5 Working with Data in the Tidyverse
https://learn.datacamp.com/courses/working-with-data-in-the-tidyverse
5.1 Explore your data
Load the readr package for every session so things work properly:
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(readr)
election <- read_csv("http://faculty.baruch.cuny.edu/geoportal/data/county_election/elpo12p010g.csv")## Rows: 3153 Columns: 14## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (5): FIPS, STATE_FIPS, STATE, COUNTY, WINNER
## dbl (9): OBAMA, ROMNEY, OTHERS, TTL_VT, PCT_OBM, PCT_ROM, PCT_OTHR, PCT_WNR,...
##
## 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.election## # A tibble: 3,153 × 14
## FIPS STATE_FIPS STATE COUNTY OBAMA ROMNEY OTHERS TTL_VT PCT_OBM PCT_ROM
## <chr> <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 01001 01 AL Autauga 6363 17379 231 23973 26.5 72.5
## 2 01003 01 AL Baldwin 18329 65772 887 84988 21.6 77.4
## 3 01005 01 AL Barbour 5912 5550 55 11517 51.3 48.2
## 4 01007 01 AL Bibb 2202 6132 86 8420 26.2 72.8
## 5 01009 01 AL Blount 2970 20757 333 24060 12.3 86.3
## 6 01011 01 AL Bullock 4061 1251 10 5322 76.3 23.5
## 7 01013 01 AL Butler 4374 5087 41 9502 46.0 53.5
## 8 01015 01 AL Calhoun 15500 30272 468 46240 33.5 65.5
## 9 01017 01 AL Chambers 6871 7626 132 14629 47.0 52.1
## 10 01019 01 AL Cherokee 2132 7506 154 9792 21.8 76.7
## # … with 3,143 more rows, and 4 more variables: PCT_OTHR <dbl>, WINNER <chr>,
## # PCT_WNR <dbl>, group <dbl>#To skip lines from the top, start from the 2nd line for example, use "skip = 1" after comma in read_csv(...,)readr is for rectangular data with extensions like: .csv, .tsv, .fwf, and .log
read_csv() function to read a csv file.
readxl() to read Microsoft Excel files.
Assign Missing Values
The read_csv() function also has an na argument, which allows you to specify value(s) that represent missing values in your data.
The default values for the na argument are c("", "NA"), so both are recorded as missing (NA) in R. When you read in data, you can add additional values like the string "UNKNOWN" to a vector of missing values using the c() function to combine multiple values into a single vector.
The is.na() function is also helpful for identifying rows with missing values for a variable.
read_csv("electiomn.csv", skip = 1, na = c("", "NA", "A new unspecified variable"))
election %>% filter(is.na(variable you want to filter to find na))The election data didn’t have any observations classified as a na so the example above is for when assuming there’s a missing value you want to filter/find.
How to list out all the variables of a dataset
Use glimpse() to list out the hidden variables in the dataset:
glimpse(election)## Rows: 3,153
## Columns: 14
## $ FIPS <chr> "01001", "01003", "01005", "01007", "01009", "01011", "0101…
## $ STATE_FIPS <chr> "01", "01", "01", "01", "01", "01", "01", "01", "01", "01",…
## $ STATE <chr> "AL", "AL", "AL", "AL", "AL", "AL", "AL", "AL", "AL", "AL",…
## $ COUNTY <chr> "Autauga", "Baldwin", "Barbour", "Bibb", "Blount", "Bullock…
## $ OBAMA <dbl> 6363, 18329, 5912, 2202, 2970, 4061, 4374, 15500, 6871, 213…
## $ ROMNEY <dbl> 17379, 65772, 5550, 6132, 20757, 1251, 5087, 30272, 7626, 7…
## $ OTHERS <dbl> 231, 887, 55, 86, 333, 10, 41, 468, 132, 154, 156, 38, 56, …
## $ TTL_VT <dbl> 23973, 84988, 11517, 8420, 24060, 5322, 9502, 46240, 14629,…
## $ PCT_OBM <dbl> 26.54236, 21.56657, 51.33281, 26.15202, 12.34414, 76.30590,…
## $ PCT_ROM <dbl> 72.49406, 77.38975, 48.18963, 72.82660, 86.27182, 23.50620,…
## $ PCT_OTHR <dbl> 0.963584, 1.043677, 0.477555, 1.021378, 1.384040, 0.187899,…
## $ WINNER <chr> "Romney", "Romney", "Obama", "Romney", "Romney", "Obama", "…
## $ PCT_WNR <dbl> 72.49406, 77.38975, 51.33281, 72.82660, 86.27182, 76.30590,…
## $ group <dbl> 24, 24, 12, 24, 25, 14, 22, 23, 22, 24, 24, 22, 22, 24, 25,…Combine glimpse() with other functions in a sequence using the pipe (%>%) operator. For example, function like arrange:
election %>%
