JJ strength_NN in_IN the_DT corresponsive_JJ dynamics_NNS and_CC that_IN/that the_DT contextual_JJ force_NN should_MD not_RB be_VB ignored_VVN in_IN future_JJ studies_NNS ._SENT 
The_DT potential_NN for_IN experiences_NNS to_TO influence_VV CSE_NN has_VHZ important_JJ practical_JJ implications_NNS for_IN organizations_NNS ._SENT In_IN contrast_NN to_TO the_DT implications_NNS of_IN the_DT dispositional_JJ perspective_NN that_IN/that it_PP is_VBZ better_JJR to_TO select_VV employees_NNS with_IN higher_JJR CSE_NN (_( e.g._FW ,_, Judge_NP ,_, 2009_CD )_) ,_, our_PP$ results_NNS support_VVP the_DT potential_NN to_TO cultivate_VV employees_NNS '_POS CSE_NN by_IN enhancing_VVG employees_NNS '_POS positive_JJ work_NN experiences_VVZ (_( e.g._FW ,_, job_NN satisfaction_NN )_) at_IN work_NN ._SENT Job_NN design_NN research_NN suggests_VVZ that_IN/that intrinsically_RB meaningful_JJ tasks_NNS not_RB only_RB contribute_VV to_TO higher_JJR job_NN satisfaction_NN (_( Hackman_NP &_CC Oldham_NP ,_, 1976_CD )_) but_CC also_RB increase_VV self-efficacy_NN at_IN work_NN (_( Parker_NP ,_, 1998_CD )_) ._SENT Therefore_RB ,_, providing_VVG intrinsically_RB meaningful_JJ tasks_NNS is_VBZ one_CD way_NN to_TO foster_VV an_DT employee_NN 's_POS positive_JJ experiences_NNS at_IN work_NN and_CC cultivate_VV an_DT employee_NN 's_POS CSE_NN ._SENT 
Third_JJ ,_, our_PP$ study_NN contributes_VVZ to_TO a_DT further_JJR understanding_NN of_IN the_DT motivational_JJ mechanisms_NNS through_IN which_WDT CSE_NP operates_VVZ ._SENT Previous_JJ studies_NNS of_IN CSE_NNS have_VHP emphasized_VVN a_DT self-verification_NN motive_NN whereby_WRB individuals_NNS higher_JJR in_IN CSE_NNS seek_VVP self-consistent_JJ information_NN and_CC experiences_VVZ (_( e.g._FW ,_, Judge_NP &_CC Hurst_NP ,_, 2008_CD )_) ._SENT In_IN addition_NN ,_, the_DT contextual_JJ effects_NNS reported_VVN in_IN our_PP$ study_NN support_VV a_DT self-enhancement_NN motive_NN whereby_WRB individuals_NNS use_VVP higher_JJR levels_NNS of_IN satisfaction_NN to_TO inform_VV judgments_NNS about_IN their_PP$ self-worth_NN (_( Sedikides_NP &_CC Strube_NP ,_, 1997_CD )_) ._SENT Our_PP$ overall_JJ results_NNS supporting_VVG a_DT corresponsive_JJ relationship_NN between_IN CSE_NN and_CC work_NN experiences_NNS further_RBR suggest_VV that_DT self-verification_NN and_CC self-enhancement_NN processes_NNS may_MD work_VV together_RB over_IN time_NN ._SENT Self-verification_NN and_CC self-enhancement_NN are_VBP often_RB treated_VVN as_IN competing_VVG motivations_NNS when_WRB they_PP are_VBP considered_VVN independently_RB (_( Swann_NP et_NP al_NP ._SENT ,_, 2003_CD )_) ._SENT In_IN contrast_NN ,_, consistent_JJ with_IN the_DT SCENT_JJ model_NN (_( Sedikides_NP &_CC Strube_NP ,_, 1997_CD )_) ,_, our_PP$ findings_NNS