The Regenerative Dignity Index™
Operationalising dignity, pride, and belonging as primary outcomes of place-based regeneration — with the Community Pride Quotient™ and Dignity Delta™
Most development interventions claim dignity as an outcome but measure only income, material change, or behavioural compliance. The literature on dignity in development is rich but largely conceptual; existing instruments tend to focus on the individual rather than the place, and on dignity violation rather than dignity uplift. The Regenerative Dignity Index™ (RDI) operationalises place-based dignity through five pillars — Pride Uplift, Belonging & Identity, Emotional Safety, Daily Experience, and Community Cohesion — measured via Likert-scaled community survey instruments and aggregated to a 0–100 composite score. Two derived metrics, the Community Pride Quotient™ (CPQ) and the Dignity Delta™, capture pride concentration and pre/post-intervention change respectively. This paper presents the conceptual basis, the survey instrument design, the scoring approach, validation methodology, and limitations. The framework is the first place-based dignity measurement system the authors are aware of in the published or grey literature.
I. The dignity gap in development measurement
Dignity appears in nearly every contemporary development framework. The Sustainable Development Goals reference it explicitly in the preamble. NGO mission statements invoke it routinely. Multilateral grant frameworks list it as a cross-cutting outcome. Yet the question of how, in practice, a programme reports on dignity at the end of a quarter has no widely accepted answer. The gap between the rhetorical centrality of dignity and the operational thinness of its measurement is large and, in our view, embarrassing for the sector.
The reasons are structural. Dignity is not directly observable; it must be inferred from self-report, observation, or proxy. Self-report instruments carry well-known biases. Observation is labour-intensive and ethically delicate. Proxies (income, education, housing) are easy to measure but only loosely correlated with the construct of interest. So practitioners default to what is measurable, and report on dignity only in narrative form — the unfortunate outcome that dignity, the thing we say matters most, becomes the thing we measure least.
The academic literature has not been silent on this. Donna Hicks's work on dignity in conflict resolution catalogues ten elements that contribute to or violate dignity in interpersonal and institutional encounters.[1] Amartya Sen's capability approach, and Martha Nussbaum's extension of it to a list of central capabilities, provide a philosophical scaffold for thinking about what dignified human functioning entails.[2][3] The place attachment literature in environmental psychology — Hidalgo and Hernandez, Scannell and Gifford — addresses the related question of how individuals relate to places, which dignity in place must in some sense extend.[4][5] Robert Sampson's work on Chicago neighbourhoods established the importance of community-level effects on individual outcomes that are not reducible to the characteristics of individuals.[6]
What is largely missing from this rich conceptual tradition is a worked, operational instrument for place-based dignity measurement at programme scale — one that a programme manager can administer with a clipboard, score with a calculator, and report to a funder with a confidence interval. The Regenerative Dignity Index™ is our attempt at such an instrument. We do not claim it captures everything dignity could mean; we claim it captures enough to be useful, with declared limitations.
II. Conceptual foundations
The RDI is built on three conceptual commitments, which together define the construct we mean to measure.
Dignity is place-based, not only personal
An individual can hold dignity privately; a community holds it publicly. The distinction matters operationally. A development programme cannot, in any direct sense, alter the inner dignity of a person — that work belongs to the person. What a programme can alter is the conditions of dignity in the place: the streets, the institutions, the shared narratives, the daily encounters. The RDI measures the latter, not the former. This is a deliberate scoping choice.
Dignity uplift matters as much as dignity violation
The literature on dignity violation — particularly in human rights and humanitarian contexts — is well developed. The literature on dignity uplift, the positive direction of the construct, is thinner. Both matter, and a regeneration programme is concerned principally with the latter: not whether dignity is being violated (a baseline humanitarian question), but whether it is rising.
Dignity is multi-dimensional
The construct cannot be collapsed to a single underlying factor. Pride in place is not the same thing as feeling emotionally safe in that place; both are constitutive of dignity but they can move in different directions and for different reasons. A five-pillar structure captures the principal facets we observed during instrument development in the Adikpo programme without inflating the construct unmanageably.
III. The five pillars
The Index is constructed from five pillars, each measured by a small set of Likert-scaled survey items and combined into a composite. The pillar definitions and rationales are below.
| Pillar | Construct | Sample item |
|---|---|---|
| Pride Uplift ★ | The extent to which the respondent feels increased pride in their place over a defined recent period. | "I feel prouder of where I live today than I did a year ago." |
| Belonging & Identity ⌂ | The depth of identification with the place and its people. | "This is my place. The people here are my people." |
| Emotional Safety ⛨ | The felt sense of safety in moving through and inhabiting the place. | "I feel safe walking through my neighbourhood after dark." |
| Daily Experience ☀ | The texture of ordinary life: cleanliness, dignity of public space, encounters. | "My street looks like a place people care about." |
| Community Cohesion ❀ | The strength of the social fabric — trust, reciprocity, collective efficacy. | "If a child here was in trouble, a neighbour would help." |
Each pillar is measured by five items (one shown above, four further), giving a total instrument of twenty-five items administered to a representative sample of community households. Item development drew on Donna Hicks's ten elements of dignity, Robert Sampson's collective efficacy scale, and the place attachment literature, with extensive field testing in Adikpo during the instrument development phase.
IV. Survey instrument design
Likert structure
Each item is rated on a 5-point Likert scale: Strongly disagree (1) · Disagree (2) · Neither (3) · Agree (4) · Strongly agree (5). The 5-point scale was chosen over 7-point on the basis that 5-point scales translate more reliably across the language and literacy contexts in which the instrument is administered, with marginal loss of granularity at the composite level.