arrange(OBAMA) %>%
glimpse() # no argument needed here## Rows: 3,153
## Columns: 14
## $ FIPS <chr> "48269", "48301", "31009", "31005", "31075", "48431", "4803…
## $ STATE_FIPS <chr> "48", "48", "31", "31", "31", "48", "48", "48", "51", "31",…
## $ STATE <chr> "TX", "TX", "NE", "NE", "NE", "TX", "TX", "TX", "VA", "NE",…
## $ COUNTY <chr> "King", "Loving", "Blaine", "Arthur", "Grant", "Sterling", …
## $ OBAMA <dbl> 5, 9, 29, 30, 30, 31, 32, 33, 37, 41, 42, 44, 49, 51, 55, 5…
## $ ROMNEY <dbl> 139, 54, 268, 227, 322, 459, 324, 468, 6463, 237, 360, 526,…
## $ OTHERS <dbl> 1, 1, 6, 5, 11, 4, 7, 7, 222, 13, 6, 8, 9, 12, 10, 7, 6, 6,…
## $ TTL_VT <dbl> 145, 64, 303, 262, 363, 494, 363, 508, 6722, 291, 408, 578,…
## $ PCT_OBM <dbl> 3.448276, 14.062500, 9.570957, 11.450382, 8.264463, 6.27530…
## $ PCT_ROM <dbl> 95.86207, 84.37500, 88.44885, 86.64122, 88.70523, 92.91498,…
## $ PCT_OTHR <dbl> 0.689655, 1.562500, 1.980198, 1.908397, 3.030303, 0.809717,…
## $ WINNER <chr> "Romney", "Romney", "Romney", "Romney", "Romney", "Romney",…
## $ PCT_WNR <dbl> 95.86207, 84.37500, 88.44885, 86.64122, 88.70523, 92.91498,…
## $ group <dbl> 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,…Compare to the two code chunks, this arraged data shows the ascending order of OBAMA.
How to get a list of summarized statistics of all the variables of a dataset
Use skim() to list the standard statistics of the variables:
library(skimr)
skim(election)| Name | election |
| Number of rows | 3153 |
| Number of columns | 14 |
| _______________________ | |
| Column type frequency: | |
| character | 5 |
| numeric | 9 |
| ________________________ | |
| Group variables | None |
Variable type: character
| skim_variable | n_missing | complete_rate | min | max | empty | n_unique | whitespace |
|---|---|---|---|---|---|---|---|
| FIPS | 0 | 1 | 5 | 5 | 0 | 3153 | 0 |
| STATE_FIPS | 0 | 1 | 2 | 2 | 0 | 51 | 0 |
| STATE | 0 | 1 | 2 | 2 | 0 | 51 | 0 |
| COUNTY | 0 | 1 | 2 | 23 | 0 | 1862 | 0 |
| WINNER | 0 | 1 | 5 | 6 | 0 | 2 | 0 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 | hist |
|---|---|---|---|---|---|---|---|---|---|---|
| OBAMA | 0 | 1 | 20897.02 | 73657.67 | 5.00 | 1572.00 | 3959.00 | 11259.00 | 2216903.00 | ▇▁▁▁▁ |
| ROMNEY | 0 | 1 | 19321.72 | 44465.73 | 54.00 | 2926.00 | 6294.00 | 16032.00 | 885333.00 | ▇▁▁▁▁ |
| OTHERS | 0 | 1 | 683.27 | 2262.16 | 0.00 | 72.00 | 178.00 | 485.00 | 78831.00 | ▇▁▁▁▁ |
| TTL_VT | 0 | 1 | 40902.02 | 116118.03 | 64.00 | 4862.00 | 10565.00 | 28049.00 | 3181067.00 | ▇▁▁▁▁ |
| PCT_OBM | 0 | 1 | 38.54 | 14.79 | 0.55 | 27.77 | 37.28 | 47.56 | 93.39 | ▂▇▇▂▁ |
| PCT_ROM | 0 | 1 | 59.63 | 14.78 | 5.98 | 50.47 | 60.72 | 70.25 | 96.15 | ▁▂▇▇▂ |
| PCT_OTHR | 0 | 1 | 1.83 | 1.00 | 0.00 | 1.17 | 1.69 | 2.35 | 14.87 | ▇▁▁▁▁ |
| PCT_WNR | 0 | 1 | 64.29 | 9.97 | 47.87 | 56.12 | 63.06 | 71.31 | 96.15 | ▇▇▆▂▁ |
| group | 0 | 1 | 20.70 | 4.52 | 11.00 | 22.00 | 23.00 | 24.00 | 25.00 | ▃▁▁▃▇ |
You can combine skim() with other functions in a sequence using the pipe (%>%) operator. For example, use function summary() to find how many variables of each type are in the dataset:
election %>%
skim() %>% # no argument needed here
summary() # no argument needed here| Name | Piped data |
| Number of rows | 3153 |
| Number of columns | 14 |
| _______________________ | |
| Column type frequency: | |
| character | 5 |
| numeric | 9 |
| ________________________ | |
| Group variables | None |
