{
 "note": "Thesia on the Artificial Societies Benchmark, sealed benchmark-split run. Lower is better (total variation distance). Summary first, then the score reports for Thesia and for the demographic model alone.",
 "correction": "2026-10-26: GSS rescored with the benchmark convention (human distribution over every asked respondent with a recorded answer); GSS trust 0.033 -> 0.034, seven-source mean 0.064 -> 0.065. Answers unchanged.",
 "summary": {
  "generated": "2026-10-26",
  "preregistration": {
   "sha256": "29d77fb924451214eb3dd3a85b1461ab354510d1487fe0c25fb9887227725de2",
   "frozen": "2026-10-26T17:39:16+02:00",
   "claim_rule": "A seven-source mean at or below 0.08 is on par with the self-reported Artificial Societies figure; strictly below is better."
  },
  "benchmark": {
   "name": "Artificial Societies Benchmark",
   "arxiv": "2609.30030",
   "companion_arxiv": "2609.27690",
   "repo": "https://github.com/artificial-societies/artificial-societies-benchmark",
   "commit": "46b8ed5",
   "self_reported_url": "https://societies.ai/as-benchmark"
  },
  "sources": [
   {
    "id": "gss_2024_confidence",
    "label": "GSS 2024, confidence in institutions",
    "people": 664,
    "answers": 7803,
    "valid": 7800,
    "thesia": 0.041,
    "artificial_societies": 0.042,
    "best_llm": {
     "tvd": 0.207,
     "model": "Qwen3.8 27B"
    },
    "per_person_error": 0.562
   },
   {
    "id": "gss_2024_interpersonal",
    "label": "GSS 2024, interpersonal trust",
    "people": 276,
    "answers": 816,
    "valid": 815,
    "thesia": 0.034,
    "artificial_societies": 0.058,
    "best_llm": {
     "tvd": 0.155,
     "model": "Gemini 3.8 Flash"
    },
    "per_person_error": 0.505
   },
   {
    "id": "sapa_traits",
    "label": "SAPA personality traits",
    "people": 512,
    "answers": 10240,
    "valid": 10119,
    "thesia": 0.057,
    "artificial_societies": 0.09,
    "best_llm": {
     "tvd": 0.376,
     "model": "Qwen3.8 27B"
    },
    "per_person_error": 0.759
   },
   {
    "id": "sapa_facets",
    "label": "SAPA personality facets",
    "people": 512,
    "answers": 4608,
    "valid": 4568,
    "thesia": 0.07,
    "artificial_societies": 0.117,
    "best_llm": {
     "tvd": 0.336,
     "model": "Qwen3.8 27B"
    },
    "per_person_error": 0.754
   },
   {
    "id": "twin_retest",
    "label": "Twin-2K-500 retest",
    "people": 512,
    "answers": 12288,
    "valid": 12225,
    "thesia": 0.09,
    "artificial_societies": 0.147,
    "best_llm": {
     "tvd": 0.268,
     "model": "Qwen3.8 27B"
    },
    "per_person_error": 0.628
   },
   {
    "id": "jordan_wording",
    "label": "Jordan survey experiment",
    "people": 512,
    "answers": 512,
    "valid": 510,
    "thesia": 0.061,
    "artificial_societies": 0.096,
    "best_llm": {
     "tvd": 0.145,
     "model": "Grok 4.3"
    },
    "per_person_error": 0.525
   },
   {
    "id": "albertson_wording",
    "label": "Albertson wording experiment",
    "people": 512,
    "answers": 512,
    "valid": 510,
    "thesia": 0.098,
    "artificial_societies": 0.025,
    "best_llm": {
     "tvd": 0.199,
     "model": "Grok 4.3"
    },
    "per_person_error": 0.569
   }
  ],
  "means": {
   "thesia": 0.065,
   "artificial_societies": 0.08,
   "artificial_societies_per_source": 0.082,
   "best_llm_per_source": 0.241,
   "prior_only": 0.059
  },
  "wins": 6,
  "secondary": {
   "subgroup_tvd": {
    "confidence": 0.061,
    "trust": 0.058,
    "artificial_societies": 0.06,
    "best_llm": 0.22
   },
   "spread_sd": {
    "thesia": 0.01,
    "artificial_societies": 0.006,
    "best_llm": 0.091
   },
   "wording_albertson": {
    "human_pp": 10.6,
    "model_pp": 26.3,
    "error_pp": 15.7,
    "signed_ci95_pp": [
     6.8,
     24.2
    ],
    "artificial_societies_pp": 1.4,
    "best_llm_pp": 6.2
   },
   "wording_jordan": {
    "human_pp": 15.0,
    "model_pp": 10.2,
    "error_pp": 4.7,
    "signed_ci95_pp": [
     -12.3,
     3.1
    ]
   }
  },
  "transport_retry": "About 4–5% of calls returned no answer under 12 parallel workers; only those were re-sent with identical prompts. Unparseable answers stayed failures."
