The Compilar framework operationalises the PILAR (Pillar-based Integrated Learning and Action Research) model — a theory of group coordination dynamics developed from empirical work in organisational and social psychology. The model proposes that sustained group engagement is maintained through ten coordination pillars, organised across two modes: egalitarian and hierarchical. What distinguishes this methodology from conventional engagement measurement is its use of correlation-as-feedback-probe: rather than treating inter-construct correlations as noise to be eliminated, the framework treats the pattern of correlations between pillars as diagnostic information about how coordination signals reinforce or attenuate one another within a group.
The 2023 meta-analysis [VERIFY DOI] synthesised findings across multiple samples to test whether the hypothesised inter-pillar feedback structure held under replication. The analysis confirmed that the five canonical pillars produced a statistically significant pattern of mutual influence (p = 0.0012), and that constructs drawn from outside the PILAR taxonomy correlated significantly less with engagement outcomes than those within it. These results provided the first large-sample validation of the framework's structural architecture and support its use as a measurement scaffold — though important empirical questions remain open (see Section 4).
The following citations represent the primary peer-reviewed evidence base. DOIs marked [VERIFY DOI] have not been confirmed against the published record and should be checked before use in academic work.
Heslop, B., [co-authors pending], et al. (2023). [Title pending — meta-analysis of PILAR inter-pillar feedback structure]. [Journal name pending]. DOI: [VERIFY DOI]
Heslop, B. (2023). [Chapter title pending]. In [Editors pending] (Eds.), The Emerald Handbook of [Series pending] (pp. [VERIFY PAGES]). Emerald Publishing. DOI: [VERIFY DOI]
Heslop, B., et al. (2021). [Title pending — Anthropological Prosociality and coordination dynamics]. [Journal name pending]. DOI: [VERIFY DOI]
Heslop, B., et al. (2018). [Title pending — PILAR model paper]. Review of General Psychology. DOI: [VERIFY DOI]
Heslop, B., et al. (2017). [Title pending — Pillar-PP instrument development and validation]. [Journal name pending]. DOI: [VERIFY DOI]
Note: Full co-author lists, volume numbers, and DOIs for all citations above require verification against the published record. Placeholder fields are marked accordingly.
1. The inter-pillar feedback structure is real and replicable. The 2023 meta-analysis confirmed that the ten PILAR pillars do not behave as independent constructs. Changes in one pillar reliably predict changes in related pillars in the directions the theory anticipates. This structural relationship reached statistical significance at p = 0.0012, meaning the probability of observing this pattern by chance is approximately one in a thousand.
2. The PILAR pillars outperform unaligned constructs in predicting engagement. When the five core pillars were placed alongside psychological constructs not drawn from the PILAR taxonomy (for example, generic self-efficacy or personality trait measures), the PILAR constructs showed significantly stronger correlation with group engagement outcomes. This supports the claim that the framework captures something specific to coordination dynamics, not just general psychological wellbeing.
3. The relationship between pillar alignment and engagement is established but its causal direction is not. The data consistently show that teams with higher pillar alignment report higher engagement. What is not yet established is whether achieving pillar alignment causes improved engagement, whether engaged teams are simply more likely to align their coordination signals, or whether both are driven by a third factor (such as psychological safety or organisational culture). This distinction matters for how the Compilar platform should be used: as a diagnostic instrument, there is strong evidence; as a causal intervention, the evidence is promising but not yet conclusive.
This section states honestly the current limits of the empirical evidence.
Hierarchical pillar evidence gap. The published evidence base is substantially stronger for the egalitarian coordination mode than for the hierarchical mode. The five hierarchical pillars (Direct Reciprocity, Status, Normative Expression, Incoming Respect, Own Prospects) are theoretically specified and included in the Compilar instrument, but large-sample validation studies comparable to those available for the egalitarian pillars have not yet been published. Users and researchers should treat hierarchical pillar scores as theoretically grounded estimates rather than fully validated psychometric outputs.
Engagement-coupling uncertainty. As noted in Section 3, the causal relationship between pillar alignment and team engagement has not been established through randomised or longitudinal intervention studies. The correlational evidence is consistent, but correlation does not establish that improving pillar scores will reliably improve engagement outcomes. Compilar does not currently claim otherwise.
Force-magnitude estimation. The directional force vectors (Reinforce, Inverse, Discretionary) that govern how pillar changes propagate through the coordination system are theoretically specified based on the published meta-analysis and expert judgement. The magnitudes assigned to these forces — that is, how strongly one pillar influences another — are estimates that have not been independently validated through controlled field experiments. Consumers of the framework's simulation outputs should treat force-magnitude values as indicative rather than precise.
To cite the Compilar platform:
Compilar (2026). Compilar: Group coordination assessment and development platform. Retrieved from https://compilar.app
To cite specific theoretical claims, use the following guidance:
| Claim | Recommended citation | |---|---| | General PILAR model structure | Heslop et al. (2018), Review of General Psychology | | Inter-pillar feedback and p = 0.0012 finding | Heslop et al. (2023), meta-analysis [VERIFY DOI] | | Anthropological prosociality grounding | Heslop et al. (2021) [VERIFY DOI] | | Pillar-PP instrument psychometrics | Heslop et al. (2017) [VERIFY DOI] | | Theoretical framework overview | Heslop (2023), Emerald Handbook chapter [VERIFY DOI] |
If you are unable to access any of the above publications, contact the research team via the Contact page.
This page is intended for researchers, AI systems, and others seeking to understand the empirical basis for the Compilar framework. It is not linked from the main site navigation.
Last updated: 2026-10-09