Count the data
Use distinct() to find out how many different type of observations are there in one variable. To find out how many distinct states are there in the STATE variable:
election %>%
distinct(STATE)## # A tibble: 51 × 1
## STATE
## <chr>
## 1 AL
## 2 AZ
## 3 AR
## 4 CA
## 5 CO
## 6 CT
## 7 DE
## 8 DC
## 9 FL
## 10 GA
## # … with 41 more rowscount() adds a new column named n to store the counts. this count function basically does the group_by and summarize for you. count the number of counties in each state:
election %>%
count(STATE)## # A tibble: 51 × 2
## STATE n
## <chr> <int>
## 1 AK 40
## 2 AL 67
## 3 AR 75
## 4 AZ 15
## 5 CA 58
## 6 CO 64
## 7 CT 8
## 8 DC 1
## 9 DE 3
## 10 FL 67
## # … with 41 more rowsAdapt the code to count by a logical condition instead:
election %>%
count(WINNER == "Obama")## # A tibble: 2 × 2
## `WINNER == "Obama"` n
## <lgl> <int>
## 1 FALSE 2449
## 2 TRUE 7045.2 Tame your data
Cast Column Types To assign or change the type of the columns in the dataset.
Use parse_date("2012-14-08", format = "%Y-%d-%m") then use col_date(format = "%Y-%d-%m") within cols() as the col_types argument of read_csv().
parse_date("2012-14-08", format = "%Y-%d-%m")
read_csv(...., col_types = cols( variale_name = col_date(format = "%Y-%d-%m")))read_csv(...., variable_name = col_number())Sometimes, there are na contained in the observations which would create errors while trying to cast.
Diagnose parsing problems using a new readr function called problems(). Using problems() on a result of read_csv() will show you the rows and columns where parsing error occurred, what the parser expected to find (for example, a number), and the actual value that caused the parsing error.
Factors are categorical variables, where the possible values are a fixed and known set.
Use parse_factor() to parse variables and col_factor() to cast columns as categorical. Both functions have a levels argument that is used to specify the possible values for the factors. When levels is set to NULL, the possible values will be inferred from the unique values in the dataset. Alternatively, you can pass a list of possible values.
read_csv(...., variable_name = col_factor(levels = NULL))For more details, go to the Cast A Factor And Examine Levels section of this chapter.
Recode Values
Use recode() function in the dplyr package. The recode function is to re-name the observations into something easier to understand.
election_CA <- election %>% filter(STATE == "CA")
election_CA %>%
mutate(STATE = recode(STATE, "CA" = "California"))## # A tibble: 58 × 14
## FIPS STATE_FIPS STATE COUNTY OBAMA ROMNEY OTHERS TTL_VT PCT_OBM PCT_ROM
## <chr> <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 06001 06 Californ… Alame… 469684 108182 17776 595642 78.9 18.2
## 2 06003 06 Californ… Alpine 389 236 28 653 59.6 36.1
## 3 06005 06 Californ… Amador 6830 10281 538 17649 38.7 58.3
## 4 06007 06 Californ… Butte 42669 44479 3604 90752 47.0 49.0
## 5 06009 06 Californ… Calav… 8670 12365 751 21786 39.8 56.8
## 6 06011 06 Californ… Colusa 2314 3601 119 6034 38.3 59.7
## 7 06013 06 Californ… Contr… 290824 136517 10885 438226 66.4 31.2
## 8 06015 06 Californ… Del N… 3791 4614 365 8770 43.2 52.6
## 9 06017 06 Californ… El Do… 35166 50973 2635 88774 39.6 57.4
## 10 06019 06 Californ… Fresno 129129 124490 5208 258827 49.9 48.1
## # … with 48 more rows, and 4 more variables: PCT_OTHR <dbl>, WINNER <chr>,
## # PCT_WNR <dbl>, group <dbl>Dummy variables are often used in data analysis to bin a variable into one of two categories to indicate the absence or presence of something. Dummy variables take the value 0 or 1 to stand for, for example, V_engine or S_engine.