 },
 "score_reports": {
  "gss": {
   "reports": [
    {
     "run": "thesia",
     "tasks": {
      "gss_2024_confidence": {
       "people": 664,
       "answers": 7803,
       "coverage": 0.9996155324875048,
       "statuses": {
        "valid": 7965
       },
       "ev1_tvd": 0.041496873298776486,
       "abs_sd_difference": 0.01020388323757349,
       "signed_sd_difference": -0.0012897105426748722,
       "per_person_error": 0.5622165067195187,
       "subgroup_tvd": 0.06138696342327872,
       "items": [
        {
         "item_id": "conbus",
         "n": 650,
         "valid_fraction": 1.0,
         "tvd": 0.02769230769230771,
         "signed_sd_difference": -0.007515919148695593,
         "human": {
          "1": 0.148,
          "2": 0.614,
          "3": 0.238
         },
         "synthetic": {
          "1": 0.152,
          "2": 0.637,
          "3": 0.211
         }
        },
        {
         "item_id": "conclerg",
         "n": 642,
         "valid_fraction": 1.0,
         "tvd": 0.01401869158878502,
         "signed_sd_difference": 0.006570842254042308,
         "human": {
          "1": 0.153,
          "2": 0.511,
          "3": 0.336
         },
         "synthetic": {
          "1": 0.165,
          "2": 0.497,
          "3": 0.338
         }
        },
        {
         "item_id": "coneduc",
         "n": 660,
         "valid_fraction": 1.0,
         "tvd": 0.056060606060606116,
         "signed_sd_difference": -0.01314230175637393,
         "human": {
          "1": 0.227,
          "2": 0.564,
          "3": 0.209
         },
         "synthetic": {
          "1": 0.171,
          "2": 0.594,
          "3": 0.235
         }
        },
        {
         "item_id": "conpress",
         "n": 647,
         "valid_fraction": 1.0,
         "tvd": 0.013910355486862466,
         "signed_sd_difference": -0.009170548368036935,
         "human": {
          "1": 0.093,
          "2": 0.403,
          "3": 0.504
         },
         "synthetic": {
          "1": 0.079,
          "2": 0.403,
          "3": 0.518
         }
        },
        {
         "item_id": "conmedic",
         "n": 651,
         "valid_fraction": 0.9984639016897081,
         "tvd": 0.062022923313245895,
         "signed_sd_difference": -0.00034204821112782113,
         "human": {
          "1": 0.315,
          "2": 0.512,
          "3": 0.174
         },
         "synthetic": {
          "1": 0.377,
          "2": 0.478,
          "3": 0.145
         }
        },
        {
         "item_id": "consci",
         "n": 643,
         "valid_fraction": 1.0,
         "tvd": 0.038880248833592534,
         "signed_sd_difference": 0.02346704242646752,
         "human": {
          "1": 0.412,
          "2": 0.493,
          "3": 0.095
         },
         "synthetic": {
          "1": 0.412,
          "2": 0.454,
          "3": 0.134
         }
        },
        {
         "item_id": "conlegis",
         "n": 649,
         "valid_fraction": 0.9984591679506933,
         "tvd": 0.07179373775418976,
         "signed_sd_difference": -0.006739806347935473,
         "human": {
          "1": 0.057,
          "2": 0.421,
          "3": 0.522
         },
         "synthetic": {
          "1": 0.049,
          "2": 0.356,
          "3": 0.594
         }
        },
        {
         "item_id": "conarmy",
         "n": 654,
         "valid_fraction": 1.0,
         "tvd": 0.05045871559633029,
         "signed_sd_difference": 0.012618523985681473,
         "human": {
          "1": 0.417,
          "2": 0.477,
          "3": 0.106
         },
         "synthetic": {
          "1": 0.456,
          "2": 0.427,
          "3": 0.118
         }
        },
        {
         "item_id": "contv",
         "n": 651,
         "valid_fraction": 1.0,
         "tvd": 0.0261136712749616,
         "signed_sd_difference": 0.00324170205010188,
         "human": {
          "1": 0.075,
          "2": 0.47,
          "3": 0.455
         },
         "synthetic": {
          "1": 0.077,
          "2": 0.444,
          "3": 0.479
         }
        },
        {
         "item_id": "conjudge",
         "n": 649,
         "valid_fraction": 1.0,
         "tvd": 0.024653312788906007,
         "signed_sd_difference": 0.007422573276111399,
         "human": {
          "1": 0.163,
          "2": 0.431,
          "3": 0.405
         },
         "synthetic": {
          "1": 0.188,
          "2": 0.431,
          "3": 0.381
         }
        },
        {
         "item_id": "confinan",
         "n": 657,
         "valid_fraction": 0.9984779299847792,
         "tvd": 0.03235790919553032,
         "signed_sd_difference": 0.0001643521769871259,
         "human": {
          "1": 0.186,
          "2": 0.588,
          "3": 0.227
         },
         "synthetic": {
          "1": 0.162,
          "2": 0.579,
          "3": 0.259
         }
        },
        {
         "item_id": "confed",
         "n": 650,
         "valid_fraction": 1.0,
