List your company
ResearchFormerly WizWand

Calibration and Measurement on 14 social science constructs (Sample Dataset)

0.186T-ECE EpsilonQwen2.5-7B
0.164520.309510.45450.59949May 12, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Qwen2.5-7BMethod=Logit (Geom.)2026.050.1860.2650.409
Gemma-2-9BMethod=Logit (Geom.)2026.050.2160.2560.429
Ministral-8BMethod=Logit (Geom.)2026.050.230.2540.254
DeepSeek-R1-7BMethod=Logit (Geom.)2026.050.2340.2310.26
GPT-5-nanoMethod=Verbal2026.050.3540.3590.451
DeepSeek-V3.2Method=Verbal2026.050.3660.3820.467
Ministral-8BMethod=Logit (P-true)2026.050.3740.3270.254
DeepSeek-R1-7BMethod=Verbal2026.050.3780.3670.283
Qwen2.5-7BMethod=Verbal2026.050.4170.4110.402
Qwen2.5-7BMethod=Self-Random2026.050.4260.4140.455
DeepSeek-V3.2Method=Self-Random2026.050.4310.4130.501
Ministral-8BMethod=Self-Random2026.050.4550.3830.291
GPT-5-miniMethod=Verbal2026.050.4660.4380.574
Gemma-2-9BMethod=Logit (P-true)2026.050.4950.4540.429
Gemma-2-9BMethod=Verbal2026.050.4980.4760.418
Ministral-8BMethod=Verbal2026.050.5110.4950.465
Qwen3.5-FlashMethod=Verbal2026.050.5280.5030.544
DeepSeek-R1-7BMethod=Self-Random2026.050.5380.4790.236
Gemma-2-9BMethod=Self-Random2026.050.5650.5430.443
GPT-5-nanoMethod=Self-Random2026.050.5680.5580.478
Qwen3.5-FlashMethod=Self-Random2026.050.6040.5880.548
Qwen2.5-7BMethod=Logit (P-true)2026.050.6260.6030.409
GPT-5-miniMethod=Self-Random2026.050.6540.6480.559
DeepSeek-R1-7BMethod=Logit (P-true)2026.050.7230.6910.26