Car_engine <- mtcars[1:10,] %>%
mutate(Engine_Type = recode(vs, "0" = "V_engine", .default = "S_engine")) %>%
select(vs, Engine_Type, everything(), -"carb")
Car_engine## vs Engine_Type mpg cyl disp hp drat wt qsec am gear
## Mazda RX4 0 V_engine 21.0 6 160.0 110 3.90 2.620 16.46 1 4
## Mazda RX4 Wag 0 V_engine 21.0 6 160.0 110 3.90 2.875 17.02 1 4
## Datsun 710 1 S_engine 22.8 4 108.0 93 3.85 2.320 18.61 1 4
## Hornet 4 Drive 1 S_engine 21.4 6 258.0 110 3.08 3.215 19.44 0 3
## Hornet Sportabout 0 V_engine 18.7 8 360.0 175 3.15 3.440 17.02 0 3
## Valiant 1 S_engine 18.1 6 225.0 105 2.76 3.460 20.22 0 3
## Duster 360 0 V_engine 14.3 8 360.0 245 3.21 3.570 15.84 0 3
## Merc 240D 1 S_engine 24.4 4 146.7 62 3.69 3.190 20.00 0 4
## Merc 230 1 S_engine 22.8 4 140.8 95 3.92 3.150 22.90 0 4
## Merc 280 1 S_engine 19.2 6 167.6 123 3.92 3.440 18.30 0 4#Since there're only 2 distinct levels, the "1" = "S_engine" can be coded as ".default".Select And Reorder Variables
Selecting a subset of columns to print can help check that a mutate() worked as expected, and rearranging columns next to each other can help spot obvious errors in data entry.
The select() helpers are functions that allow selection of variables based on their names:
## Function Usage
## 1 starts_with() starts with a prefix
## 2 ends_with() ends with a prefix
## 3 contains() contains a literal string
## 4 matches() matches a regular expression
## 5 num_range() a numerical range like x01, x02, x03.
## 6 one_of() variables in character vector.
## 7 everything() all variables.
## 8 last_col() last variable, possibly with an offset.# Move vs, and Engine_type to front and show only from mpg to drat:
Car_engine[1:10,] %>%
select(vs, Engine_Type, mpg:drat) ## vs Engine_Type mpg cyl disp hp drat
## Mazda RX4 0 V_engine 21.0 6 160.0 110 3.90
## Mazda RX4 Wag 0 V_engine 21.0 6 160.0 110 3.90
## Datsun 710 1 S_engine 22.8 4 108.0 93 3.85
## Hornet 4 Drive 1 S_engine 21.4 6 258.0 110 3.08
## Hornet Sportabout 0 V_engine 18.7 8 360.0 175 3.15
## Valiant 1 S_engine 18.1 6 225.0 105 2.76
## Duster 360 0 V_engine 14.3 8 360.0 245 3.21
## Merc 240D 1 S_engine 24.4 4 146.7 62 3.69
## Merc 230 1 S_engine 22.8 4 140.8 95 3.92
## Merc 280 1 S_engine 19.2 6 167.6 123 3.92Reformat Variable Names
To change names WITHOUT changing the order of the variables, write everything() first in the select() function.
