THE RESULT
What the data shows
The largest computed metric is 0.9981 for red / (5.0, 10.0]; the smallest is 0.9918 for white / (2.0, 5.0]. Metric: median_density (Source density units).
Bands are prespecified and sparse groups are excluded; density differences are not quality improvements.
| wine type | sugar band | samples | median density | p25 | p75 | group |
|---|---|---|---|---|---|---|
| red | (0.0, 2.0] | 620 | 0.9962 | 0.9952 | 0.997 | red / (0.0, 2.0] |
| red | (2.0, 5.0] | 895 | 0.9971 | 0.9958 | 0.9983 | red / (2.0, 5.0] |
| red | (5.0, 10.0] | 73 | 0.9981 | 0.9972 | 0.999 | red / (5.0, 10.0] |
| white | (0.0, 2.0] | 1,500 | 0.9919 | 0.9907 | 0.993 | white / (0.0, 2.0] |
| white | (10.0, 20.0] | 1,175 | 0.9978 | 0.9968 | 0.9986 | white / (10.0, 20.0] |
| white | (2.0, 5.0] | 910 | 0.9918 | 0.9906 | 0.9936 | white / (2.0, 5.0] |
| white | (5.0, 10.0] | 1,295 | 0.9949 | 0.9934 | 0.9958 | white / (5.0, 10.0] |
THE METHOD
From source to answer
Stratified median density and interquartile ranges.
The Python source exposes this study’s transformations. The complete project download includes shared preparation and evaluation routines.
THE NEXT DECISION
What follows from the finding
Use measurement structure to check plausibility and collinearity in a laboratory model.
Where the conclusion stops
Sensory scores are ordinal and concentrated in the middle. Producer and batch IDs are unavailable. Associations are not recipes for changing quality or evidence of market price.
Related studies may reuse observations or holdouts. These are historical analyses; associations and backtests do not demonstrate commercial impact. Further model tuning needs new, untouched evaluation data.