The present study attempts to see both the forest and the trees
that is, describe, explain, and model alternative, configurational, asymmetric, casebased configurations of how individual and industry sub-categories, job stressors, core self-evaluation theory, and job strain identify high as well as low job satisfaction (JS). The study’s use of asymmetric case-based modeling also includes separate models indicating either high or low JS. The study provides case-level model profiles that are high in accuracy consistently in predicting managers high (and separate models for managers low) in JS. Thus, the study focuses on case-based modeling using somewhat precise outcome testing (SPOT, Woodside, 2016) and avoids the fatal flaws in using null hypothesis statistical testing (NHST) (Armstrong, 2012; Gigerenzer & Brighton, 2009; Hubbard, 2016; Trafimow, 2014; Trafimow & Marks, 2015) and the flaws in examining the relative sizes of betas in regression models (Armstrong, 2012; Hubbard, 2016). The study contributes to the literature by describing how complexity theory and configurational analysis applies in constructing asymmetric models in case-based research on JS. The study advances McClelland’s (1998) algorithm asymmetric analysis, with predictive validation using additional samples, to solve the pervasive current mismatch between theory and analysis (Fiss, 2011) in human resource management (HRM) research. This asymmetric research perspective rests on a foundation of complexity theory. Adopting asymmetric perspective goes beyond the dominant logic in the literature of symmetric, variable-based, theory construction/testing. The asymmetric approach to theory construction and data analysis recognizes and models cases supporting main effects hypothesis (e.g., generalized self-efficacy associates positively with JS) as well as cases exhibiting relationships contrarian to such symmetric hypothesis (e.g., high-generalized self-efficacy contributes to low JS in some contexts). Complexity theory and asymmetric analysis go beyond the empirically support of small, medium, and large main effects of relationships of independent on dependent variables. For example, a complexity theory tenet suggests the need for modeling the configuration of causes that include contrarian associations in JS research, such as for cases (employees or managers) where high job stress associates with high job performance; such cases occur in possibly all studies with moderate-to-large sample sizes but are typically ignored in studies focusing on the general finding of a modest effect size, negative, main effect for job stress and JS. Rather than adopting a symmetric stance, complexity theory supports the perspective that a configurational asymmetric perspective is necessary for examining complex antecedent conditions to achieve deep understanding and for reporting complex wholes of causes
because different cases occur whereby job stressors and job satisfaction relationships support and run counter to intuitive associations as well as cases where the same job stressors do not associate with job satisfaction.