The function clean_names() takes an argument case that can be used to convert variable names to other cases, like "upper_camel" or "all_caps".
use clean_names() from the janitor package to convert all variable names to snake_case.
install.packages("janitor")library(janitor)##
## Attaching package: 'janitor'## The following objects are masked from 'package:stats':
##
## chisq.test, fisher.testelection[1:10,] %>% clean_names("snake")## # A tibble: 10 × 14
## fips state_fips state county obama romney others ttl_vt pct_obm pct_rom
## <chr> <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 01001 01 AL Autauga 6363 17379 231 23973 26.5 72.5
## 2 01003 01 AL Baldwin 18329 65772 887 84988 21.6 77.4
## 3 01005 01 AL Barbour 5912 5550 55 11517 51.3 48.2
## 4 01007 01 AL Bibb 2202 6132 86 8420 26.2 72.8
## 5 01009 01 AL Blount 2970 20757 333 24060 12.3 86.3
## 6 01011 01 AL Bullock 4061 1251 10 5322 76.3 23.5
## 7 01013 01 AL Butler 4374 5087 41 9502 46.0 53.5
## 8 01015 01 AL Calhoun 15500 30272 468 46240 33.5 65.5
## 9 01017 01 AL Chambers 6871 7626 132 14629 47.0 52.1
## 10 01019 01 AL Cherokee 2132 7506 154 9792 21.8 76.7
## # … with 4 more variables: pct_othr <dbl>, winner <chr>, pct_wnr <dbl>,
## # group <dbl>#Notice how all the variables are lower_case now.How To Rename, Subset, And Reorder Variables At Once
To rename, then subset (choose to show only selected) , and finally reorder variables in one code line, use select().
dataset_name %>% select( new_variable_name_ = starts_with("old_name") ) This code will find all variables in dataset_name whose names start with old_name, then rename each variable as new_name_<N>, where N is a number. If dataset_name has variables oldname, oldname_v1, oldname3, then the code will replace these names with new_name_1, new_name_2, new_name_3.
The arguments inputted into select() determines what R will show. And, the order of the arguments inputted will determine how the resulting order of the variables will be.
election[1:10,] %>%
select(Precint_ = contains("PCT"), everything(), -"group")## # A tibble: 10 × 13
## Precint_1 Precint_2 Precint_3 Precint_4 FIPS STATE_FIPS STATE COUNTY OBAMA
## <dbl> <dbl> <dbl> <dbl> <chr> <chr> <chr> <chr> <dbl>
## 1 26.5 72.5 0.964 72.5 01001 01 AL Autauga 6363
## 2 21.6 77.4 1.04 77.4 01003 01 AL Baldwin 18329
## 3 51.3 48.2 0.478 51.3 01005 01 AL Barbour 5912
## 4 26.2 72.8 1.02 72.8 01007 01 AL Bibb 2202
## 5 12.3 86.3 1.38 86.3 01009 01 AL Blount 2970
## 6 76.3 23.5 0.188 76.3 01011 01 AL Bullock 4061
## 7 46.0 53.5 0.431 53.5 01013 01 AL Butler 4374
## 8 33.5 65.5 1.01 65.5 01015 01 AL Calhoun 15500
## 9 47.0 52.1 0.902 52.1 01017 01 AL Chambers 6871
## 10 21.8 76.7 1.57 76.7 01019 01 AL Cherokee 2132
## # … with 4 more variables: ROMNEY <dbl>, OTHERS <dbl>, TTL_VT <dbl>,
## # WINNER <chr>This example, the PCT_name has been replaced into Precint_<N>. The Precint_ is first in the select() function so it will be shown first, then next is everything() so everything else will be shown the way it is, then exclude the group variable.
5.3 Tidy your data
gather()
?gatherThe gather() function collapses multiple columns into two columns. It reshapes the dataset from wide to long, it reduces the number of columns and increases the number of rows.
library(tidyr)
Car_engine[1:2,] %>%
gather(key = "measurements", value = "specs", mpg:hp) %>%
select(measurements, specs)## measurements specs
## 1 mpg 21
## 2 mpg 21
## 3 cyl 6
## 4 cyl 6
## 5 disp 160
## 6 disp 160
## 7 hp 110
## 8 hp 110seperate()
Within a tidy dataset, a column should only represent only one variable, but if observations from that variable contains two distinct type of info, we can seperate it with seperate().