Administration
The instrument is administered face-to-face by a trained field officer, in the respondent's primary language, with verbal informed consent recorded by audio. Reading the items aloud equalises the burden across literacy levels and removes the literacy bias that self-administered instruments introduce. Field officers are trained in non-leading administration, with periodic recalibration sessions.
Sampling
A stratified random sample of households within the catchment is drawn for each measurement wave. Stratification variables are: street/zone (to ensure spatial coverage), household head gender, and a programme-engagement variable (engaged vs. non-engaged households, where the latter serve as a within-community comparison group). Target sample size is calibrated to produce a 95% confidence interval of ±5 points on the composite RDI for the relevant population.
Language
The instrument is developed in English and translated into the locally dominant languages by a back-translation process: an independent translator returns the translated items to English, and discrepancies are reconciled with the original drafter. We retain the original English items as the canonical version; the field versions are flagged as locally-adapted with the dates of last back-translation check.
V. Scoring and composite construction
Raw Likert scores (1–5) are transformed to a 0–100 scale per item using a simple linear mapping:
Each pillar score is the arithmetic mean of its five constituent items, averaged across respondents in the sample. The composite RDI is the arithmetic mean of the five pillar scores.
The default equal weighting follows the same rationale as the ACES Circularity Index™: we have no empirical basis for differential weighting, the five pillars are conceptually distinct, and the OECD/JRC composite indicator handbook recommends equal weighting in the absence of theoretical or statistical justification for variation.[7]
VI. The Community Pride Quotient™ and the Dignity Delta™
The composite RDI gives a single-number summary, but two complementary metrics carry information the composite hides.
Community Pride Quotient™ (CPQ)
The CPQ is the proportion of respondents in the sample who score 80 or above on the Pride Uplift pillar. Where the RDI tells us the average pride level, the CPQ tells us how concentrated pride is at the high end: a community can have a moderate average and a thin tail of strongly-proud people, or it can have the same moderate average with many people clustered around the threshold. The two have different implications for what the programme should do next.
Dignity Delta™
The Dignity Delta is the difference between two RDI measurements, taken at different points in time on the same community (and where possible, the same respondents). It is the principal metric for evaluating the effect of a regeneration intervention.
We strongly prefer paired-respondent designs (the same individuals interviewed at t0 and t1) where logistically feasible, as paired designs have substantially more statistical power for detecting modest changes. Where respondent turnover or migration makes pairing impractical, we fall back to independent samples drawn by identical stratification.
VII. Validation: triangulation, reflexivity, and field testing
Self-report instruments measuring subjective constructs face a particular validity challenge: there is no external "true score" against which to validate. We address this through a combination of strategies.
Construct triangulation
The survey instrument is administered alongside two structurally different data collections in each wave: (i) a short structured observation of public space by an independent observer, scoring cleanliness, presence of new investment, and signs of communal care; (ii) a small set of qualitative interviews with key informants (teachers, traditional leaders, business owners) using a semi-structured guide. The three sources should broadly agree. Persistent divergence between survey results and observational/interview signals is treated as a validity concern, not as evidence that the survey is the truth.
Internal consistency
Each pillar's five items should hang together as a coherent scale. We target Cronbach's α of 0.70 or above per pillar, calculated at each measurement wave. Items consistently failing internal consistency are candidates for revision in a future version of the instrument.
Test-retest stability
A small subset of respondents (10%) is re-interviewed within 14 days by a different field officer. We target an intraclass correlation coefficient of 0.75 or above at the pillar level. The 14-day window is short enough that genuine change is unlikely; substantial discordance suggests instrument unreliability.
Reflexivity
The Foundation maintains a reflexive practice: each quarterly measurement is accompanied by a short reflective note from the field officer team identifying contextual events during the measurement period that may have shifted respondent mood independently of the programme (a public funeral, an election, a security incident, a national news event). The reflective note travels with the data so that downstream readers can interpret the numbers in context.
VIII. Limitations and open questions
Self-report bias. All Likert-scaled subjective measurement is vulnerable to social desirability bias, acquiescence bias, and central tendency bias. We mitigate through item wording, administration training, and triangulation, but we cannot eliminate. RDI scores should be read as a measurement of self-reported place-based dignity, and the qualifier should not be dropped.
Cultural translation. The construct of dignity does not map identically across languages and cultures. The English-language instrument and its locally translated versions are not strictly equivalent; we treat the local-language versions as the canonical field instrument and disclose the back-translation date in every published RDI report. Cross-cultural comparison of RDI scores between communities with markedly different linguistic and cultural contexts is not currently advised.
The pride/dignity boundary. Pride and dignity are conceptually adjacent but not identical. The RDI's first pillar — Pride Uplift — treats pride as the principal observable proxy for the dignified felt-sense of place. This is a defensible operational simplification but is not the only available choice. We anticipate that future versions may distinguish them more sharply.
The control community question. A Dignity Delta is most interpretable when compared against a non-intervention community measured at the same two time points. We acknowledge this and are evaluating the feasibility of paired-community designs in future programme waves. The methodological burden is substantial and the budget envelope is real.
What we do not yet know. The Index has been developed against a single context (Adikpo). Item-level performance, internal consistency, and the calibration of the CPQ threshold (currently 80) have not been tested at scale. We expect a v1.1 of this paper following the first complete annual cycle of measurement.