What was your dataset?

Load your dataset in with the function below. The input is the date the dataset was issued. You should be able to get this from the tt_available() function.

coffee <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-07-07/coffee_ratings.csv')

Rows: 1339 Columns: 43
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (24): species, owner, country_of_origin, farm_name, lot_number, mill, ic...
dbl (19): total_cup_points, number_of_bags, aroma, flavor, aftertaste, acidi...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

What was your question?

Given your inital exploration of the data, what was the question you wanted to answer?

Does processing method affect overall coffee rating?

Initial Skim of Data

skimr::skim(coffee)

Data summary
Name	coffee
Number of rows	1339
Number of columns	43
_______________________
Column type frequency:
character	24
numeric	19
________________________
Group variables	None

Variable type: character

skim_variable	n_missing	complete_rate	min	max	n_unique
species	0	1.00	7	7	2
owner	7	0.99	3	50	315
country_of_origin	1	1.00	4	28	36
farm_name	359	0.73	1	73	571
lot_number	1063	0.21	1	71	227
mill	315	0.76	1	77	460
ico_number	151	0.89	1	40	847
company	209	0.84	3	73	281
altitude	226	0.83	1	41	396
region	59	0.96	2	76	356
producer	231	0.83	1	100	691
bag_weight	0	1.00	1	8	56
in_country_partner	0	1.00	7	85	27
harvest_year	47	0.96	3	24	46
grading_date	0	1.00	13	20	567
owner_1	7	0.99	3	50	319
variety	226	0.83	4	21	29
processing_method	170	0.87	5	25	5
color	218	0.84	4	12	4
expiration	0	1.00	13	20	566
certification_body	0	1.00	7	85	26
certification_address	0	1.00	40	40	32
certification_contact	0	1.00	40	40	29
unit_of_measurement	0	1.00	1	2	2

Variable type: numeric

skim_variable	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
total_cup_points	0	1.00	82.09	3.50	0	81.08	82.50	83.67	90.58	▁▁▁▁▇
number_of_bags	0	1.00	154.18	129.99	0	14.00	175.00	275.00	1062.00	▇▇▁▁▁
aroma	0	1.00	7.57	0.38	0	7.42	7.58	7.75	8.75	▁▁▁▁▇
flavor	0	1.00	7.52	0.40	0	7.33	7.58	7.75	8.83	▁▁▁▁▇
aftertaste	0	1.00	7.40	0.40	0	7.25	7.42	7.58	8.67	▁▁▁▁▇
acidity	0	1.00	7.54	0.38	0	7.33	7.58	7.75	8.75	▁▁▁▁▇
body	0	1.00	7.52	0.37	0	7.33	7.50	7.67	8.58	▁▁▁▁▇
balance	0	1.00	7.52	0.41	0	7.33	7.50	7.75	8.75	▁▁▁▁▇
uniformity	0	1.00	9.83	0.55	0	10.00	10.00	10.00	10.00	▁▁▁▁▇
clean_cup	0	1.00	9.84	0.76	0	10.00	10.00	10.00	10.00	▁▁▁▁▇
sweetness	0	1.00	9.86	0.62	0	10.00	10.00	10.00	10.00	▁▁▁▁▇
cupper_points	0	1.00	7.50	0.47	0	7.25	7.50	7.75	10.00	▁▁▁▇▁
moisture	0	1.00	0.09	0.05	0	0.09	0.11	0.12	0.28	▃▇▅▁▁
category_one_defects	0	1.00	0.48	2.55	0	0.00	0.00	0.00	63.00	▇▁▁▁▁
quakers	1	1.00	0.17	0.83	0	0.00	0.00	0.00	11.00	▇▁▁▁▁
category_two_defects	0	1.00	3.56	5.31	0	0.00	2.00	4.00	55.00	▇▁▁▁▁
altitude_low_meters	230	0.83	1750.71	8669.44	1	1100.00	1310.64	1600.00	190164.00	▇▁▁▁▁
altitude_high_meters	230	0.83	1799.35	8668.81	1	1100.00	1350.00	1650.00	190164.00	▇▁▁▁▁
altitude_mean_meters	230	0.83	1775.03	8668.63	1	1100.00	1310.64	1600.00	190164.00	▇▁▁▁▁