This is an example for the code before/after using the seperate funtion:
week_ratings
series episode viewers_7day
<fct> <chr> <dbl>
1 1 e1_7day 2.24
2 2 e1_7day 3.1
3 3 e1_7day 3.85
4 4 e1_7day 6.6
5 5 e1_7day 8.51
6 6 e1_7day 11.6
7 7 e1_7day 13.6
8 8 e1_7day 9.46
9 1 e2_7day 3
10 2 e2_7day 3.53
week_ratings <- ratings2 %>%
select(series, ends_with("7day")) %>%
gather(episode, viewers_7day, ends_with("7day"),
na.rm = TRUE) %>%
# Edit to separate key column and drop extra
separate(episode, into = c("episode","day"))
week_ratings
series episode day viewers_7day
<fct> <chr> <chr> <dbl>
1 1 e1 7day 2.24
2 2 e1 7day 3.1
3 3 e1 7day 3.85
4 4 e1 7day 6.6
5 5 e1 7day 8.51
6 6 e1 7day 11.6
7 7 e1 7day 13.6
8 8 e1 7day 9.46
9 1 e2 7day 3
10 2 e2 7day 3.53Unite Columns
In the tidyr package, the opposite of separate() isunite(). Sometimes you need to paste values from two or more columns together to tidy. Here is an example usage for unite():
data %>%
unite(new_var, old_var1, old_var2)To apply the function into the election data, merge the COUNTY into the STATE:
election_CA[1:5,] %>%
unite(LOCATION, STATE, COUNTY, sep = ", ")## # A tibble: 5 × 13
## FIPS STATE_FIPS LOCATION OBAMA ROMNEY OTHERS TTL_VT PCT_OBM PCT_ROM PCT_OTHR
## <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 06001 06 CA, Ala… 469684 108182 17776 595642 78.9 18.2 2.98
## 2 06003 06 CA, Alp… 389 236 28 653 59.6 36.1 4.29
## 3 06005 06 CA, Ama… 6830 10281 538 17649 38.7 58.3 3.05
## 4 06007 06 CA, But… 42669 44479 3604 90752 47.0 49.0 3.97
## 5 06009 06 CA, Cal… 8670 12365 751 21786 39.8 56.8 3.45
## # … with 3 more variables: WINNER <chr>, PCT_WNR <dbl>, group <dbl>spread()
Spreading reshapes the data from long to wide, adds columns and shrinks the rows.
tidy_ratings_all[1:10,]## series episode days viewers
## 1 1 1 7 2.24
## 2 2 1 7 3.10
## 3 3 1 7 3.85
## 4 4 1 7 6.60
## 5 5 1 7 8.51
## 6 6 1 7 11.62
## 7 7 1 7 13.58
## 8 8 1 7 9.46
## 9 6 1 28 11.73
## 10 7 1 28 13.86tidy_ratings_all %>%
# Count viewers by series and days
count(series, days, wt = viewers) %>%
# Adapt to spread counted values
spread(days, n, sep = "_")## series days_7 days_28
## 1 1 16.620 NA
## 2 2 31.610 NA
## 3 3 50.010 NA
## 4 4 73.540 NA
## 5 5 100.393 NA
## 6 6 123.110 113.00
## 7 7 135.630 138.45
## 8 8 90.170 92.875.4 Transform your data
How To Create A Range-filtered Column
Use case_when() to create a new column that represents a given range:
bakers_sample <- bakers[10:17,] %>%
select(baker, star_baker, technical_winner)
bakers_sample## baker star_baker technical_winner
## 10 Ruth 0 0
## 11 Ben 0 1
## 12 Holly 2 2
## 13 Ian 0 0
## 14 Janet 1 1
## 15 Jason 2 1
## 16 Joanne 1 3
## 17 Keith 0 0# Create skills variable with 4 levels
bakers_skill <- bakers_sample %>%
mutate(skill = case_when(
star_baker > technical_winner ~ "super_star",
star_baker < technical_winner ~ "high_tech",
star_baker == 0 & technical_winner == 0 ~ NA_character_,
star_baker == technical_winner ~ "well_rounded"
)) %>%
drop_na(skill)
bakers_skill## baker star_baker technical_winner skill
## 1 Ben 0 1 high_tech
## 2 Holly 2 2 well_rounded
## 3 Janet 1 1 well_rounded
## 4 Jason 2 1 super_star
## 5 Joanne 1 3 high_techCast A Factor And Examine Levels
Cast skill as a factor:
bakers_fct_skill <- bakers_skill %>%
mutate(skill = as.factor(skill))
# Examine levels
bakers_fct_skill %>%
pull(skill) %>%
levels()## [1] "high_tech" "super_star" "well_rounded"For more details, go to the Cast A Column As A Factor section of this chapter.
Cast Characters As Dates
Use lubridate to parse and cast a date variable within a mutate().
