To better understand the mediating effects of emotions and job satisfaction on themotivation- performance association, it is critical to examine both within- and between-person effects as they address different research questions. As the within-person effectsfocus on short-term changes, they are appropriate for examining the linkage betweendaily fluctuations of motivations, affective emotions and daily performance. In contrast,the between-person effects are better suited for addressing the more lasting associationsobserved among employees, such as the relationships between motivation, chronicemotionalexpressionandgeneralperformanceevaluation.Thus,simultaneousconsiderationofwithin-andbetween-personrelationshipshasstronglybeenrecommendedforfullyevaluatingtheoreticalmodels(CurranandBauer,2011).Analyzinglongitudinalnesteddatawithamultilevelapproachwillenableanexaminationoftheextenttowhichemployee’sintra-individualmotivationandperformance varies in real-time situations, as well as allowing a comparison of thedifferentlevels.However,previousmultilevelresearchdesignshaveprovidedinconclusive and mixed results in comparing within and between levels of analysis.Some investigators have suggested that variables at the within-person level are morestrongly associated than at the between-person level (e.g. Fisher and Noble, 2004).419Employeemotivation,emotions, andperformance
Yet, others have argued that motivational mechanisms showed stronger relationships atthe between-person level than at the within level of change (Broseet al., 2010). Thus wehypothesize the following:H5.Within- person vs between-person effects are manifested differently such that thebetween-subject analysis will explain more of the job performance variance than thewithin-subject analysis.MethodParticipantsParticipants comprised employees from a variety of occupations working for differentIsraeli organizations. A total of 116 respondents completed the questionnaire for all ninedaysandwereincludedinoursample.Intotal,1,044responseswerecollected(116 respondents×9 days). Of the final sample, 59 percent were female and 58 percent weremarried or in a relationship. Respondents’mean age was 31.76 years (SD¼11.02), andaverage weekly work hours was 39.46 (SD¼13.50) and an average 4.86-year work tenure(SD¼3.20). Education levels ranged from 10 to 22 years (M¼13.99, SD¼2.88). Of therespondents, 15.5 percent had managerial positions.ProcedureResearch assistants distributed the questionnaire and diary using referral sampling. Datacollectors personally described the research objectives to potential respondents, and then sentemails that included a description of the purpose of the study and assurance of anonymity ofall responses, instructions regarding completion of the surveys and a general questionnaire.