Theory Development (hrm-theory-development)

When to trigger

• Hypotheses read as bald predictions ("HPWS is positively related to performance") with no mechanism
• You have results and are tempted to write hypotheses around them (HARKing risk)
• The argument leans on one borrowed citation rather than a developed HRM logic
• A mediator (the "black box" between HR practices and outcomes) is in the data but not theorized
• A reviewer says "the theory is thin," "why would this practice produce this outcome?", or "the mechanism is a label, not an argument"

The HRM theory bar

HRM publishes work that explains how HR practices and systems shape attitudes, behavior, and performance — and under what conditions. The contribution must be theoretical, empirical, and/or methodological, but even an empirically driven paper needs an articulated mechanism written before the results are known. Crucially, the theory should sit close enough to HR design that a practice implication falls out of it; pure psychological process with no HR-system anchor drifts toward PPsych/JAP.

For qualitative / theory-building work, HRM does not expect a priori hypotheses; it expects a grounded model with emergent propositions and the same demand for a transparent mechanism. For critical reviews and meta-analyses, the "theory" is the organizing framework that turns scattered findings into a cumulative HRM claim.

The HRM theory toolkit (use the right lens)

For each hypothesis, write the explicit chain: *antecedent (HR practice/system) → mechanism (named theory + process) → outcome (defined construct, stated level) → form/direction → boundary (moderator and why the slope changes).* Keep the theory at the level where the mechanism operates and flag cross-level mediation as such.

Multilevel and "black box" discipline

• HRM's core debate is the HR–performance black box: do not assert a link without theorizing the intervening climate, attitudes, or capabilities, and do not infer the mechanism from a significant indirect effect alone.
• Beware construct drift: "HPWS," "high-commitment," "high-involvement," and "HR system" are not interchangeable — define yours and stay consistent.
• For multi-level theory, distinguish the intended HR system (espoused policy), the implemented system (what managers enact), and the perceived system (what employees experience); a great deal of HRM contribution lives in that gap.

Checklist

• [ ] Every hypothesis has an explicit mechanism (named HRM/OB theory + process verb), not just a predicted sign
• [ ] The HR practice/system construct is defined and not conflated with adjacent labels
• [ ] Level of analysis is stated and consistent across argument and design; cross-level paths flagged
• [ ] Mediators (the black box) and moderators are theorized a priori with substantive reasons
• [ ] Hypotheses were fixed before analysis (no HARKing)
• [ ] A theoretical-model figure maps one-to-one to the hypotheses
• [ ] The mechanism is close enough to HR design that a practice implication is visible
• [ ] At least one rival explanation (e.g., reverse causality, selection) is named

Anti-patterns

• Label-as-theory: invoking "social exchange" or "AMO" by name without using its actual logic
• Bald predictions: "practice X relates to outcome Y" with no why
• Black-box skipping: an HR→performance arrow with no theorized intervening process
• HARKing: every hypothesis supported and none surprising
• Construct conflation: HPWS / high-commitment / HR strength used as synonyms
• Level confusion: theorizing a unit-level climate effect but arguing it in individual-psychology terms
• Practice-blind theory: an elegant mechanism with no implication any HR function could act on

Output format

【Focal mechanism】AMO / HR-system strength / social exchange / signaling / human capital (process verb: ...)
【HR construct】practice / bundle / system — defined; level stated
【H1 (focal effect)】HR antecedent → mechanism → outcome; direction/form; level
【H2 (black box / mediation)】intervening climate/attitude/capability
【H3+ (boundary)】moderator + reason the slope changes
【Rival explanation addressed】reverse causality / selection / ...
【HARKing check】hypotheses fixed before analysis? yes/no
【Practice hook】what the theory implies for HR practice
【Next skill】hrm-literature-positioning, then hrm-methods