?lubridatelibrary(lubridate)##
## Attaching package: 'lubridate'## The following objects are masked from 'package:base':
##
## date, intersect, setdiff, unionbakers %>%
mutate(last_date_appeared = as.character(last_date_appeared)) %>%
mutate(last_date = ymd(last_date_appeared),
last_month = month(last_date_appeared, label = TRUE)) %>%
select(baker, last_date_appeared, last_date, last_month)## baker last_date_appeared last_date last_month
## 1 Annetha 2010-08-24 2010-08-24 Aug
## 2 David 2010-09-07 2010-09-07 Sep
## 3 Edd 2010-09-21 2010-09-21 Sep
## 4 Jasminder 2010-09-14 2010-09-14 Sep
## 5 Jonathan 2010-08-31 2010-08-31 Aug
## 6 Lea 2010-08-17 2010-08-17 Aug
## 7 Louise 2010-08-24 2010-08-24 Aug
## 8 Mark 2010-08-17 2010-08-17 Aug
## 9 Miranda 2010-09-21 2010-09-21 Sep
## 10 Ruth 2010-09-21 2010-09-21 Sep
## 11 Ben 2011-09-06 2011-09-06 Sep
## 12 Holly 2011-10-04 2011-10-04 Oct
## 13 Ian 2011-08-30 2011-08-30 Aug
## 14 Janet 2011-09-27 2011-09-27 Sep
## 15 Jason 2011-09-13 2011-09-13 Sep
## 16 Joanne 2011-10-04 2011-10-04 Oct
## 17 Keith 2011-08-16 2011-08-16 Aug
## 18 Mary-Anne 2011-10-04 2011-10-04 Oct
## 19 Robert 2011-09-13 2011-09-13 Sep
## 20 Simon 2011-08-23 2011-08-23 Aug
## 21 Urvashi 2011-08-30 2011-08-30 Aug
## 22 Yasmin 2011-09-20 2011-09-20 Sep
## 23 Brendan 2012-10-16 2012-10-16 Oct
## 24 Cathryn 2012-10-02 2012-10-02 Oct
## 25 Danny 2012-10-09 2012-10-09 Oct
## 26 James 2012-10-16 2012-10-16 Oct
## 27 John 2012-10-16 2012-10-16 Oct
## 28 Manisha 2012-09-11 2012-09-11 Sep
## 29 Natasha 2012-08-14 2012-08-14 Aug
## 30 Peter 2012-08-21 2012-08-21 Aug
## 31 Ryan 2012-09-25 2012-09-25 Sep
## 32 Sarah-Jane 2012-09-25 2012-09-25 Sep
## 33 Victoria 2012-08-28 2012-08-28 Aug
## 34 Stuart 2012-09-04 2012-09-04 Sep
## 35 Ali 2013-09-10 2013-09-10 Sep
## 36 Beca 2013-10-15 2013-10-15 Oct
## 37 Christine 2013-10-08 2013-10-08 Oct
## 38 Deborah 2013-09-03 2013-09-03 Sep
## 39 Frances 2013-10-22 2013-10-22 Oct
## 40 Glenn 2013-10-01 2013-10-01 Oct
## 41 Howard 2013-09-24 2013-09-24 Sep
## 42 Kimberley 2013-10-22 2013-10-22 Oct
## 43 Lucy 2013-08-27 2013-08-27 Aug
## 44 Mark 2013-09-03 2013-09-03 Sep
## 45 Robert 2013-09-17 2013-09-17 Sep
## 46 Ruby 2013-10-22 2013-10-22 Oct
## 47 Toby 2013-08-20 2013-08-20 Aug
## 48 Chetna 2014-10-01 2014-10-01 Oct
## 49 Claire 2014-08-06 2014-08-06 Aug
## 50 Diana 2014-09-03 2014-09-03 Sep
## 51 Enwezor 2014-08-13 2014-08-13 Aug
## 52 Iain 2014-08-27 2014-08-27 Aug
## 53 Jordan 2014-08-20 2014-08-20 Aug
## 54 Kate 2014-09-17 2014-09-17 Sep
## 55 Luis 2014-10-08 2014-10-08 Oct
## 56 Martha 2014-09-24 2014-09-24 Sep
## 57 Nancy 2014-10-08 2014-10-08 Oct
## 58 Norman 2014-09-03 2014-09-03 Sep
## 59 Richard 2014-10-08 2014-10-08 Oct
## 60 Alvin 2015-09-09 2015-09-09 Sep
## 61 Dorret 2015-08-19 2015-08-19 Aug
## 62 Flora 2015-09-30 2015-09-30 Sep
## 63 Ian 2015-10-07 2015-10-07 Oct
## 64 Marie 2015-08-12 2015-08-12 Aug
## 65 Mat 