EV-0014
Staying because you feel you ought to is its own kind of commitment
Meyer, J. P., Stanley, D. J., Herscovitch, L., & Topolnytsky, L. (2002). Journal of Vocational Behavior, 61(1), 20–52 · Source
What this entitles us to say
A meta-analysis of 155 studies finding that commitment arising from felt obligation (normative) is distinguishable from commitment arising from attachment (affective) and from cost of leaving (continuance), and that all three relate negatively to turnover — evidence that obligation-based commitment is a real and separable thing.
What it found
A meta-analysis of 155 research reports on the three-component model of organizational commitment. The three forms — affective (I want to), continuance (it would cost me to leave) and normative (I ought to) — were found to be related but distinguishable from one another, and from job satisfaction, job involvement and occupational commitment. All three related negatively to withdrawal thinking and to turnover. Affective commitment had the strongest and most favorable relationships with organization-relevant outcomes. No unique antecedents of normative commitment were identified.
Why it matters here
Normative commitment is obligation arising from an agreement — the closest measured cousin to CDL's definition of faithfulness. Its existence as a separable construct supports the framework's claim that keeping your word is not merely a symptom of liking your job.
It also draws the line between Faithfulness and Commitment, which are neighbors in the model and sit in different regions. Commitment in the book's sense is closer to affective — the wanting. Faithfulness is closer to normative — the owing, and specifically the owing you took on yourself. Jeff's phrasing is "loyalty by choice," and normative commitment is very nearly that in academic language.
The honest complication: the authors could not find anything that predicts normative commitment uniquely. If the trait cannot be told apart from affective commitment by its causes, that is a genuine question for a model that separates them — and it should be carried into the module design rather than buried here.
The limits
Correlational and self-reported throughout. Importantly, the authors found no unique antecedents of normative commitment — nothing predicted it that did not also predict affective commitment — which is the weakest point in the construct and must be conceded. Affective commitment, not normative, had the strongest and most favorable relationships with organizational outcomes; obligation is not the most valuable form of commitment, only a distinct one. This literature is about commitment to an employer, which is a narrower thing than faithfulness to a promise. Verified to abstract level.
What we do not say
"Obligation-based commitment is what drives performance." The opposite is closer to true in this study — affective commitment had the stronger relationships with outcomes. Faithfulness is being claimed here as foundational, not as the most productive trait a person can have.