GSS (2024) — Political-policy trio (death penalty + gun permit + marijuana)
https://gss.norc.org/get-the-data/ →benchmark:gss-policy-2024
Shape mix:
2
bimodal-asymmetric
1
multimodal-symmetric
Δ 0.25
3.74
()
→ 3.48
()
within-fixture gradient
Acceptance rubric
An SSR re-run acceptably reproduces this benchmark when it clears every check below on the same 3 paired questions the source publishes. The chip badges name the checks the round-38 → round-46 shape-mismatch finding argued are necessary but not sufficient in isolation.
CI overlap
3/3
Shape match
3/3
SSR mean inside CI · shape badge reproduces the source label.
What this benchmark discloses
Recruitment screener (verbatim)
GSS uses an address-based national probability sample of U.S. adults 18+. The 2024 cross-section ran mixed-mode (in-person + web push). No content screener. Sampling frame and selection probability are documented in the GSS 2024 Codebook R3 (NORC). cappun and gunlaw are on the larger ballot share (n ≈ 2,070 / 2,191); grass is on a smaller ballot share (n ≈ 862). Don't-know / no-answer respondents drop from each variable's n per the standard GSS convention for these dichotomous items.
Sample weighting & demographics
Weighted to wtssps (NORC's recommended GSS 2024 post-stratification weight, person-level). High_share is the weighted proportion answering 'favor' (cappun, gunlaw) or 'legal' (grass). Dichotomous → 1-5 SSR mapping via 1 + 4 × high_share with 'depends/dk' dropped.
Fieldwork dates, country, language
United States, 2024 cross-section. Fieldwork March–November 2024 per the GSS 2024 Codebook R3 (page 1). Modes: in-person and web push (push-to-web for the subsample without prior in-person contact).
Original question text & translations
cappun (codebook verbatim): "Do you favor or oppose the death penalty for persons convicted of murder?" gunlaw: "Would you favor or oppose a law which would require a person to obtain a police permit before he or she could buy a gun?" grass: "Do you think the use of marijuana should be made legal or not?"
Likert / scale labelling
All three items are dichotomous yes/no. cappun: 1 = 'favor' (high), 2 = 'oppose' (low). gunlaw: 1 = 'favor' (high), 2 = 'oppose' (low). grass: 1 = 'legal' (high), 2 = 'not legal' (low). The SSR rescale puts the policy-favoring answer at anchor 5 and the policy-opposing answer at anchor 1 across all three, so the 1-5 axis reads as 'support for the policy described' regardless of which direction is conservative vs progressive.
FGI moderator guide & coding rules
Closed-ended forced-choice, no FGI. No moderator guide. Each respondent answers each item independently in the same wave. cappun and gunlaw have run since 1972; grass since 1973. The 'support for the policy described' framing is an SSR-axis convention — the raw codebook does not direction-flip; we direction-flip only for visualisation comparability.
Data source & license
Public domain (NSF-funded). Required citation: Davern, Michael; Bautista, Rene; Freese, Jeremy; Herd, Pamela; and Morgan, Stephen L. General Social Survey 1972-2024. NORC ed., Chicago, 2025.
Anything not disclosed here was not available in the source dataset and is recorded as such.
SSR run — reproducibility keys
- template_version
- 3.0.0
- llm_seed
- 4216
- bootstrap_seed
- 8432
- llm_model
- deepseek-v4-flash
- embedding_model
- nvidia/nv-embed-v1
SSR score vs ground truth, by question
Each row holds one question. The teal mark and range are SSR. The sky mark and range are the source FGI/survey.
§
Do you favor or oppose the death penalty for persons convicted of murder?
SSR
—
Source
3.48 (3.40–3.57)
Source distribution
SD 1.94
skew -0.50
2 cells
multimodal-symmetric
1.02.03.04.05.0
§
Would you favor or oppose a law which would require a person to obtain a police permit before he or she could buy a gun?
SSR
—
Source
3.74 (3.66–3.82)
Source distribution
SD 1.86
skew -0.79
2 cells
bimodal-asymmetric
1.02.03.04.05.0
§
Do you think the use of marijuana should be made legal or not?
SSR
—
Source
3.68 (3.56–3.81)
Source distribution
SD 1.88
skew -0.73
2 cells
bimodal-asymmetric
1.02.03.04.05.0