ANALYTICS REPORT
TO: E & J GALLO WINERY
FROM: CHRISTIAN FARLIN
SUBJECT: PRELIMINARY WINE RATING AND PRICE ANALYSIS
DATE: 29 OCTOBER 2022
Introduction
I was given data on over 13,000 wines from around the world that included specific
attributes, such as ratings, types, and prices. I was tasked with determining how ratings and
prices compare across the wines and if there was a relevant relationship between ratings and
prices. To determine the correlation between wine rating and prices, I utilized descriptive
statistics, ANOVA testing, and visualizations. By testing the average wine rating and average
wine prices across the wine types, I concluded that there is a positive relationship between price
and rating, which means that if one attribute increases, then the other follows suit. I recommend
that the high-price high-rating relationship of Gallo’s Sparkling Wine warrants further analysis,
as an increase of price or of units available could be extremely profitable for the company.
Data Analysis
Descriptive Statistics
Comparison of Averages: The average price between the wine types reveals that we
should be able to find significant differences in price whereas the average rating between
the wine types do not clearly differ. We will confirm or reject these theories with
ANOVA testing.
Skewness and Kurtosis: The skewness and kurtosis values of the ratings of each wine
approach 0, which reveals that the rating data approaches normality. However, the
skewness and kurtosis values of the average wine prices are large and positive. Thus, they
are all right-tails that are stretched out due to outliers.
Possibility of Outliers: The presence of outliers is notable in the prices of each wine. For
example, the average price of the red wine is approximately $39 whereas the highest
priced—or the maximum price value of—red wine is $3,410. The difference between the
average and the maximum rating values for each wine is less pronounced. Thus, we need
stronger tests to determine if the wine ratings significantly differ.
Other Noteworthy Details: The average ratings for the red, white, rose, and sparkling
wines differ by less than one point. However, the average prices for each wine
significantly differ by at least $30. We intuitively notice the large average price
differences, but the slim differences between each average rating value is why we need to
employ ANOVA tests and detailed visualizations.
Rating ANOVA Results
Alternative Hypothesis HA: There is at least one difference between the average ratings of the
four wine types.
Null Hypothesis HO: There is no difference between the average ratings of the four tine types.
Because the p-value of 1.01E-74 is less than the significance level of 0.05, we can reject the Null
Hypothesis and can conclude the Alternative Hypothesis’ claim that there is at least one
difference between the average ratings of the four wine types.
Rating ANOVA Recommendation
Although we cannot determine which specific wine types differ with ANOVA, I noticed the high
average ratings of the Sparkling Wines. I recommend that E & J Gallo Winery focus their
marketing campaigns on the high ratings of this product line, raise the prices of these products,
and increase production of more Sparkling Wines to sell.
Price ANOVA Results
Alternative Hypothesis HA: There is at least one difference between the average prices of the
four wine types.
Null Hypothesis HO: There is no difference between the average prices of the four tine types.
Because the p-value of 1.3-52 is less than the significance level of 0.05, We can reject the Null
Hypothesis and can conclude the Alternative Hypothesis’ claim that there is at least one
difference between the average prices of the four wine types.
Price ANOVA Recommendation
Although we cannot determine which specific wine types differ with ANOVA, I noticed the
correlation between high ratings and high prices of Sparkling Wine. I recommend that E & J
Gallo Winery focus their marketing campaigns on the high-quality yet reasonably priced
Sparkling Wines. Due to the overwhelmingly positive ratings, I recommend increasing retail
prices and producing more Sparkling Wine to sell.
Bar Graph Analysis and Recommendation
The bar graph visually confirms our findings in ANOVA: there are clear relationships between
the average price and the average rating of each wine type. The increase in the average wine
price causes the average rating to simultaneously increase. This visualization also reveals that the
Sparkling Wine’s high ratings allow it to be sold for higher prices. E & J Gallo Winery can
justify increasing the price of their Sparkling Wines with their positive consumer feedback.
Conclusion
By testing the average wine rating and average wine prices across the wine types, I concluded
that there is a positive relationship between price and rating, which means that if one attribute
increases, then the other follows suit. I recommend that the high-price high-rating relationship of
Gallo’s Sparkling Wine warrants further analysis, as an increase in the price of units available is
extremely profitable for the company.
I recommend that E & J Gallo Winery focus their marketing campaigns on the high ratings of
their Sparkling Wines, increase the price of their Sparkling Wines, and produce more Sparkling
Wines to sell.
