Current replication
Spearman rank correlation
Not measuredNo current measurement is available in this registry.
Political Science · 2015
Causal Inference in Conjoint Analysis= Understanding Multidimensional Choices via Stated Preference Experiments
Voters preferred Ivy League candidates, not car dealers, Mormons or the old.
Read the paper ↗01 / Human finding
What the study found
Voters preferred Ivy League candidates, not car dealers, Mormons or the old.
Answer-key review: AI-reviewed
02 / Experiment design
Who chose, and what varied
- Country
- United States
- Human respondents
- 311
- Attributes
- 8
- Task
- discrete choice
03 / Synthetic replication
Registry replication results
Paper-era reference
2023 CLM mean Spearman ρ
0.05Historical results use their original estimator and reference. They are not a matched comparison with the current run.
Detailed replication evidence · 2026-10-02
Right direction on 13 of 14 significant effects, weak ranking (ρ = 0.55); effect sizes not yet settled (1.28× human).
Use it to screen which attributes move choice; calibrate sizes against a small human study before forecasting, and confirm the reversed effect with people.
- direction
- holds
- rank
- fails
- size
- inconclusive
Supported uses
- Screening which attributes move choice, and in which direction.
Limits to keep in view
- Ranking close levels against each other; the order diverges from humans.
- Forecasting the size of a shift without a human calibration anchor.
Effects on a common scale
Human effects: change in choice probability. Synthetic effects are multiplied by the record’s calibration factor, k = 0.7805. Calibration aligns magnitude; it does not establish agreement in unadjusted effect sizes.
| Attribute / level | Human | Human 95% half-width | Calibrated synthetic |
|---|---|---|---|
| Military Service: This candidate has never served in the military (reference) | 0.000 | Not measured | 0.000 |
| Military Service: This candidate served in the military | 0.090 | 0.040 | 0.163 |
| Religion: This candidate has no religion (reference) | 0.000 | Not measured | 0.000 |
| Religion: This candidate's religion is Jewish | -0.040 | 0.055 | 0.002 |
| Religion: This candidate's religion is Catholic | -0.035 | 0.055 | 0.015 |
| Religion: This candidate's religion is Mainline Protestant | -0.010 | 0.055 | 0.016 |
| Religion: This candidate's religion is Evangelical Protestant | -0.118 | 0.055 | 0.010 |
| Religion: This candidate's religion is Mormon | -0.142 | 0.055 | -0.016 |
| Higher Education: This candidate did not complete a bachelor's degree at any college or university (reference) | 0.000 | Not measured | 0.000 |
| Higher Education: This candidate completed a degree from a Baptist college | 0.145 | 0.055 | 0.137 |
| Higher Education: This candidate completed a degree from a community college | 0.150 | 0.055 | 0.088 |
| Higher Education: This candidate completed a degree from a state university | 0.195 | 0.055 | 0.151 |
| Higher Education: This candidate completed a degree from a small college | 0.190 | 0.055 | 0.068 |
| Higher Education: This candidate completed a degree from an Ivy League university | 0.280 | 0.055 | 0.207 |
| Profession: This candidate's profession is business owner (reference) | 0.000 | Not measured | 0.000 |
| Profession: This candidate's profession is lawyer | -0.022 | 0.060 | -0.076 |
| Profession: This candidate's profession is doctor | -0.023 | 0.050 | 0.006 |
| Profession: This candidate's profession is high school teacher | -0.040 | 0.050 | -0.053 |
| Profession: This candidate's profession is farmer | -0.090 | 0.060 | -0.043 |
| Profession: This candidate's profession is car dealer | -0.240 | 0.060 | -0.086 |
| Gender: This candidates' gender is male (reference) | 0.000 | Not measured | 0.000 |
| Gender: This candidate's gender is female | 0.002 | 0.040 | 0.035 |
| Income: This candidate's annual income is $32K (reference) | 0.000 | Not measured | 0.000 |
| Income: This candidate's annual income is $54K | 0.020 | 0.055 | 0.102 |
| Income: This candidate's annual income is $65K | 0.060 | 0.055 | 0.088 |
| Income: This candidate's annual income is $92K | 0.050 | 0.055 | 0.096 |
| Income: This candidate's annual income is $210K | 0.070 | 0.055 | 0.144 |
| Income: This candidate's annual income is $5.1M | 0.008 | 0.055 | 0.089 |
| Race: This candidate's race is White (reference) | 0.000 | Not measured | 0.000 |
| Race: This candidate's race is Native American | 0.040 | 0.060 | -0.178 |
| Race: This candidate's race is Black | 0.038 | 0.060 | -0.098 |
| Race: This candidate's race is Hispanic | -0.010 | 0.060 | -0.148 |
| Race: This candidate's race is Caucasian | 0.005 | 0.060 | -0.010 |
| Race: This candidate's race is Asian American | 0.040 | 0.060 | -0.209 |
| Age: This candidate's age is 36 (reference) | 0.000 | Not measured | 0.000 |
| Age: This candidate's age is 45 | 0.020 | 0.060 | -0.014 |
| Age: This candidate's age is 52 | 0.030 | 0.060 | 0.016 |
| Age: This candidate's age is 60 | 0.001 | 0.060 | -0.013 |
| Age: This candidate's age is 68 | -0.080 | 0.060 | -0.003 |
| Age: This candidate's age is 75 | -0.150 | 0.060 | -0.054 |
04 / Evidence and limitations
Follow the evidence
Missing results mean unmeasured, not zero agreement.