         "tvd": 0.07999999999999999,
         "signed_sd_difference": -0.03205093884932042,
         "human": {
          "1": 0.148,
          "2": 0.437,
          "3": 0.415
         },
         "synthetic": {
          "1": 0.082,
          "2": 0.423,
          "3": 0.495
         }
        }
       ],
       "subgroups": [
        {
         "field": "age_band",
         "value": "30–44",
         "n": 189,
         "tvd": 0.060106187773836854
        },
        {
         "field": "age_band",
         "value": "45–59",
         "n": 148,
         "tvd": 0.05921427527642522
        },
        {
         "field": "age_band",
         "value": "60–74",
         "n": 186,
         "tvd": 0.05331405584283235
        },
        {
         "field": "sex",
         "value": "Female",
         "n": 335,
         "tvd": 0.058468482649928745
        },
        {
         "field": "sex",
         "value": "Male",
         "n": 329,
         "tvd": 0.046912573464799275
        },
        {
         "field": "ethnicity",
         "value": "Black",
         "n": 100,
         "tvd": 0.09125919202461234
        },
        {
         "field": "ethnicity",
         "value": "White",
         "n": 437,
         "tvd": 0.04350614710773584
        },
        {
         "field": "education",
         "value": "Bachelor’s degree",
         "n": 169,
         "tvd": 0.06499755600154201
        },
        {
         "field": "education",
         "value": "Graduate degree",
         "n": 120,
         "tvd": 0.0716524038994129
        },
        {
         "field": "education",
         "value": "High School",
         "n": 175,
         "tvd": 0.06592790061825801
        },
        {
         "field": "education",
         "value": "Some College or Associate’s degree",
         "n": 200,
         "tvd": 0.06115941037966885
        },
        {
         "field": "urbanicity",
         "value": "Suburban",
         "n": 381,
         "tvd": 0.04526228386126143
        },
        {
         "field": "urbanicity",
         "value": "Urban",
         "n": 214,
         "tvd": 0.0757224106132
        },
        {
         "field": "politics",
         "value": "Democrat",
         "n": 188,
         "tvd": 0.05466882306664503
        },
        {
         "field": "politics",
         "value": "Independent",
         "n": 307,
         "tvd": 0.051957051705922214
        },
        {
         "field": "politics",
         "value": "Republican",
         "n": 169,
         "tvd": 0.07806266048637849
        }
       ]
      },
      "gss_2024_interpersonal": {
       "people": 276,
       "answers": 816,
       "coverage": 0.9987745098039216,
       "statuses": {
        "valid": 827
       },
       "ev1_tvd": 0.03353714538171095,
       "abs_sd_difference": null,
       "signed_sd_difference": null,
       "per_person_error": 0.5054347826086957,
       "subgroup_tvd": 0.05836951337261549,
       "items": [
        {
         "item_id": "trust",
         "n": 274,
         "valid_fraction": 1.0,
         "tvd": 0.043795620437956165,
         "signed_sd_difference": null,
         "human": {
          "1": 0.277,
          "2": 0.588,
          "3": 0.135
         },
         "synthetic": {
          "1": 0.255,
          "2": 0.631,
          "3": 0.113
         }
        },
        {
         "item_id": "fair",
         "n": 270,
         "valid_fraction": 1.0,
         "tvd": 0.003703703703703709,
         "signed_sd_difference": null,
         "human": {
          "1": 0.437,
          "2": 0.448,
          "3": 0.115
         },
         "synthetic": {
          "1": 0.433,
          "2": 0.448,
          "3": 0.119
         }
        },
        {
         "item_id": "helpful",
         "n": 272,
         "valid_fraction": 0.9963235294117647,
         "tvd": 0.05311211200347298,
         "signed_sd_difference": null,
         "human": {
          "1": 0.393,
          "2": 0.441,
          "3": 0.165
         },
         "synthetic": {
          "1": 0.446,
          "2": 0.435,
          "3": 0.118
         }
        }
       ],
       "subgroups": [
        {
         "field": "sex",
         "value": "Female",
         "n": 138,
         "tvd": 0.03903838471874203
        },
        {
         "field": "sex",
         "value": "Male",
         "n": 138,
         "tvd": 0.07443924328233993
        },
        {
         "field": "ethnicity",
         "value": "White",
         "n": 150,
         "tvd": 0.03816922558303453
        },
        {
         "field": "urbanicity",
         "value": "Suburban",
         "n": 149,
         "tvd": 0.07107617821903535
        },
        {
         "field": "politics",
         "value": "Independent",
         "n": 123,
         "tvd": 0.06912453505992561
        }
       ]
      }
     }
    },
    {
     "run": "demographic_model_alone",
     "tasks": {
      "gss_2024_confidence": {
       "people": 664,
       "answers": 7803,
