# The Value Attribution Problem
## How organizations assign credit for collective outcomes

Proposed by Jean Nickson Pijeau
Version 0.1 | October 2026 | Research update October 19 2026
Conceptual working paper 3 of 5 | Pending empirical validation

## Abstract
Collective outcomes create an identification problem: observing success does not reveal each participant's contribution. The value attribution account within the Corporate Value Triad separates the production of evidence, evaluator exposure, credit judgments, recorded recognition, and later allocation. It proposes that information structure and narrative control affect credit when contributions are ambiguous, while allowing relational work to improve the underlying outcome. The paper distinguishes inaccurate attribution from legitimate accountability and develops a randomized experiment that holds contribution evidence constant. It also addresses joint production, preventive work, unequal status, and the danger of treating organizational records as ground truth. The proposed attribution record is a governance aid, not an algorithm for distributing merit. A generalized reported episode supplies a hypothesis-generating illustration rather than verified findings. No original empirical study results are reported, and the framework remains open to rejection if information structure adds no explanatory value beyond existing accounts.

## 1 Collective success and individual inference
An outcome can depend on several complementary actions. A designer makes a process usable, an operator detects a defect, a coordinator secures a dependency, and a manager accepts risk. Removing any one action could change the result. Assigning all of the value to the last visible action is generally an insufficient description, but adding up separate counterfactual claims can also exceed the total outcome. Joint necessity does not supply unique additive shares.

CVT treats contribution attribution as an inference under incomplete information. An evaluator observes selected traces, brings prior beliefs, and makes a judgment about who did what and why it mattered. Attribution can be accurate, incomplete, inflated, or unresolved. Those categories concern the relationship between a claim and its evidence. They do not establish a person's general competence or intentions.

The model distinguishes accountability from authorship. A formally accountable project owner may deserve credit for decisions and coordination while other people deserve credit for execution. Stating that the owner was accountable is not equivalent to claiming that the owner originated every idea or produced every work product. A useful record permits both statements without forcing a single winner.

## 2 The proposed attribution mechanism
The mechanism has five stages. Work creates traces. Reporting practices select and summarize traces. Evaluators encounter selected information. They assign credit to people and actions. Formal recognition preserves some judgments for future use. Later decisions may rely on these records, creating a feedback loop in which earlier attribution affects subsequent assignment and visibility.

Each stage has an alternative explanation. A sparse trace may indicate missing documentation or little contribution. A summary may omit an action because it is irrelevant, because space is limited, or because its author lacks knowledge. Evaluator exposure may improve accuracy by adding context. The theory predicts distortion only under specified information conditions. It does not assume that every manager's summary is self-serving.

Narrative control means the practical ability to select, frame, or disseminate the account of an outcome. It can reside with a project spokesperson, a customer-facing employee, or an evaluator. Its productive use makes complex work understandable. Its potentially distorting use obscures material contributors or treats unsupported claims as settled fact. Researchers should measure the reporting process rather than infer narrative control from seniority alone.

### Attribution without intentional appropriation
The Collective Attribution Paradox, Attribution Dilution Effect, and Contribution Signaling are proposed extensions. A person can acknowledge a team's success while leaving their own role unspecified. Another can truthfully name both an earlier creator and their own improvements. If an evaluator encounters those accounts differently, unequal visibility may arise even when neither person misrepresents the work. Whether that difference changes actual recognition remains an empirical question.

Intentional appropriation, unintentional omission, legitimate attention to later work, and accurate shared credit are distinct possibilities. The presence of explicit individual attribution does not identify deception. Silence about one's actions does not prove humility, competence, or a deliberate sacrifice. A study must measure statements, available evidence, exposure, and later judgments separately. It should not infer motives from a single exchange.

### Reported qualitative illustration and source limits
A specialist reportedly developed an internal information-management utility independently. A colleague subsequently enhanced it and discussed those enhancements in a group setting that included an evaluator. According to the supplied account, the colleague explicitly acknowledged the original developer and described their own improvements. The original developer reported preferring collective attribution, partly shaped by previous team-oriented leadership experience. The colleague was described as favoring more explicit individual attribution.

This is a generalized participant-reported episode, not a completed qualitative case study. It has no independently checked artifacts, interview transcript, consented second perspective, evaluator account, or outcome measures. The description of the colleague's general style is an informant interpretation. It cannot establish stable traits, deliberate manipulation, omitted credit, unfair promotion decisions, actual recognition differences, or compensation disparities. Because the creator was reportedly credited, the episode does not itself demonstrate attribution dilution.

The episode motivates prospective comparison of communication styles under shared credit. It is not evidence of prevalence or causation. Employee names, employer identifiers, screenshots, product names, exact technical functions, dates of the episode, and distinctive career details are omitted. Generalization reduces identification risk but cannot guarantee anonymity. Before any external use, obtain the company's required confidentiality clearance and appropriate participant permissions, or omit the episode. No such clearance or consent is claimed here. Do not publish the underlying internal exchange as supporting data.