Total Number of Samples Per Country

coffee %>%
  janitor::tabyl(country_of_origin) %>%
  arrange(desc(n)) %>%
  gt::gt()

country_of_origin	n	percent	valid_percent
Mexico	236	0.176250934	0.1763826607
Colombia	183	0.136669156	0.1367713004
Guatemala	181	0.135175504	0.1352765321
Brazil	132	0.098581031	0.0986547085
Taiwan	75	0.056011949	0.0560538117
United States (Hawaii)	73	0.054518297	0.0545590433
Honduras	53	0.039581777	0.0396113602
Costa Rica	51	0.038088125	0.0381165919
Ethiopia	44	0.032860344	0.0328849028
Tanzania, United Republic Of	40	0.029873040	0.0298953662
Uganda	36	0.026885736	0.0269058296
Thailand	32	0.023898432	0.0239162930
Nicaragua	26	0.019417476	0.0194319880
Kenya	25	0.018670650	0.0186846039
El Salvador	21	0.015683346	0.0156950673
Indonesia	20	0.014936520	0.0149476831
China	16	0.011949216	0.0119581465
India	14	0.010455564	0.0104633782
Malawi	11	0.008215086	0.0082212257
Peru	10	0.007468260	0.0074738416
United States	10	0.007468260	0.0074738416
Myanmar	8	0.005974608	0.0059790732
Vietnam	8	0.005974608	0.0059790732
Haiti	6	0.004480956	0.0044843049
Philippines	5	0.003734130	0.0037369208
Panama	4	0.002987304	0.0029895366
United States (Puerto Rico)	4	0.002987304	0.0029895366
Ecuador	3	0.002240478	0.0022421525
Laos	3	0.002240478	0.0022421525
Burundi	2	0.001493652	0.0014947683
Cote d?Ivoire	1	0.000746826	0.0007473842
Japan	1	0.000746826	0.0007473842
Mauritius	1	0.000746826	0.0007473842
Papua New Guinea	1	0.000746826	0.0007473842
Rwanda	1	0.000746826	0.0007473842
Zambia	1	0.000746826	0.0007473842
NA	1	0.000746826	NA

Distribution of `total_cup_points` versus `processing_method`

ggplot(coffee) + 
  aes(y=total_cup_points, x=processing_method, fill=processing_method) +
  geom_boxplot() +
  theme(axis.text.x = element_text(angle=90, hjust = 1)) +
  coord_flip()

Counts of Country of Origin versus Processing Method

coffee %>%
  mutate(country_of_origin= fct_rev(country_of_origin)) %>%
  ggplot() +
  aes(y=country_of_origin, x=processing_method, 
      color=processing_method) +
  geom_count() +
  theme(axis.text.x = element_text(angle=90))

Here’s a sortable table of the above table

library(reactable)
coffee %>%
  janitor::tabyl(country_of_origin, processing_method) %>%
  reactable::reactable()

Sorted Heatmap of scores by `total_cup_points`

coffee %>% mutate(sample_id = rownames(coffee)) %>%
  select(sample_id, country_of_origin, total_cup_points, aroma, flavor, acidity, body, balance, uniformity, clean_cup, sweetness, cupper_points)%>%
  pivot_longer(cols = c(aroma, flavor, acidity, body, balance, uniformity, clean_cup, sweetness, cupper_points), names_to="type", values_to="score") %>%
  mutate(sample_id = fct_reorder(sample_id, total_cup_points)) %>%
  ggplot() +
  aes(y=sample_id, x=type, fill=score) +
    geom_tile()

Bi-clustered Heatmap of Scores

library(heatmaply)

Loading required package: plotly


Attaching package: 'plotly'

The following object is masked from 'package:ggplot2':

    last_plot

The following object is masked from 'package:stats':

    filter

The following object is masked from 'package:graphics':

    layout

Loading required package: viridis

Loading required package: viridisLite


======================
Welcome to heatmaply version 1.4.2

Type citation('heatmaply') for how to cite the package.
Type ?heatmaply for the main documentation.

