Datasets:
Formats:
json
Size:
< 1K
Tags:
spectral-brand-theory
exp-cross-language
r15
brand-perception
cross-language
measurement-invariance
DOI:
License:
Upload analysis/exp_cross_language_summary.md
Browse files
analysis/exp_cross_language_summary.md
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Experiment B Results: Cross-Language Semantic Drift
|
| 2 |
+
|
| 3 |
+
**Total valid records**: 441
|
| 4 |
+
|
| 5 |
+
## H1: Configural Invariance Under Bilingual Anchoring
|
| 6 |
+
|
| 7 |
+
- Overall median cosine: 0.992
|
| 8 |
+
- Overall min cosine: 0.966
|
| 9 |
+
- **SUPPORTED** (threshold: median > .90)
|
| 10 |
+
|
| 11 |
+
## H2: Native-Only Drift Magnitude
|
| 12 |
+
|
| 13 |
+
- Significant languages: 0/7 (threshold: 4/7)
|
| 14 |
+
- **NOT SUPPORTED**
|
| 15 |
+
|
| 16 |
+
| Language | Mean Abs Drift (pp) | p-value | Significant |
|
| 17 |
+
|----------|-------------------|---------|-------------|
|
| 18 |
+
| zh | 1.23 | 0.837 | No |
|
| 19 |
+
| ru | 1.59 | 0.833 | No |
|
| 20 |
+
| ja | 1.29 | 0.980 | No |
|
| 21 |
+
| ko | 0.73 | 0.985 | No |
|
| 22 |
+
| ar | 1.36 | 0.880 | No |
|
| 23 |
+
| hi | 1.19 | 0.974 | No |
|
| 24 |
+
| es | 1.71 | 0.926 | No |
|
| 25 |
+
|
| 26 |
+
## H3: Dimension-Specific Cultural Drift
|
| 27 |
+
|
| 28 |
+
- Confirmed predictions: 0/3 (threshold: 1/3)
|
| 29 |
+
- **NOT SUPPORTED**
|
| 30 |
+
|
| 31 |
+
## H4: Collectivist Social Amplification
|
| 32 |
+
|
| 33 |
+
- Collectivist mean Social: 12.14
|
| 34 |
+
- Individualist mean Social: 12.10
|
| 35 |
+
- p = 0.956, d = 0.009
|
| 36 |
+
- **NOT SUPPORTED**
|
| 37 |
+
|
| 38 |
+
---
|
| 39 |
+
*Generated by exp_cross_language_analysis.py*
|