{
  "id": "social-findings-publication-2026-10-10-v1",
  "publicationVersion": "v1",
  "phase": 7,
  "publishedOn": "2026-10-10",
  "baseEvidenceRelease": "lens-2026-10-09-cbec43fe76c4",
  "status": "Exploratory research interpretation",
  "humanReviewRecorded": false,
  "originalHypothesesPromoted": [],
  "validatedPricingFactors": [],
  "headline": "Social context carries a temporal signal. Geographic transfer remains unresolved.",
  "measure": "Percentage reduction in held-out Poisson deviance versus each stated comparator. Positive means better model fit; negative means worse. It is not a percentage change in collision frequency, claims or premiums.",
  "outcome": "Police-reported personal-injury collisions at road location per 100,000 matched motor vehicle miles.",
  "timing": {
    "temporalYears": [
      2023,
      2024,
      2025
    ],
    "spatialYear": 2025,
    "diagnosticOutcomesAlreadyInspected": true,
    "explanation": "Original models were refitted before each annual assessment. The later diagnostic holds 2023 parameters fixed while updating lagged inputs and using realised mileage. Its design followed inspection of the original results."
  },
  "selection": "Four decision-relevant comparisons beyond collision history are summarised. They are not the full analysis: all 11 primary comparisons and the 29-force recording sensitivity remain available in v2. Authority and force studies have different rows and are not a league table.",
  "findings": [
    {
      "id": "combined",
      "study": "authority",
      "comparison": "combined-history",
      "title": "Deprivation and education together",
      "comparator": "Road-context baseline and collision history",
      "added": "Income and employment deprivation plus the 2011 no-qualifications share",
      "reading": "Temporal signal; regional transfer unresolved",
      "implication": "Retain the combined block as a research candidate. Resolve road-context confounding and test a frozen specification on fresh outcomes before operational use.",
      "areas": 133,
      "geography": "English local authorities",
      "model": "DEH",
      "comparatorModel": "BH",
      "originalStatus": "Not consistently supported",
      "temporal": {
        "gain": 0.10965763397448036,
        "interval95": [
          0.05992130992415709,
          0.150150421471265
        ],
        "observedExpected": 1.0295586540395616
      },
      "spatial": {
        "gain": -0.043483808158396586,
        "interval95": [
          -0.15582203372125308,
          0.10585245389542129
        ],
        "observedExpected": 1.140840931243812
      },
      "fixedParameters": {
        "gain": 0.12202093349347654,
        "interval95": [
          0.07116943441095444,
          0.1635675750585255
        ],
        "annual": [
          {
            "year": 2023,
            "gain": 0.05698824167230743
          },
          {
            "year": 2024,
            "gain": 0.12239570278605338
          },
          {
            "year": 2025,
            "gain": 0.163565033124613
          }
        ]
      },
      "links": {
        "original": "/data-lab/social-factors?version=v2&study=authority&comparison=combined-history&release=lens-2026-10-09-cbec43fe76c4",
        "diagnostics": "/data-lab/social-factors/robustness?version=v3&study=authority&comparison=combined-history&release=lens-2026-10-09-cbec43fe76c4"
      }
    },
    {
      "id": "deprivation",
      "study": "authority",
      "comparison": "deprivation-history",
      "title": "Deprivation",
      "comparator": "Road-context baseline and collision history",
      "added": "Income and employment deprivation",
      "reading": "Later-year improvement; little original transfer gain",
      "implication": "Test the deprivation block alongside road mix and history. The combined evidence does not identify separate causal effects of income and employment.",
      "areas": 133,
      "geography": "English local authorities",
      "model": "DH",
      "comparatorModel": "BH",
      "originalStatus": "Not consistently supported",
      "temporal": {
        "gain": 0.06820024739525876,
        "interval95": [
          0.030014490302054067,
          0.10029049596615111
        ],
        "observedExpected": 1.033700431238181
      },
      "spatial": {
        "gain": -0.00010373290975218552,
        "interval95": [
          -0.07253689447575834,
          0.08994692459532071
        ],
        "observedExpected": 1.145802222909968
      },
      "fixedParameters": {
        "gain": 0.07708571924997953,
        "interval95": [
          0.0367910708492962,
          0.11274787318609342
        ],
        "annual": [
          {
            "year": 2023,
            "gain": 0.019870771985061597
          },
          {
            "year": 2024,
            "gain": 0.08494957930713165
          },
          {
            "year": 2025,
            "gain": 0.10744854435905415
          }
        ]
      },
      "links": {
        "original": "/data-lab/social-factors?version=v2&study=authority&comparison=deprivation-history&release=lens-2026-10-09-cbec43fe76c4",
        "diagnostics": "/data-lab/social-factors/robustness?version=v3&study=authority&comparison=deprivation-history&release=lens-2026-10-09-cbec43fe76c4"
      }
    },
    {
      "id": "education",
      "study": "authority",
      "comparison": "education-history",
      "title": "Education",
      "comparator": "Road-context baseline, deprivation and collision history",
      "added": "2011 share of residents aged 16+ with no qualifications",
      "reading": "Incremental temporal signal; regional deterioration",
      "implication": "Keep qualifications as an area-context descriptor for investigation. It is neither a current annual education measure nor a driver-level effect.",
      "areas": 133,
      "geography": "English local authorities",
      "model": "DEH",
      "comparatorModel": "DH",
      "originalStatus": "Not consistently supported",
      "temporal": {
        "gain": 0.04449173383372562,
        "interval95": [
          0.019377377999863787,
          0.07309963667713959