## 3 What existing research permits us to say
Heilman and Haynes (2005 [R05]) show in experimental mixed-sex team scenarios that ambiguous individual contribution can support unequal credit judgments. The setting and manipulated information matter. The result does not establish that ambiguity always benefits seniority, extroversion, or technical status. These would be additional hypotheses requiring their own tests.

Burt (2004 [R02]) links brokerage with valued ideas and career outcomes. For attribution research, the important constraint is that information access may improve the work itself. An employee who brings a new perspective may earn recognition because the idea is better. Munyon et al. (2015 [R04]) likewise make it untenable to define political skill as proof of low productive contribution. CVT must allow information and relationships to affect both production and credit.

Exley and Kessler (2022 [R19]) document context-dependent differences in self-evaluations, motivating separation of communication from performance. VanEpps et al. (2024 [R20]) study impressions formed when self-promotion accompanies credit to others. This makes combined attribution an essential comparator. Ross and Sicoly (1979 [R21]) examine availability and responsibility judgments in joint work. These findings motivate alternatives to an intent-based explanation. They do not validate the proposed adoption mechanism or workplace reward effects.

These are supporting literatures with different constructs. CVT does not introduce an established scale of narrative control or a validated measure of deserved credit. The proposed contribution is a falsifiable synthesis at the level of an outcome episode. A study must show incremental explanation and practical benefit to justify using new terminology.

## 4 Attribution records as evidence structures
A contribution record documents actions and dependencies before a final narrative hardens. It names the objective, evidence window, relevant outcome, uncertainty, contributors, and sources. Participants can correct factual omissions. Reviewers distinguish an observed action from an inferred effect. Where sources disagree, the record preserves competing accounts.

| Record element | Example of an admissible statement | Inference to avoid |
| --- | --- | --- |
| Action | A participant tested the failover procedure on a stated date | The participant caused all subsequent uptime |
| Outcome | A documented defect did not recur during the observation window | The intervention eliminated every future risk |
| Coordination | A dependency owner confirmed a handoff that had been blocked | Attendance at the meeting proves value |
| Joint dependency | Two actions were required before release | Each participant independently created the full outcome |
| Uncertainty | No reliable record identifies who originated the change | The person who spoke first must have originated it |

The record should remain proportionate to the decision. A routine handoff does not justify a burdensome dossier. Where work is sensitive, authorized reviewers can inspect evidence without reproducing it in a general file. The audit must avoid collecting customer secrets, private messages, or personal information simply because more documentation is possible.

The record itself can distort behavior. People may favor documented tasks, spend time writing persuasive accounts, or crowd out contributors less comfortable with the reporting format. A field evaluation must measure administrative burden, omitted work, and access to correction. Documentation is an intervention with costs, not a neutral window onto reality.

### Preserving origin and later improvements
A contribution history records the original action and the later actions that changed, maintained, deployed, or coordinated it. Distinguish factual provenance, formal accountability, current stewardship, and legal ownership. A record can preserve who originated a design while recognizing that others created most of its current utility. The scholarly CRediT standard [R22] is a precedent for describing differentiated roles; it is not an employment evaluation standard or a validated CVT instrument.

Use an authorized, minimal record with a study code, stage, broad time window, originating action, later action, corroboration, uncertainty, and correction status. Record unavailable history as unavailable, not as proof that no originator exists. Keep source evidence in authorized systems rather than copying private correspondence. Once an action is corroborated, later narratives can be checked for acknowledgment of that same action without requiring a fixed share of total credit.

The new Contribution Attribution Assessment compares an employee's neutral episode description, a manager's independent account, and this history. Agreement is evidence of consistency, not proof of complete truth. Contradictions require review. No percentage of matched actions can become a fairness score or employee ranking. The protocol's H9 to H13 separates observational questions from a factorial experiment on communication, exposure, and structured records.

## 5 A randomized test of information structure
The proposed H2 experiment crosses presenter prominence with evidence format. Adult evaluators receive a fictional project in which contributions are explicitly equal in relevant quality and scope. In the prominence condition one participant presents the result, with the script stating that presentation adds no extra work or responsibility. In the control condition neither participant receives distinctive presentation exposure. Within each, evaluators see either a concise narrative or the same facts organized as a structured contribution record.

The primary outcome is credit assigned to the focal participant on a 0 to 100 scale. Because this is a designed judgment outcome with known fictional facts, its interpretation is narrower than an organizational fairness measure. H2 predicts that the prominence-induced difference is smaller under the structured record. Random assignment identifies the effect of presentation and format in this vignette population, subject to implementation, compliance, and missing outcomes.

All substantive facts, word counts as closely as feasible, and outcome quality must remain constant across formats. A pretest checks readability and whether evaluators interpret the presenter as having additional responsibilities despite the script. If they do, the manipulation does not isolate irrelevant prominence and requires redesign before confirmation. An attention failure should not become a post hoc exclusion that changes the result. The primary analysis includes randomized participants with observed outcomes, with exclusions and missing-data sensitivity specified in advance.