       "coverage": 0.9996155324875048,
       "statuses": {
        "valid": 7965
       },
       "ev1_tvd": 0.03947055261009409,
       "abs_sd_difference": 0.008569519578897072,
       "signed_sd_difference": -0.0010213598481094942,
       "per_person_error": 0.5578536005493836,
       "subgroup_tvd": 0.060460219134077536,
       "items": [
        {
         "item_id": "conbus",
         "n": 650,
         "valid_fraction": 1.0,
         "tvd": 0.040000000000000036,
         "signed_sd_difference": -0.004538163348681956,
         "human": {
          "1": 0.148,
          "2": 0.614,
          "3": 0.238
         },
         "synthetic": {
          "1": 0.169,
          "2": 0.632,
          "3": 0.198
         }
        },
        {
         "item_id": "conclerg",
         "n": 642,
         "valid_fraction": 1.0,
         "tvd": 0.01090342679127726,
         "signed_sd_difference": 0.004312360622472422,
         "human": {
          "1": 0.153,
          "2": 0.511,
          "3": 0.336
         },
         "synthetic": {
          "1": 0.164,
          "2": 0.505,
          "3": 0.332
         }
        },
        {
         "item_id": "coneduc",
         "n": 660,
         "valid_fraction": 1.0,
         "tvd": 0.056060606060606075,
         "signed_sd_difference": -0.015249312575449203,
         "human": {
          "1": 0.227,
          "2": 0.564,
          "3": 0.209
         },
         "synthetic": {
          "1": 0.171,
          "2": 0.6,
          "3": 0.229
         }
        },
        {
         "item_id": "conpress",
         "n": 647,
         "valid_fraction": 1.0,
         "tvd": 0.02936630602782072,
         "signed_sd_difference": -0.007367710096409685,
         "human": {
          "1": 0.093,
          "2": 0.403,
          "3": 0.504
         },
         "synthetic": {
          "1": 0.08,
          "2": 0.386,
          "3": 0.533
         }
        },
        {
         "item_id": "conmedic",
         "n": 651,
         "valid_fraction": 0.9984639016897081,
         "tvd": 0.0697152310055536,
         "signed_sd_difference": -0.0039024432550843757,
         "human": {
          "1": 0.315,
          "2": 0.512,
          "3": 0.174
         },
         "synthetic": {
          "1": 0.385,
          "2": 0.48,
          "3": 0.135
         }
        },
        {
         "item_id": "consci",
         "n": 643,
         "valid_fraction": 1.0,
         "tvd": 0.049766718506998465,
         "signed_sd_difference": 0.021106964555316854,
         "human": {
          "1": 0.412,
          "2": 0.493,
          "3": 0.095
         },
         "synthetic": {
          "1": 0.432,
          "2": 0.443,
          "3": 0.124
         }
        },
        {
         "item_id": "conlegis",
         "n": 649,
         "valid_fraction": 0.9984591679506933,
         "tvd": 0.04710237972949839,
         "signed_sd_difference": -0.0007902194489978753,
         "human": {
          "1": 0.057,
          "2": 0.421,
          "3": 0.522
         },
         "synthetic": {
          "1": 0.056,
          "2": 0.375,
          "3": 0.569
         }
        },
        {
         "item_id": "conarmy",
         "n": 654,
         "valid_fraction": 1.0,
         "tvd": 0.04740061162079512,
         "signed_sd_difference": 0.004926147226529898,
         "human": {
          "1": 0.417,
          "2": 0.477,
          "3": 0.106
         },
         "synthetic": {
          "1": 0.465,
          "2": 0.431,
          "3": 0.104
         }
        },
        {
         "item_id": "contv",
         "n": 651,
         "valid_fraction": 1.0,
         "tvd": 0.02150537634408605,
         "signed_sd_difference": 0.005751871669910591,
         "human": {
          "1": 0.075,
          "2": 0.47,
          "3": 0.455
         },
         "synthetic": {
          "1": 0.081,
          "2": 0.449,
          "3": 0.47
         }
        },
        {
         "item_id": "conjudge",
         "n": 649,
         "valid_fraction": 1.0,
         "tvd": 0.02619414483821264,
         "signed_sd_difference": 0.008477828350317351,
         "human": {
          "1": 0.163,
          "2": 0.431,
          "3": 0.405
         },
         "synthetic": {
          "1": 0.19,
          "2": 0.428,
          "3": 0.382
         }
        },
        {
         "item_id": "confinan",
         "n": 657,
         "valid_fraction": 0.9984779299847792,
         "tvd": 0.011016445780896139,
         "signed_sd_difference": 0.0007137859601783481,
         "human": {
          "1": 0.186,
          "2": 0.588,
          "3": 0.227
         },
         "synthetic": {
          "1": 0.178,
          "2": 0.584,
          "3": 0.238
         }
        },
        {
         "item_id": "confed",
         "n": 650,
         "valid_fraction": 1.0,
         "tvd": 0.06461538461538459,
         "signed_sd_difference": -0.025697427837416298,
         "human": {
          "1": 0.148,
          "2": 0.437,
