Scripts and visualizations to analyze a wine tasting experiment.
-
samples/survey_1_2.csv
- station_id,experiment_no
- 1A,1 => votes_1 == A, votes_2 == B
- 1A,2 => votes_1 == A, votes_2 == C
- 1B,1 => votes_1 == A, votes_2 == B, added_1 == votes added to votes_1 == A
- 1B,2 => votes_1 == C, votes_2 == A, added_1 == votes added to votes_1 == C
- 2A,1 => votes_1 == A, votes_2 == C special
- 2B,2 => votes_1 == A, votes_2 == B special
- votes_1
- votes_2
- added_1
- station_id,experiment_no
-
samples/survey_3.csv
-
treament - indicates first or second half of the experiment
-
wine - the wine shown
- 1 - earthy picture (D when treatment==1, E when treatment==2),
- 2 - berry picture (D when treatment==2, E when treatment==1)
wine,treatment 1,1 => D earthy 1,2 => E earthy 2,1 => E berry 2,2 => D berry
-
1,...,12 - the row index of the word used (lookup in
samples/descriptors.csv)
-
-
samples/descriptors.csv
- word - the word being used to describe. The row index can be used as a key to find out what word we are referring to in
survey_3.csv. - weight - the associated sweetness weight of the word. This is subjective but it should be a rough measure of how much the word connotates sweetness.
- word - the word being used to describe. The row index can be used as a key to find out what word we are referring to in
It turned out that the actual data we got back was a little different
than the sample data (particularly for experiments 1 and 2).
Everything outputted for the actual data is done in the actual
folder.