The first experiment need not manipulate protected characteristics. A later study of status or demographic expectations would require a separate ethical review, careful stimuli development, and adequate precision for interactions. The current design must not borrow evidence from gender research to assert untested effects for every group.

## 6 Connecting judgment to workplace recognition
The vignette establishes only a bounded judgment effect. A field study can ask whether contribution records change formal recognition while preserving outcomes and reducing omissions. Teams could receive a standardized record process or retain usual practice where an independent review permits. Randomization should occur at a level that limits sharing of the process between conditions. The record must supplement normal safeguards and must not withhold required acknowledgment or employee rights.

The principal field endpoint could be the proportion of independently corroborated actions that receive accurate acknowledgment in a review. The denominator requires a prespecified relevance rule and blinded adjudication where feasible. It should be reported with unresolved cases and source coverage. Researchers must not call a reviewer consensus ground truth. Agreement can reflect shared assumptions as well as accuracy.

A further distinction concerns mediation. A record might change both evaluator credit and final recognition, but that pattern alone does not identify a causal indirect effect. Mediation requires additional assumptions about confounding between mediator and outcome and about post-treatment variables (Imai et al., 2010 [R15]). The protocol treats mediation as secondary and assumption-dependent.

## 7 Rival explanations and disconfirmation
The comprehension rival predicts that a structured record improves general understanding rather than specifically correcting prominence. Measure understanding separately and compare matched formats. The accountability rival predicts that evaluators rationally treat presentation as responsibility. The script and pretest must address this possibility. The demand-characteristic rival predicts that participants infer the experiment's desired response. Neutral instructions and questions about interpretation help evaluate that concern.

The productive communication rival predicts that presentation itself contributes value. That rival is plausible in real organizations and only held constant in the vignette. A field analysis must record communication that secures adoption, clarifies decisions, or prevents rework. Removing that work from contribution would unfairly privilege visible technical artifacts and bias the result.

H2 would receive evidence against its predicted direction if a sufficiently precise interaction showed that the record amplifies the prominence effect. It would receive evidence of practically negligible benefit if an equivalence analysis, justified before results, placed the attenuation within a prespecified negligible range. An imprecise null supports neither conclusion. Persistent failure to separate contribution facts from format effects would undermine the mechanism's testability.

## 8 Implications for organizational governance
The most defensible immediate application is a correction process. An employee should be able to identify an omitted action, supply appropriate evidence, and receive a documented response. A reviewer should distinguish factual correction from disagreement over the value of the action. The organization can acknowledge shared work without deciding a universal ranking.

Attribution review should also examine who controls the evidence channel. If only managers can submit records, structured documentation may preserve existing asymmetry. If everyone must publicly contest a powerful colleague's claim, nominal access may be unusable. Confidential review, alternative evidence formats, and separation from punitive decisions are proposed safeguards. Their usefulness requires local evaluation.

## 9 Conclusion
The value attribution problem concerns the inference from collective outcomes to individual credit. CVT proposes that information structure can change this inference even when relevant work remains constant. The test is demanding: contribution evidence, legitimate relational work, comprehension, and role responsibility must remain distinguishable. A structured record offers a testable intervention, while uncertainty and joint production remain part of the result rather than defects to conceal.


## References

[R02] Burt, R. S. (2004). Structural holes and good ideas. American Journal of Sociology, 110(2), 349-399. https://doi.org/10.1086/421787

[R04] Munyon, T. P., Summers, J. K., Thompson, K. M., & Ferris, G. R. (2015). Political skill and work outcomes: A theoretical extension, meta-analytic investigation, and agenda for the future. Personnel Psychology, 68(1), 143-184. https://doi.org/10.1111/peps.12066

[R05] Heilman, M. E., & Haynes, M. C. (2005). No credit where credit is due: Attributional rationalization of women's success in male-female teams. Journal of Applied Psychology, 90(5), 905-916. https://doi.org/10.1037/0021-9010.90.5.905

[R15] Imai, K., Keele, L., & Tingley, D. (2010). A general approach to causal mediation analysis. Psychological Methods, 15(4), 309-334. https://doi.org/10.1037/a0020761

[R19] Exley, C. L., & Kessler, J. B. (2022). The gender gap in self-promotion. The Quarterly Journal of Economics, 137(3), 1345-1381. https://doi.org/10.1093/qje/qjac003

[R20] VanEpps, E. M., Hart, E., & Schweitzer, M. E. (2024). Dual-promotion: Bragging better by promoting peers. Journal of Personality and Social Psychology, 126(4), 603-623. https://doi.org/10.1037/pspi0000431

[R21] Ross, M., & Sicoly, F. (1979). Egocentric biases in availability and attribution. Journal of Personality and Social Psychology, 37(3), 322-336. https://doi.org/10.1037/0022-3514.37.3.322

[R22] National Information Standards Organization (2022). ANSI/NISO Z39.104-2022 CRediT Contributor Roles Taxonomy. NISO. https://doi.org/10.3789/ansi.niso.z39.104-2022