Thank you for engaging my services on this analysis. If you have queries about future steps
based upon this analysis or have questions on how I arrived to my conclusions and
recommendations, please e-mail me at [email protected]
Appendix
Red Wine Descriptive Statistics
Rating Price
Mean 3.89014784 Mean 39.14979
Standard Error 0.003314722 Standard Error 0.913111
Median 3.9 Median 18.2
Mode 3.8 Mode 9.9
Standard Deviation 0.308429494 Standard Deviation 84.9635
Sample Variance 0.095128753 Sample Variance 7218.797
Kurtosis 0.004399514 Kurtosis 353.2909
Skewness -0.058572728 Skewness 13.4079
Range 2.3 Range 3407.24
Minimum 2.5 Minimum 3.55
Maximum 4.8 Maximum 3410.79
Sum 33680.9 Sum 338958.9
Count 8658 Count 8658
Christian Farlin
White Wine Descriptive Statistics
Rating Price
Mean 3.818648577 Mean 20.64125565
Standard Error 0.004321079 Standard Error 0.504702267
Median 3.8 Median 13.15
Mode 3.8 Mode 9.9
Standard Deviation 0.264928287 Standard Deviation 30.94364131
Sample Variance 0.070186997 Sample Variance 957.5089376
Kurtosis 0.327932451 Kurtosis 142.2589129
Skewness -0.047218339 Skewness 9.713985369
Range 2.2 Range 677.63
Minimum 2.7 Minimum 3.74
Maximum 4.9 Maximum 681.37
Sum 14354.3 Sum 77590.48
Count 3759 Count 3759
Christian Farlin
Rose Wine Descriptive Statistics
Rating Price
Mean 3.741624365 Mean 12.58555838
Standard Error 0.013774061 Standard Error 0.810288402
Median 3.75 Median 8.92
Mode 3.8 Mode 7.9
Standard Deviation 0.273407303 Standard Deviation 16.08376555
Sample Variance 0.074751553 Sample Variance 258.6875143
Kurtosis 0.710464629 Kurtosis 124.8212095
Skewness 0.093996189 Skewness 9.610330501
Range 2.1 Range 245.3
Minimum 2.7 Minimum 3.7
Maximum 4.8 Maximum 249
Sum 1474.2 Sum 4958.71
Count 394 Count 394
Christian Farlin
Sparkling Wine Descriptive Statistics
Rating Price
Mean 4.065232975 Mean 60.53028674
Standard Error 0.017067116 Standard Error 4.43817948
Median 4.1 Median 34.9
Mode 3.9 Mode 14.9
Standard Deviation 0.285077043 Standard Deviation 74.13221263
Sample Variance 0.08126892 Sample Variance 5495.58495
Kurtosis -0.58992625 Kurtosis 8.996675801
Skewness 0.126229266 Skewness 2.715493188
Range 1.4 Range 488.5
Minimum 3.3 Minimum 6.5
Maximum 4.7 Maximum 495
Sum 1134.2 Sum 16887.95
Count 279 Count 279
Christian Farlin
Rating ANOVA Output
Anova: Single Factor
SUMMARY
Groups Count Sum Average Variance
Red Rating 8658 33680.9 3.89014784 0.095128753
White Rating 3759 14354.3 3.818648577 0.070186997
Rose Rating 394 1474.2 3.741624365 0.074751553
Sparkling Rating 279 1134.2 4.065232975 0.08126892
ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 30.53802618 3 10.17934206 116.9237767 1.00905E-74 2.605587681
Within Groups 1139.262466 13086 0.087059641
Total 1169.800492 13089
Christian Farlin
Price ANOVA Output
Anova: Single Factor
SUMMARY
Groups Count Sum Average Variance
Red Price 8658 338958.87 39.14978863 7218.796853
White Price 3759 77590.48 20.64125565 957.5089376
Rose Price 394 4958.71 12.58555838 258.6875143
Sparkling Price 279 16887.95 60.53028674 5495.58495
ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 1274092.389 3 424697.4629 82.06613706 1.35112E-52 2.605587681
Within Groups 67720879.75 13086 5175.063408
Total 68994972.14 13089
Christian Farlin
Overlapping Bar Graph
Farlin Comparison of Price and Rating Averages of Wine Types
70 4.1
60 4
50
Average Price (USD)
Average Rating
3.9
40
3.8
30
3.7
20
10 3.6
0 3.5
Rose White Red Sparkli ng
Type of Wine
Average of Price Average of Rating