suggest_VVP that_IN/that self-verification_NN may_MD serve_VV self-enhancement_NN ends_NNS because_IN people_NNS can_MD increase_VV their_PP$ levels_NNS of_IN CSE_NN after_IN experiencing_VVG higher_JJR job_NN satisfaction_NN over_IN time_NN ._SENT These_DT findings_NNS suggest_VVP that_IN/that people_NNS high_JJ in_IN CSE_NNS tend_VVP to_TO experience_VV greater_JJR job_NN satisfaction_NN not_RB only_RB as_IN verification_NN of_IN their_PP$ positive_JJ self-views_NNS ,_, but_CC through_IN a_DT process_NN of_IN enhancing_VVG their_PP$ positive_JJ self-views_NNS ._SENT For_IN example_NN ,_, people_NNS high_JJ in_IN CSE_NP might_MD perceive_VV more_RBR positive_JJ job_NN characteristics_NNS and_CC report_VV higher_JJR levels_NNS of_IN job_NN satisfaction_NN not_RB only_RB because_IN they_PP �seek_NN and_CC categorize_VV information_NN in_IN their_PP$ work_NN environment_NN that_WDT will_MD lead_VV to_TO positive_JJ conclusions_NNS about_IN their_PP$ work�_NN (_( Judge_NP et_NP al_NP ._SENT ,_, 1998_CD ,_, p_NN ._SENT 21_LS )_) ,_, but_CC also_RB because_IN they_PP actively_RB seek_VVP favorable_JJ work_NN environments_NNS that_WDT allow_VVP them_PP to_TO make_VV positive_JJ judgments_NNS about_IN themselves_PP ._SENT Our_PP$ study_NN design_NN and_CC measures_NNS did_VVD not_RB allow_VV us_PP to_TO test_VV these_DT possibilities_NNS directly_RB ._SENT Further_JJR research_NN should_MD examine_VV these_DT motivational_JJ processes_NNS in_IN more_JJR detail_NN to_TO explore_VV how_WRB self-verification_NN and_CC self-enhancement_NN interact_VVP to_TO influence_VV positive_JJ work_NN experiences_NNS ._SENT 
Limitations_NNS and_CC Future_NP Research_NP 
In_IN addition_NN to_TO the_DT above_JJ implications_NNS ,_, the_DT limitations_NNS of_IN this_DT study_NN should_MD be_VB noted_VVN ._SENT First_RB ,_, we_PP measured_VVD CSE_NNS on_IN only_RB two_CD occasions_NNS ;_: thus_RB ,_, we_PP could_MD not_RB incorporate_VV a_DT growth_NN process_NN for_IN CSE_NN in_IN our_PP$ analysis_NN as_IN we_PP did_VVD for_IN job_NN satisfaction_NN ._SENT Because_IN a_DT growth_NN factor_NN of_IN CSE_NP denotes_VVZ a_DT within-person_NN change_NN of_IN CSE_NN over_IN time_NN ,_, it_PP can_MD be_VB used_VVN as_IN an_DT indicator_NN to_TO reflect_VV the_DT level_NN of_IN self-enhancement_NN and_CC thus_RB could_MD help_VV to_TO unpack_VV the_DT potential_JJ self-enhancement_NN motivational_JJ process_NN behind_IN CSE_NN at_IN work_NN ._SENT In_IN addition_NN ,_, our_PP$ assessment_NN of_IN trait_NN variance_NN in_IN CSE_NP might_MD have_VH been_VBN underestimated_VVN because_IN of_IN the_DT wording_NN of_IN some_DT items_NNS ._SENT To_TO operationalize_VV our_PP$ CSE_JJ construct_NN at_IN the_DT trait_NN level_NN ,_, we_PP used_VVD a_DT CT-C_NP (_( M_NP -_: 1_LS )_) model_VV (_( Eid_NP et_NP al_NP ._SENT ,_, 2003_CD )_) to_TO separate_VV the_DT momentary_JJ variance_NN of_IN CSE_NNS resulting_VVG from_IN the_DT wording_NN of_IN the_DT four_CD items_NNS with_IN a_DT shorter_JJR time_NN frame_NN ._SENT This_DT approach_NN seems_VVZ justified_VVN because_IN the_DT test�retest_JJ reliability_NN of_IN the_DT latent_JJ CSE_NN factor_NN was_VBD relatively_RB high_JJ (_( r_LS =_SYM .63_CD )_) and_CC comparable_JJ to_TO the_DT test�retest_JJ reliability_NN of_IN Big_NP Five_CD personality_NN measures_NNS reported_VVN by_IN Roberts_NP and_CC DelVecchio_NP (_( 2000_CD )_) ,_, whose_WP$ findings_NNS ranged_VVD from_IN .55_CD to_TO .74_CD over_IN a_DT time_NN interval_NN of_IN 6.7_CD years_NNS ._SENT However_RB ,_, future_JJ research_NN should_MD measure_VV CSE_NN at_IN the_DT trait_NN level_NN using_VVG items_NNS without_IN momentary_JJ wording_NN ._SENT 
Second_RB ,_, job_NN satisfaction_NN is_VBZ a_DT relatively_RB simple_JJ indicator_NN that_WDT might_MD not_RB be_VB sufficient_JJ to_TO assess_VV the_DT complex_JJ processes_NNS between_IN CSE_NN and_CC work_NN experiences_NNS ._SENT For_IN example_NN ,_, work_NN experiences_NNS can_MD be_VB captured_VVN at_IN different_JJ levels_NNS of_IN specificity_NN ,_, such_JJ as_IN actual_JJ events_NNS (_( e.g._FW ,_, a_DT job_NN change_NN )_) ,_, perceived_VVN or_CC interpreted_VVN experiences_NNS (_( e.g._FW ,_, perceived_VVN organizational_JJ justice_NN )_) ,_, and_CC an_DT overall_JJ evaluative_JJ judgment_NN (_( e.g._FW ,_, job_NN satisfaction_NN )_) ._SENT In_IN addition_NN ,_, different_JJ work_NN experiences_NNS can_MD influence_VV one_PP another_DT as_RB well_RB as_IN CSE_NN ._SENT For_IN example_NN ,_, Schinkel_NP et_CC al_NP ._SENT (_( 2004_CD )_) reported_VVD that_DT perceived_VVN procedural_JJ fairness_NN will_MD lead_VV to_TO different_JJ interpretations_NNS of_IN a_DT rejection_NN ,_, which_WDT then_RB influences_VVZ the_DT level_NN of_IN CSE_NP ._SENT Their_PP$ finding_NN shows_VVZ that_IN/that the_DT relationship_NN between_IN CSE_NN and_CC work_NN experiences_NNS is_VBZ more_RBR complex_JJ than_IN we_PP have_VHP outlined_VVN here_RB ._SENT Thus_RB ,_, in_IN future_JJ studies_NNS ,_, the_DT indicators_NNS of_IN work_NN experiences_NNS should_MD be_VB expanded_VVN ._SENT 
Finally_RB ,_, the_DT dynamics_NNS between_IN CSE_NN and_CC job_NN satisfaction_NN might_MD be_VB more_RBR complex_JJ than_IN that_DT tested_VVN in_IN our_PP$ research_NN model_NN ._SENT Factors_NNS that_WDT might_MD mediate_VV or_CC moderate_VV paths_NNS in_IN the_DT reciprocal_JJ relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN and_CC its_PP$ growth_NN were_VBD not_RB examined_VVN in_IN our_PP$ study_NN ._SENT For_IN example_NN ,_, it_PP has_VHZ been_VBN theorized_VVN