          "3": 0.415
         },
         "synthetic": {
          "1": 0.092,
          "2": 0.428,
          "3": 0.48
         }
        }
       ],
       "subgroups": [
        {
         "field": "age_band",
         "value": "30–44",
         "n": 189,
         "tvd": 0.06320665386147103
        },
        {
         "field": "age_band",
         "value": "45–59",
         "n": 148,
         "tvd": 0.050549350193809996
        },
        {
         "field": "age_band",
         "value": "60–74",
         "n": 186,
         "tvd": 0.05278802242059882
        },
        {
         "field": "sex",
         "value": "Female",
         "n": 335,
         "tvd": 0.05490562776716137
        },
        {
         "field": "sex",
         "value": "Male",
         "n": 329,
         "tvd": 0.0524983684249753
        },
        {
         "field": "ethnicity",
         "value": "Black",
         "n": 100,
         "tvd": 0.08950462342650933
        },
        {
         "field": "ethnicity",
         "value": "White",
         "n": 437,
         "tvd": 0.04604200382985859
        },
        {
         "field": "education",
         "value": "Bachelor’s degree",
         "n": 169,
         "tvd": 0.05999547480952142
        },
        {
         "field": "education",
         "value": "Graduate degree",
         "n": 120,
         "tvd": 0.07250451027868775
        },
        {
         "field": "education",
         "value": "High School",
         "n": 175,
         "tvd": 0.0620477903984579
        },
        {
         "field": "education",
         "value": "Some College or Associate’s degree",
         "n": 200,
         "tvd": 0.06113996256058105
        },
        {
         "field": "urbanicity",
         "value": "Suburban",
         "n": 381,
         "tvd": 0.04526093477252277
        },
        {
         "field": "urbanicity",
         "value": "Urban",
         "n": 214,
         "tvd": 0.07211632862882947
        },
        {
         "field": "politics",
         "value": "Democrat",
         "n": 188,
         "tvd": 0.05516063098770795
        },
        {
         "field": "politics",
         "value": "Independent",
         "n": 307,
         "tvd": 0.05308599422787117
        },
        {
         "field": "politics",
         "value": "Republican",
         "n": 169,
         "tvd": 0.07655722955667658
        }
       ]
      },
      "gss_2024_interpersonal": {
       "people": 276,
       "answers": 816,
       "coverage": 0.9987745098039216,
       "statuses": {
        "valid": 827
       },
       "ev1_tvd": 0.0389731120070134,
       "abs_sd_difference": null,
       "signed_sd_difference": null,
       "per_person_error": 0.5066425120772946,
       "subgroup_tvd": 0.06222566030587181,
       "items": [
        {
         "item_id": "trust",
         "n": 274,
         "valid_fraction": 1.0,
         "tvd": 0.054744525547445244,
         "signed_sd_difference": null,
         "human": {
          "1": 0.277,
          "2": 0.588,
          "3": 0.135
         },
         "synthetic": {
          "1": 0.248,
          "2": 0.642,
          "3": 0.109
         }
        },
        {
         "item_id": "fair",
         "n": 270,
         "valid_fraction": 1.0,
         "tvd": 0.01481481481481483,
         "signed_sd_difference": null,
         "human": {
          "1": 0.437,
          "2": 0.448,
          "3": 0.115
         },
         "synthetic": {
          "1": 0.43,
          "2": 0.463,
          "3": 0.107
         }
        },
        {
         "item_id": "helpful",
         "n": 272,
         "valid_fraction": 0.9963235294117647,
         "tvd": 0.04735999565878012,
         "signed_sd_difference": null,
         "human": {
          "1": 0.393,
          "2": 0.441,
          "3": 0.165
         },
         "synthetic": {
          "1": 0.439,
          "2": 0.443,
          "3": 0.118
         }
        }
       ],
       "subgroups": [
        {
         "field": "sex",
         "value": "Female",
         "n": 138,
         "tvd": 0.0415804190725206
        },
        {
         "field": "sex",
         "value": "Male",
         "n": 138,
         "tvd": 0.07442135291451397
        },
        {
         "field": "ethnicity",
         "value": "White",
         "n": 150,
         "tvd": 0.04040636204835892
        },
        {
         "field": "urbanicity",
         "value": "Suburban",
         "n": 149,
         "tvd": 0.08013115155972299
        },
        {
         "field": "politics",
         "value": "Independent",
         "n": 123,
         "tvd": 0.07458901593424254
        }
       ]
      }
     }
    }
   ]
  },
  "other_sources": {
   "thesia": {
    "albertson_wording": {
     "answers": 512,
     "valid": 510,
     "ev1_tvd": 0.09803921568627452,
     "ev1_numeric_w1": null,
     "human_retest_loss": null,
     "model_loss_vs_wave4": null,
     "ev1_status": "estimated",
     "trait_corr_error": null,
