Althing (Sonnet 5.5)

4 runs · 4 datasets · 1 model

slug: althing-sonnet-5-5

0.805

Best SPS · opinionsqa

Disaggregated subgroup scorecard. Each card below is one published run for this vendor; expand the question-type and demographic-subgroup sections to see the matrix beneath the headline SPS. Where coverage permits, 95% CI bands accompany the point estimate.

What each cell measures

Each cell compares this vendor's synthetic responses against real human respondents in that demographic subgroup — published survey ground truth, not a model's guess about the subgroup. Higher = closer to how that real subgroup actually answered.

Demographic conditioning here is not stereotyping: every conditioned score is checked against what real subgroup members said, not against assumptions about them — and where the vendor's output diverges from the real subgroup, the score drops.

globalopinionqa — ground truth: Durmus et al. 2023, Anthropic — llm_global_opinions
gss — ground truth: NORC at the University of Chicago — General Social Survey
opinionsqa — ground truth: Santurkar et al., ICML 2023 — Whose Opinions Do LLMs Reflect? (derived from Pew American Trends Panel)
subpop — ground truth: Suh et al., ACL 2025 — SubPOP: Subpopulation-Level Opinion Prediction

Low-n cells: cells with n < 30 are shown muted and tagged low n — suggestive only instead of color-graded; they are reported for transparency, not as findings.

CIs: "no CI — single run" marks point estimates from a single run with no confidence interval yet; treat the uncertainty as unknown, never zero.

Columns: Score = distributional parity (p_dist) restricted to the subgroup · p_cond = conditioning strength vs the unconditioned baseline · N = questions answered under that conditioning · Cov. = coverage dot derived from N (green = high, ≥100 · amber = medium, 50–99 · red = low, <50). Topic tables: SPS = Survey Parity Score for the topic · p_dist = 1 − mean(JSD) · p_rank = (1 + mean(τ)) / 2 · p_refuse = 1 − mean(|R_model − R_human|).

Multiple comparisons: with this many subgroup cells, a few extreme cells are expected by chance alone — read patterns across a dimension, not single cells.

globalopinionqa product

althing--claude-sonnet-5-5--tdefault--tplcurrent--74dcb5d5

0.750 ± 0.032

SPS · 95% CI [0.717, 0.782] · n = 100

Question-type breakdown (6 topics)

Topic	SPS	p_dist	p_rank	p_refuse	N
Health & Science low n — suggestive only	0.887	0.775	1.000	1.000	1
General Attitudes low n — suggestive only	0.720	0.747	0.693	0.990	4
Economy & Work low n — suggestive only	0.702	0.737	0.667	0.949	3
International Relations & Security	0.700	0.675	0.725	0.905	60
Politics & Governance	0.618	0.615	0.620	0.860	30
Trust & Wellbeing low n — suggestive only	0.616	0.489	0.743	1.000	2

Topics with N < 10 are muted and tagged "low n — suggestive only": too few questions for a stable 3-decimal score.

Not yet measured

This vendor has no demographic-conditioned runs for age, geography, education, or any other subgroup dimension on globalopinionqa. No cells are fabricated — scores appear here only when a conditioned run actually measured them.

How to submit a demographic-conditioned run →

gss product

althing--claude-sonnet-5-5--tdefault--tplcurrent--1c3b6edc

0.721 ± 0.040

SPS · 95% CI [0.682, 0.762] · n = 75

Question-type breakdown (7 topics)

Topic	SPS	p_dist	p_rank	p_refuse	N
International Relations & Security	0.680	0.656	0.705	0.974	33
Trust & Wellbeing low n — suggestive only	0.595	0.590	0.600	0.951	3
Economy & Work low n — suggestive only	0.564	0.537	0.591	0.981	4
Health & Science low n — suggestive only	0.560	0.619	0.500	0.967	2
Politics & Governance low n — suggestive only	0.543	0.502	0.584	0.979	9
General Attitudes	0.517	0.508	0.527	0.960	10
Social Values & Religion	0.501	0.500	0.501	0.978	14

Topics with N < 10 are muted and tagged "low n — suggestive only": too few questions for a stable 3-decimal score.

Not yet measured

This vendor has no demographic-conditioned runs for age, geography, education, or any other subgroup dimension on gss. No cells are fabricated — scores appear here only when a conditioned run actually measured them.

How to submit a demographic-conditioned run →

opinionsqa product

althing--claude-sonnet-5-5--tdefault--tplcurrent--b88b652b

0.805 ± 0.025

SPS · 95% CI [0.778, 0.829] · n = 100

Question-type breakdown (9 topics)

Topic	SPS	p_dist	p_rank	p_refuse	N
Media & Information low n — suggestive only	0.851	0.841	0.862	0.996	2
Health & Science low n — suggestive only	0.806	0.758	0.854	0.997	1
Technology & Digital Life low n — suggestive only	0.786	0.756	0.816	0.995	4
Economy & Work	0.755	0.715	0.794	0.995	21
Trust & Wellbeing low n — suggestive only	0.743	0.669	0.816	0.994	1
General Attitudes	0.690	0.649	0.732	0.994	35
International Relations & Security	0.689	0.658	0.719	0.994	20
Social Values & Religion	0.682	0.641	0.722	0.993	15
Politics & Governance low n — suggestive only	0.576	0.494	0.658	0.996	1

Topics with N < 10 are muted and tagged "low n — suggestive only": too few questions for a stable 3-decimal score.

Not yet measured

This vendor has no demographic-conditioned runs for age, geography, education, or any other subgroup dimension on opinionsqa. No cells are fabricated — scores appear here only when a conditioned run actually measured them.

How to submit a demographic-conditioned run →

subpop product

althing--claude-sonnet-5-5--tdefault--tplcurrent--26d94120

0.695 ± 0.040

SPS · 95% CI [0.654, 0.734] · n = 100

Question-type breakdown (9 topics)

Topic	SPS	p_dist	p_rank	p_refuse	N
Economy & Work	0.738	0.703	0.773	0.992	13
Technology & Digital Life	0.729	0.685	0.772	0.993	16
General Attitudes	0.704	0.697	0.711	0.936	17
Politics & Governance low n — suggestive only	0.570	0.540	0.601	0.981	4
International Relations & Security	0.519	0.489	0.550	0.988	15
Media & Information low n — suggestive only	0.461	0.465	0.457	0.978	1
Health & Science low n — suggestive only	0.460	0.487	0.433	0.994	6
Social Values & Religion	0.324	0.274	0.374	0.985	27
Identity & Demographics low n — suggestive only	0.085	0.023	0.146	0.995	1

Topics with N < 10 are muted and tagged "low n — suggestive only": too few questions for a stable 3-decimal score.

Not yet measured

This vendor has no demographic-conditioned runs for age, geography, education, or any other subgroup dimension on subpop. No cells are fabricated — scores appear here only when a conditioned run actually measured them.

How to submit a demographic-conditioned run →

No demographic conditioning data has been published for this vendor yet. The question-type matrix above shows topic-level parity; subgroup rows fill in once Althing-style conditioned runs land.

← Back to leaderboard