that_IN/that CSE_NP can_MD contribute_VV to_TO higher_JJR job_NN satisfaction_NN and_CC greater_JJR growth_NN by_IN guiding_VVG an_DT individual_NN 's_POS attention_NN to_TO seek_VV positive_JJ features_NNS in_IN the_DT environment_NN and_CC take_VV advantage_NN of_IN opportunities_NNS (_( e.g._FW ,_, Judge_NP &_CC Hurst_NP ,_, 2008_CD ;_: Judge_NP et_NP al_NP ._SENT ,_, 1998_CD )_) ._SENT However_RB ,_, we_PP did_VVD not_RB empirically_RB test_VV these_DT mediated_VVN processes_NNS ._SENT 
In_IN addition_NN ,_, we_PP did_VVD not_RB include_VV potential_JJ moderators_NNS in_IN our_PP$ model_NN ._SENT Judge_NP and_CC Kammeyer-Mueller_NP (_( 2004_CD )_) and_CC Tolli_NP and_CC Schmidt_NP (_( 2008_CD )_) have_VHP indicated_VVN that_IN/that internal_JJ attribution_NN ,_, a_DT tendency_NN to_TO attribute_VV the_DT cause_NN of_IN events_NNS to_TO oneself_PP ,_, can_MD moderate_VV the_DT effect_NN of_IN success_NN or_CC failure_NN in_IN increasing_VVG or_CC decreasing_VVG CSE_NN or_CC self-efficacy_NN ._SENT Also_RB ,_, according_VVG to_TO the_DT contingencies_NNS of_IN the_DT self-worth_NN model_NN (_( Crocker_NP &_CC Wolfe_NP ,_, 2001_CD )_) ,_, self-evaluations_NNS are_VBP made_VVN according_VVG to_TO what_WP people_NNS believe_VVP they_PP need_VVP to_TO be_VB or_CC do_VVP to_TO have_VH worth_JJ as_IN a_DT person_NN ._SENT Thus_RB ,_, concepts_NNS such_JJ as_IN the_DT importance_NN of_IN job_NN performance_NN ,_, involvement_NN in_IN a_DT job_NN ,_, and_CC personal_JJ identity_NN at_IN work_NN may_MD amplify_VV or_CC limit_VV the_DT reciprocal_JJ relationship_NN between_IN CSE_NN and_CC job_NN satisfaction_NN ._SENT 
Conclusion_NN 
In_IN keeping_VVG with_IN the_DT corresponsive_JJ principle_NN of_IN personality_NN development_NN (_( Caspi_NP et_NP al_NP ._SENT ,_, 2005_CD )_) ,_, we_PP found_VVD that_IN/that individuals_NNS '_POS CSE_NNS influenced_VVD their_PP$ work_NN experiences_NNS ,_, which_WDT in_IN turn_RB shaped_VVN their_PP$ CSE_NN ._SENT The_DT corresponsive_JJ perspective_NN incorporates_VVZ the_DT conventional_JJ dispositional_JJ perspective_NN of_IN CSE_NN (_( i.e._FW ,_, CSE_NNS will_MD influence_VV one_CD 's_POS work_NN experiences_NNS ;_: Judge_NP ,_, 2009_CD )_) and_CC the_DT contextual_JJ perspective_NN of_IN CSE_NN (_( i.e._FW ,_, CSE_NP is_VBZ shaped_VVN by_IN one_CD 's_POS work_NN experiences_NNS ;_: Schinkel_NP et_NP al_NP ._SENT ,_, 2004_CD )_) ._SENT Moreover_RB ,_, the_DT corresponsive_JJ perspective_NN suggests_VVZ that_IN/that CSE_NP is_VBZ malleable_JJ and_CC involves_VVZ both_DT self-verification_NN and_CC self-enhancement_NN motivations_NNS ._SENT Overall_RB ,_, this_DT study_NN provides_VVZ a_DT dynamic_JJ view_NN of_IN the_DT relationship_NN between_IN CSE_NN and_CC work_NN experiences_NNS ._SENT 