     "person_loss": 0.5686274509803921,
     "human_effect_pp": 10.588235294117643,
     "model_effect_pp": 26.274509803921575,
     "effect_error_pp": 15.686274509803933,
     "subgroups": [
      [
       "Education",
       "2-year college degree",
       66,
       NaN
      ],
      [
       "Education",
       "4-year college degree",
       191,
       0.092
      ],
      [
       "Education",
       "High school graduate",
       62,
       NaN
      ],
      [
       "Education",
       "Postgraduate degree (MA, MBA, MD, JD, PhD, etc.)",
       80,
       NaN
      ],
      [
       "Education",
       "Some college but no degree (yet)",
       111,
       0.163
      ],
      [
       "Gender",
       "Female",
       286,
       0.106
      ],
      [
       "Gender",
       "Male",
       220,
       0.15
      ],
      [
       "Race or origin",
       "Black or African-American",
       57,
       NaN
      ],
      [
       "Race or origin",
       "White",
       363,
       0.15
      ]
     ]
    },
    "jordan_wording": {
     "answers": 512,
     "valid": 510,
     "ev1_tvd": 0.06094672736220472,
     "ev1_numeric_w1": null,
     "human_retest_loss": null,
     "model_loss_vs_wave4": null,
     "ev1_status": "estimated",
     "trait_corr_error": null,
     "person_loss": 0.5254901960784314,
     "human_effect_pp": 14.960629921259837,
     "model_effect_pp": 10.225455216535428,
     "effect_error_pp": 4.735174704724409,
     "subgroups": [
      [
       "Education",
       "2-year college degree (Associate’s degree)",
       62,
       NaN
      ],
      [
       "Education",
       "4-year college degree (Bachelor’s degree)",
       189,
       0.095
      ],
      [
       "Education",
       "High school degree or equivalent (for example: GED)",
       60,
       NaN
      ],
      [
       "Education",
       "Postgraduate degree (MA, MBA, MD, JD, PhD, etc.)",
       94,
       NaN
      ],
      [
       "Education",
       "Some college, but no degree",
       103,
       0.107
      ],
      [
       "Gender",
       "Female",
       276,
       0.075
      ],
      [
       "Gender",
       "Male",
       230,
       0.055
      ],
      [
       "Political affiliation",
       "Democrat",
       175,
       0.044
      ],
      [
       "Political affiliation",
       "Independent",
       189,
       0.107
      ],
      [
       "Political affiliation",
       "Republican",
       144,
       0.179
      ],
      [
       "Race or origin",
       "Black or African-American",
       70,
       NaN
      ],
      [
       "Race or origin",
       "White",
       359,
       0.085
      ]
     ]
    },
    "sapa_facets": {
     "answers": 4608,
     "valid": 4568,
     "ev1_tvd": 0.07005795478534749,
     "ev1_numeric_w1": null,
     "human_retest_loss": null,
     "model_loss_vs_wave4": null,
     "ev1_status": "estimated",
     "trait_corr_error": null,
     "person_loss": 0.7539876302083334,
     "human_effect_pp": NaN,
     "model_effect_pp": NaN,
     "effect_error_pp": NaN,
     "subgroups": [
      [
       "Age band",
       "19–24 years",
       234,
       0.121
      ],
      [
       "Age band",
       "25–29 years",
       88,
       0.143
      ],
      [
       "Age band",
       "30–34 years",
       61,
       0.171
      ],
      [
       "Age band",
       "40–49 years",
       53,
       0.173
      ],
      [
       "English proficiency",
       "Very well / fluent / native",
       304,
       0.089
      ],
      [
       "English proficiency",
       "Well",
       208,
       0.122
      ],
      [
       "Sex",
       "female",
       335,
       0.075
      ],
      [
       "Sex",
       "male",
       171,
       0.116
      ]
     ]
    },
    "sapa_traits": {
     "answers": 10240,
     "valid": 10119,
     "ev1_tvd": 0.05718047984767229,
     "ev1_numeric_w1": null,
     "human_retest_loss": null,
     "model_loss_vs_wave4": null,
     "ev1_status": "estimated",
     "trait_corr_error": null,
     "person_loss": 0.7587437113003096,
     "human_effect_pp": NaN,
     "model_effect_pp": NaN,
     "effect_error_pp": NaN,
     "subgroups": [
      [
       "Age band",
       "19–24 years",
       239,
       0.097
      ],
      [
       "Age band",
       "25–29 years",
       97,
       0.175
      ],
      [
       "Age band",
       "30–34 years",
       56,
       0.171
      ],
      [
       "English proficiency",
       "Very well / fluent / native",
       296,
       0.085
      ],
      [
       "English proficiency",
       "Well",
       216,
       0.098
      ],
      [
       "Sex",
       "female",
       329,
       0.082
      ],
      [
       "Sex",
       "male",
       176,
       0.098
      ]
     ]
    },
    "twin_retest": {
     "answers": 12288,
     "valid": 12225,
     "ev1_tvd": 0.090316993664313,
     "ev1_numeric_w1": 0.19800541996908338,
     "human_retest_loss": 0.44789162320972437,