The github page is: https://github.com/talgalili/heatmaply/
Please submit your suggestions and bug-reports at: https://github.com/talgalili/heatmaply/issues
You may ask questions at stackoverflow, use the r and heatmaply tags: 
     https://stackoverflow.com/questions/tagged/heatmaply
======================

coffee %>% mutate(sample_id = rownames(coffee)) %>%
  select(aroma, flavor, acidity, body, balance, uniformity, clean_cup, sweetness, cupper_points) %>% heatmaply()

Processing Method: `Dry / Natural`

coffee %>%
  filter(processing_method == "Natural / Dry") %>%
  mutate(country_of_origin = fct_reorder(country_of_origin, total_cup_points, median)) %>%
ggplot() + 
  aes(y=total_cup_points, x=country_of_origin, fill=country_of_origin) +
  geom_boxplot() +
  theme(axis.text.x = element_text(angle=90), legend.position = "none") +
  coord_flip() +
  labs(title="Tanzania leads with ratings in Natural/Dry")

Processing Method: `Washed / Wet`

coffee %>%
  filter(processing_method == "Washed / Wet") %>%
  mutate(country_of_origin = fct_reorder(country_of_origin, total_cup_points, median)) %>%
ggplot() + 
  aes(y=total_cup_points, x=country_of_origin, fill=country_of_origin) +
  geom_boxplot() +
  theme(axis.text.x = element_text(angle=90), legend.position = "none") +
  coord_flip() +
  labs(title="US leads in Ratings in Washed/Wet")

Mexico: Processing Methods

coffee %>%
  filter(country_of_origin == "Mexico") %>%
  mutate(processing_method = fct_reorder(processing_method, total_cup_points, median)) %>%
  ggplot() +
  aes(y=total_cup_points, x=processing_method,  fill=processing_method) +
  geom_boxplot(color="black") +
  coord_flip()

Linear model of `total_cup_points`

broom::tidy(lm(total_cup_points ~
                 country_of_origin +
                 category_one_defects, data=coffee)) %>%
  filter(p.value < 0.05) %>%
  arrange(p.value)

# A tibble: 9 × 5
  term                       estimate std.error statistic      p.value
  <chr>                         <dbl>     <dbl>     <dbl>        <dbl>
1 (Intercept)                  82.4      0.286     288.   0           
2 country_of_originHonduras    -3.06     0.535      -5.72 0.0000000133
3 country_of_originEthiopia     3.10     0.573       5.41 0.0000000757
4 category_one_defects         -0.163    0.0371     -4.40 0.0000116   
5 country_of_originMexico      -1.40     0.359      -3.90 0.000100    
6 country_of_originHaiti       -5.00     1.37       -3.64 0.000284    
7 country_of_originNicaragua   -1.92     0.706      -2.73 0.00649     
8 country_of_originKenya        1.89     0.718       2.63 0.00859     
9 country_of_originColombia     0.741    0.376       1.97 0.0488

Citation

BibTeX citation:

@online{laderas2020,
  author = {Laderas, Ted and Laderas, Ted},
  title = {Coffee {Data} {Exploration}},
  date = {2020-07-08},
  url = {https://laderast.github.io//articles/2020-07-08_coffee},
  langid = {en}
}

For attribution, please cite this work as:

Laderas, Ted, and Ted Laderas. 2020. “Coffee Data Exploration.” July 8, 2020. https://laderast.github.io//articles/2020-07-08_coffee.