EV-0031
Helping shows up in unit performance, and the timing suggests it causes it
Podsakoff, N. P., Whiting, S. W., Podsakoff, P. M., & Blume, B. D. (2009). Journal of Applied Psychology, 94(1), 122-141 · Source
What this entitles us to say
A meta-analysis of 168 samples and 51,235 individuals finding that discretionary helping behavior relates to individual outcomes such as performance ratings and reward decisions, and to unit-level performance - with the relationship stronger in longitudinal than cross-sectional studies.
What it found
168 independent samples, 51,235 individuals. Citizenship behavior related to managerial ratings of performance, reward allocation decisions, and withdrawal-related outcomes at the individual level, and to unit-level performance measures at the organizational level.
The methodological note is the valuable part: relationships with unit performance were stronger in longitudinal than in cross-sectional studies, which the authors offer as some evidence that citizenship behaviors are causally related to those outcomes. Most of this library is cross-sectional; this is one of the few entries with a directional hint in it.
Why it matters here
It is the business case for the trait, and it is stated at the level a buyer cares about - the unit, not the individual. It also gives the Engagement region its first outcome-level anchor.
Read alongside EV-0030, the pattern is consistent: the behaviors that make up selflessness are among the better-supported things in the leadership literature, which is unusual for a virtue.
The caution worth teaching: part of what this measures is that helpful people get rated higher. That is a real organizational fact, and it is also exactly the sort of finding a cynic can use to argue that citizenship is impression management. Both readings should be in the module.
The limits
The individual-level findings include how managers rate and reward people, which is not the same as how well those people actually do their jobs - helpful people may simply be liked more. The longitudinal hint is evidence toward causality, not proof of it. Nothing here says how much helping is optimal, and EV-0032 documents what it costs the helper. Verified to abstract level.
What we do not say
"Helping causes performance." A hint from longitudinal designs is not a causal claim. Nor should the individual-level findings be described as performance when part of what was measured is how managers rate and reward people.