     "model_loss_vs_wave4": 0.6312765783416696,
     "ev1_status": "estimated",
     "trait_corr_error": null,
     "person_loss": 0.6280218047492588,
     "human_effect_pp": NaN,
     "model_effect_pp": NaN,
     "effect_error_pp": NaN,
     "subgroups": [
      [
       "Age band",
       "18-29",
       103,
       0.167
      ],
      [
       "Age band",
       "30-49",
       193,
       0.133
      ],
      [
       "Age band",
       "50-64",
       157,
       0.153
      ],
      [
       "Age band",
       "65+",
       59,
       0.179
      ],
      [
       "Education",
       "Associate's degree",
       62,
       0.122
      ],
      [
       "Education",
       "College graduate/some postgrad",
       182,
       0.144
      ],
      [
       "Education",
       "High school graduate",
       66,
       0.161
      ],
      [
       "Education",
       "Postgraduate",
       76,
       0.156
      ],
      [
       "Education",
       "Some college, no degree",
       122,
       0.167
      ],
      [
       "Political affiliation",
       "Democrat",
       211,
       0.135
      ],
      [
       "Political affiliation",
       "Independent",
       145,
       0.149
      ],
      [
       "Political affiliation",
       "Republican",
       135,
       0.134
      ],
      [
       "Race or origin",
       "Black",
       77,
       0.159
      ],
      [
       "Race or origin",
       "Hispanic",
       55,
       0.128
      ],
      [
       "Race or origin",
       "White",
       318,
       0.112
      ],
      [
       "Sex assigned at birth",
       "Female",
       253,
       0.131
      ],
      [
       "Sex assigned at birth",
       "Male",
       259,
       0.129
      ],
      [
       "US region",
       "Midwest (ND, SD, NE, KS, MN, IA, MO, WI, IL, MI, IN, OH)",
       104,
       0.144
      ],
      [
       "US region",
       "Northeast (PA, NY, NJ, RI, CT, MA, VT, NH, ME)",
       78,
       0.145
      ],
      [
       "US region",
       "South (TX, OK, AR, LA, KY, TN, MS, AL, WV, DC, MD, DE, VA, NC, SC, GA, FL)",
       205,
       0.143
      ],
      [
       "US region",
       "West (WA, OR, ID, MT, WY, CA, NV, UT, CO, AZ, NM)",
       119,
       0.148
      ]
     ]
    }
   },
   "demographic_model_alone": {
    "albertson_wording": {
     "answers": 512,
     "valid": 510,
     "ev1_tvd": 0.08235294117647057,
     "ev1_numeric_w1": null,
     "human_retest_loss": null,
     "model_loss_vs_wave4": null,
     "ev1_status": "estimated",
     "trait_corr_error": null,
     "person_loss": 0.5745098039215686,
     "human_effect_pp": 10.588235294117643,
     "model_effect_pp": 23.13725490196078,
     "effect_error_pp": 12.549019607843137,
     "subgroups": [
      [
       "Education",
       "2-year college degree",
       66,
       NaN
      ],
      [
       "Education",
       "4-year college degree",
       191,
       0.047
      ],
      [
       "Education",
       "High school graduate",
       62,
       NaN
      ],
      [
       "Education",
       "Postgraduate degree (MA, MBA, MD, JD, PhD, etc.)",
       80,
       NaN
      ],
      [
       "Education",
       "Some college but no degree (yet)",
       111,
       0.164
      ],
      [
       "Gender",
       "Female",
       286,
       0.098
      ],
      [
       "Gender",
       "Male",
       220,
       0.092
      ],
      [
       "Race or origin",
       "Black or African-American",
       57,
       NaN
      ],
      [
       "Race or origin",
       "White",
       363,
       0.105
      ]
     ]
    },
    "jordan_wording": {
     "answers": 512,
     "valid": 510,
     "ev1_tvd": 0.04916646161417322,
     "ev1_numeric_w1": null,
     "human_retest_loss": null,
     "model_loss_vs_wave4": null,
     "ev1_status": "estimated",
     "trait_corr_error": null,
     "person_loss": 0.5333333333333333,
     "human_effect_pp": 14.960629921259837,
     "model_effect_pp": 9.261195866141737,
     "effect_error_pp": 5.699434055118102,
     "subgroups": [
      [
       "Education",
       "2-year college degree (Associate’s degree)",
       62,
       NaN
      ],
      [
       "Education",
       "4-year college degree (Bachelor’s degree)",
       189,
       0.085
      ],
      [
       "Education",
       "High school degree or equivalent (for example: GED)",
       60,
       NaN
      ],
      [
       "Education",
       "Postgraduate degree (MA, MBA, MD, JD, PhD, etc.)",
       94,
       NaN
      ],
      [
       "Education",
       "Some college, but no degree",
       103,
       0.143
      ],
      [
       "Gender",
       "Female",
       276,
       0.057
      ],
      [
       "Gender",
       "Male",
       230,
       0.059
      ],
      [
       "Political affiliation",
       "Democrat",
       175,
       0.04
      ],
      [
       "Political affiliation",
       "Independent",
       189,
       0.107
      ],
      [