        ],
        "observedExpected": 1.0295586540395616
      },
      "spatial": {
        "gain": -0.04337557577395712,
        "interval95": [
          -0.09667299547711901,
          0.040476828566436966
        ],
        "observedExpected": 1.140840931243812
      },
      "fixedParameters": {
        "gain": 0.04868839412364456,
        "interval95": [
          0.020599969797450702,
          0.08246551053423672
        ],
        "annual": [
          {
            "year": 2023,
            "gain": 0.037869975332150974
          },
          {
            "year": 2024,
            "gain": 0.04092247009795136
          },
          {
            "year": 2025,
            "gain": 0.06287199288164436
          }
        ]
      },
      "links": {
        "original": "/data-lab/social-factors?version=v2&study=authority&comparison=education-history&release=lens-2026-10-09-cbec43fe76c4",
        "diagnostics": "/data-lab/social-factors/robustness?version=v3&study=authority&comparison=education-history&release=lens-2026-10-09-cbec43fe76c4"
      }
    },
    {
      "id": "burglary",
      "study": "crime",
      "comparison": "burglary-history",
      "title": "Recorded burglary",
      "comparator": "Force-level road-context baseline, deprivation and collision history",
      "added": "Two-year-lagged burglary intensity relative to fixed 2011 population",
      "reading": "No consistent added value",
      "implication": "Do not promote burglary as a stable predictor from this study. This finding cannot settle the predictive value of all crime or smaller local crime measures.",
      "areas": 28,
      "geography": "Fully matched English police-force areas",
      "model": "DCH",
      "comparatorModel": "DH",
      "originalStatus": "Not consistently supported",
      "temporal": {
        "gain": -0.053042444638806474,
        "interval95": [
          -0.7338408826364957,
          0.25619329033175336
        ],
        "observedExpected": 0.9741785309203165
      },
      "spatial": {
        "gain": -0.033722074340855945,
        "interval95": [
          -0.06276404426363191,
          0.23207341391488268
        ],
        "observedExpected": 0.6827523099403602
      },
      "fixedParameters": {
        "gain": -0.04110512449100012,
        "interval95": [
          -0.7232149104458917,
          0.25406329206375694
        ],
        "annual": [
          {
            "year": 2023,
            "gain": 0.003914703494535932
          },
          {
            "year": 2024,
            "gain": 0.024761553468688136
          },
          {
            "year": 2025,
            "gain": -0.1262022132257674
          }
        ]
      },
      "links": {
        "original": "/data-lab/social-factors?version=v2&study=crime&comparison=burglary-history&release=lens-2026-10-09-cbec43fe76c4",
        "diagnostics": "/data-lab/social-factors/robustness?version=v3&study=crime&comparison=burglary-history&release=lens-2026-10-09-cbec43fe76c4"
      }
    }
  ],
  "transfer": {
    "london": {
      "areas": 33,
      "outsideAreas": 24,
      "outsideMileageShare": 0.6390016997974493
    },
    "combinedSpatialSensitivities": [
      {
        "id": "poisson",
        "gain": 0.06510173121559792
      },
      {
        "id": "fixed-imd2015",
        "gain": 0.03899872429642948
      },
      {
        "id": "no-region",
        "gain": 0.07109587347381052
      }
    ],
    "explanation": "Alternative assumptions change the geographic result. They were examined after seeing the original outcomes; a better diagnostic score does not replace the original failed transfer test. London remains included."
  },
  "sources": [
    {
      "version": "v2",
      "id": "social-exploratory-2026-10-10-v2",
      "url": "/research/social-factors/exploratory-v2/results.json",
      "sha256": "ccdb03786ced5860dddb0b677b3207cd56544988b1d258d1d5b695d565b9d126"
    },
    {
      "version": "v3",
      "id": "social-robustness-2026-10-10-v3",
      "url": "/research/social-factors/robustness-v3/results.json",
      "sha256": "442fdcc27fc37d85da13cf9fe8a8751fb8edae0b7d5496464395d818902834be"
    }
  ],
  "uncertainty": "95% intervals use nine region blocks and fixed predictions. They omit refitting and input measurement uncertainty, may be optimistic under regional dependence, and are unadjusted for multiple exploratory comparisons.",
  "limitations": [
    "The reduced baseline omits historical joint road-class and urban/rural controls; they could explain some social signal.",
    "2011 Census context is held fixed. Historical input series were retrieved in 2026 and exact forecast-origin vintages have not been reconstructed.",
    "Area residents are not necessarily the drivers on those roads. The study excludes damage-only events and does not measure insured claims.",
    "The fixed complete-history cohort is a restricted sample and uses knowledge of later data availability. It does not establish a whole-England result.",
    "All 2023–2025 outcomes have been inspected. Reusing them cannot provide fresh confirmation."
  ],
  "nextEvidence": [
    {
      "need": "Comparable road context",
      "requirement": "Add historically matched joint road-class and urban/rural mileage controls, or design a study on a network where matched exposure is observable."
    },
    {
      "need": "Fresh validation",
      "requirement": "Freeze the next specification and acceptance criteria before inspecting new outcomes or independent data; assess temporal performance, geographic transfer and calibration together."
    },
    {
      "need": "Insurer applicability",
      "requirement": "Test relevant features against geocoded insurer mileage/exposure and developed claims, with portfolio calibration and review of intended use."
    }
  ],
  "websiteUse": "Publish the research interpretation and negative findings with exact comparison links. No risk score, postcode ranking, individual-driver relativity or automatic pricing factor is created.",
  "maintenance": "Phase 8 remains pending. Recurring source monitoring and model refresh are not enabled by this publication."
}