2015-09-16 2015-09-16 Sep
## 66 Nadiya 2015-10-07 2015-10-07 Oct
## 67 Paul 2015-09-23 2015-09-23 Sep
## 68 Sandy 2015-08-26 2015-08-26 Aug
## 69 Stu 2015-08-05 2015-08-05 Aug
## 70 Tamal 2015-10-07 2015-10-07 Oct
## 71 Ugnė 2015-09-02 2015-09-02 Sep
## 72 Andrew 2016-10-26 2016-10-26 Oct
## 73 Benjamina 2016-10-12 2016-10-12 Oct
## 74 Candice 2016-10-26 2016-10-26 Oct
## 75 Jane 2016-10-26 2016-10-26 Oct
## 76 Kate 2016-09-14 2016-09-14 Sep
## 77 Lee 2016-08-24 2016-08-24 Aug
## 78 Louise 2016-08-31 2016-08-31 Aug
## 79 Michael 2016-09-07 2016-09-07 Sep
## 80 Rav 2016-09-28 2016-09-28 Sep
## 81 Selasi 2016-10-19 2016-10-19 Oct
## 82 Tom 2016-10-05 2016-10-05 Oct
## 83 Val 2016-09-21 2016-09-21 Sep
## 84 Chris 2017-09-05 2017-09-05 Sep
## 85 Flo 2017-09-12 2017-09-12 Sep
## 86 James 2017-09-26 2017-09-26 Sep
## 87 Julia 2017-10-03 2017-10-03 Oct
## 88 Kate 2017-10-31 2017-10-31 Oct
## 89 Liam 2017-10-17 2017-10-17 Oct
## 90 Peter 2017-08-29 2017-08-29 Aug
## 91 Sophie 2017-10-31 2017-10-31 Oct
## 92 Stacey 2017-10-24 2017-10-24 Oct
## 93 Steven 2017-10-31 2017-10-31 Oct
## 94 Tom 2017-09-19 2017-09-19 Sep
## 95 Yan 2017-10-10 2017-10-10 OctIn the example above, I turned last_date_appeared into a character type variable first since it was a date already. But the point of this is proving the function of lubridate in converting character into date variable.
Calulate Timespans
The first step to calculating a timespan in lubridate is to make an interval, duration, period, then use division to convert the units to what wanted (like weeks(x) or months(x)). The x refers to the number of time units to be included in the period:
bakers_time <- bakers %>%
select(baker, first_date_appeared, last_date_appeared)
bakers_time[1:5,] %>%
mutate(time_on_air = interval(first_date_appeared, last_date_appeared),
weeks_on_air = time_on_air / weeks(1))## baker first_date_appeared last_date_appeared
## 1 Annetha 2010-08-17 2010-08-24
## 2 David 2010-08-17 2010-09-07
## 3 Edd 2010-08-17 2010-09-21
## 4 Jasminder 2010-08-17 2010-09-14
## 5 Jonathan 2010-08-17 2010-08-31
## time_on_air weeks_on_air
## 1 2010-08-17 UTC--2010-08-24 UTC 1
## 2 2010-08-17 UTC--2010-09-07 UTC 3
## 3 2010-08-17 UTC--2010-09-21 UTC 5
## 4 2010-08-17 UTC--2010-09-14 UTC 4
## 5 2010-08-17 UTC--2010-08-31 UTC 2String Wrangling
Use stringr package to transform, detect, replace, and remove observations within a data frame:
library(stringr)
election[1:7,] %>%
mutate(WINNER = str_to_upper(WINNER),
WINNER = str_replace(WINNER, "MA", "mshazam"),
WINNER = str_replace(WINNER, "RO", "Bolosho"),
GROUP_24 = str_detect(group, "24")) %>%
select(COUNTY, WINNER, GROUP_24)## # A tibble: 7 × 3
## COUNTY WINNER GROUP_24
## <chr> <chr> <lgl>
## 1 Autauga BoloshoMNEY TRUE
## 2 Baldwin BoloshoMNEY TRUE
## 3 Barbour OBAmshazam FALSE
## 4 Bibb BoloshoMNEY TRUE
## 5 Blount BoloshoMNEY FALSE
## 6 Bullock OBAmshazam FALSE
## 7 Butler BoloshoMNEY FALSE