       "Political affiliation",
       "Republican",
       144,
       0.158
      ],
      [
       "Race or origin",
       "Black or African-American",
       70,
       NaN
      ],
      [
       "Race or origin",
       "White",
       359,
       0.076
      ]
     ]
    },
    "sapa_facets": {
     "answers": 4608,
     "valid": 4568,
     "ev1_tvd": 0.06634818436821859,
     "ev1_numeric_w1": null,
     "human_retest_loss": null,
     "model_loss_vs_wave4": null,
     "ev1_status": "estimated",
     "trait_corr_error": null,
     "person_loss": 0.75732421875,
     "human_effect_pp": NaN,
     "model_effect_pp": NaN,
     "effect_error_pp": NaN,
     "subgroups": [
      [
       "Age band",
       "19–24 years",
       234,
       0.117
      ],
      [
       "Age band",
       "25–29 years",
       88,
       0.148
      ],
      [
       "Age band",
       "30–34 years",
       61,
       0.171
      ],
      [
       "Age band",
       "40–49 years",
       53,
       0.171
      ],
      [
       "English proficiency",
       "Very well / fluent / native",
       304,
       0.088
      ],
      [
       "English proficiency",
       "Well",
       208,
       0.113
      ],
      [
       "Sex",
       "female",
       335,
       0.069
      ],
      [
       "Sex",
       "male",
       171,
       0.11
      ]
     ]
    },
    "sapa_traits": {
     "answers": 10240,
     "valid": 10119,
     "ev1_tvd": 0.0539076310854867,
     "ev1_numeric_w1": null,
     "human_retest_loss": null,
     "model_loss_vs_wave4": null,
     "ev1_status": "estimated",
     "trait_corr_error": null,
     "person_loss": 0.7589555853865669,
     "human_effect_pp": NaN,
     "model_effect_pp": NaN,
     "effect_error_pp": NaN,
     "subgroups": [
      [
       "Age band",
       "19–24 years",
       239,
       0.093
      ],
      [
       "Age band",
       "25–29 years",
       97,
       0.167
      ],
      [
       "Age band",
       "30–34 years",
       56,
       0.168
      ],
      [
       "English proficiency",
       "Very well / fluent / native",
       296,
       0.082
      ],
      [
       "English proficiency",
       "Well",
       216,
       0.098
      ],
      [
       "Sex",
       "female",
       329,
       0.077
      ],
      [
       "Sex",
       "male",
       176,
       0.096
      ]
     ]
    },
    "twin_retest": {
     "answers": 12288,
     "valid": 12225,
     "ev1_tvd": 0.08592682003540622,
     "ev1_numeric_w1": 0.19800541996908338,
     "human_retest_loss": 0.44789162320972437,
     "model_loss_vs_wave4": 0.6390313595316702,
     "ev1_status": "estimated",
     "trait_corr_error": null,
     "person_loss": 0.6375857483633893,
     "human_effect_pp": NaN,
     "model_effect_pp": NaN,
     "effect_error_pp": NaN,
     "subgroups": [
      [
       "Age band",
       "18-29",
       103,
       0.162
      ],
      [
       "Age band",
       "30-49",
       193,
       0.129
      ],
      [
       "Age band",
       "50-64",
       157,
       0.147
      ],
      [
       "Age band",
       "65+",
       59,
       0.162
      ],
      [
       "Education",
       "Associate's degree",
       62,
       0.12
      ],
      [
       "Education",
       "College graduate/some postgrad",
       182,
       0.132
      ],
      [
       "Education",
       "High school graduate",
       66,
       0.14
      ],
      [
       "Education",
       "Postgraduate",
       76,
       0.139
      ],
      [
       "Education",
       "Some college, no degree",
       122,
       0.172
      ],
      [
       "Political affiliation",
       "Democrat",
       211,
       0.134
      ],
      [
       "Political affiliation",
       "Independent",
       145,
       0.147
      ],
      [
       "Political affiliation",
       "Republican",
       135,
       0.127
      ],
      [
       "Race or origin",
       "Black",
       77,
       0.155
      ],
      [
       "Race or origin",
       "Hispanic",
       55,
       0.113
      ],
      [
       "Race or origin",
       "White",
       318,
       0.108
      ],
      [
       "Sex assigned at birth",
       "Female",
       253,
       0.125
      ],
      [
       "Sex assigned at birth",
       "Male",
       259,
       0.121
      ],
      [
       "US region",
       "Midwest (ND, SD, NE, KS, MN, IA, MO, WI, IL, MI, IN, OH)",
       104,
       0.134
      ],
      [
       "US region",
       "Northeast (PA, NY, NJ, RI, CT, MA, VT, NH, ME)",
       78,
       0.125
      ],
      [
       "US region",
       "South (TX, OK, AR, LA, KY, TN, MS, AL, WV, DC, MD, DE, VA, NC, SC, GA, FL)",
       205,
       0.14
      ],
      [
       "US region",
       "West (WA, OR, ID, MT, WY, CA, NV, UT, CO, AZ, NM)",
       119,
       0.139
      ]
     ]
    }
   }
  }
 },
 "run_labels": {
  "thesia": "Thesia",
  "demographic_model_alone": "demographic model alone, no language model"
 }
}