forked from sfirke/janitor
-
Notifications
You must be signed in to change notification settings - Fork 0
/
index.Rmd
193 lines (135 loc) · 7.59 KB
/
index.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
---
output:
github_document
---
<!-- index.md is generated from index.Rmd. Please edit that file -->
```{r, echo = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>",
fig.path = "README-"
)
options(width = 110)
```
# janitor <img src="man/figures/logo_small.png" align="right" />
> Data scientists, according to interviews and expert estimates, spend from 50 percent to 80 percent of their time mired in this more mundane labor of collecting and preparing unruly digital data, before it can be explored for useful nuggets.
>
> -- "[For Big-Data Scientists, 'Janitor Work' Is Key Hurdle to Insight"](https://www.nytimes.com/2014/08/18/technology/for-big-data-scientists-hurdle-to-insights-is-janitor-work.html) *(New York Times, 2014)*
***********************
[](https://travis-ci.org/sfirke/janitor)
[](https://codecov.io/github/sfirke/janitor?branch=master)
[](https://lifecycle.r-lib.org/articles/stages.html#stable)
[](https://cran.r-project.org/package=janitor)
**janitor** has simple functions for examining and cleaning dirty data. It was built with beginning and intermediate R users in mind and is optimized for user-friendliness. Advanced R users can already do everything covered here, but with janitor they can do it faster and save their thinking for the fun stuff.
The main janitor functions:
* perfectly format data.frame column names;
* create and format frequency tables of one, two, or three variables - think an improved `table()`; and
* provide other tools for cleaning and examining data.frames.
The tabulate-and-report functions approximate popular features of SPSS and Microsoft Excel.
janitor is a [#tidyverse](https://cran.r-project.org/web/packages/tidyverse/vignettes/manifesto.html)-oriented package. Specifically, it plays nicely with the `%>%` pipe and is optimized for cleaning data brought in with the [readr](https://github.com/tidyverse/readr) and [readxl](https://github.com/tidyverse/readxl) packages.
## <i class="fa fa-cog" aria-hidden="true"></i> Getting Started
You can install:
* the most recent officially-released version from CRAN with
```R
install.packages("janitor")
````
* the latest development version from GitHub with
```R
install.packages("devtools")
devtools::install_github("sfirke/janitor")
```
## Using janitor
A full description of each function, organized by topic, can be found in janitor's [catalog of functions vignette](https://sfirke.github.io/janitor/articles/janitor.html). There you will find functions not mentioned in this README, like `compare_df_cols()` which provides a summary of differences in column names and types when given a set of data.frames.
Below are quick examples of how janitor tools are commonly used.
### Cleaning dirty data
Take this roster of teachers at a fictional American high school, stored in the Microsoft Excel file [dirty_data.xlsx](https://github.com/sfirke/janitor/blob/master/dirty_data.xlsx):
Dirtiness includes:
* A header at the top
* Dreadful column names
* Rows and columns containing Excel formatting but no data
* Dates in two different formats in a single column (MM/DD/YYYY and numbers)
* Values spread inconsistently over the "Certification" columns
Here's that data after being read in to R:
```{r, warning = FALSE, message = FALSE}
library(readxl); library(janitor); library(dplyr); library(here)
roster_raw <- read_excel(here("dirty_data.xlsx")) # available at https://github.com/sfirke/janitor
glimpse(roster_raw)
```
Now, to clean it up, starting with the column names.
Name cleaning comes in two flavors. `make_clean_names()` operates on character vectors and can be used during data import:
```{r, warning = FALSE, message = FALSE}
roster_raw_cleaner <- read_excel(here("dirty_data.xlsx"),
skip = 1,
.name_repair = make_clean_names)
glimpse(roster_raw_cleaner)
```
`clean_names()` is a convenience version of `make_clean_names()` that can be used for piped data.frame workflows. The equivalent steps with `clean_names()` would be:
```{r, warning = FALSE}
roster_raw <- roster_raw %>%
row_to_names(row_number = 1) %>%
clean_names()
```
The data.frame now has clean names. Let's tidy it up further:
```{r}
roster <- roster_raw %>%
remove_empty(c("rows", "cols")) %>%
remove_constant(na.rm = TRUE, quiet = FALSE) %>% # remove the column of all "Yes" values
mutate(hire_date = convert_to_date(hire_date, # handle the mixed-format dates
character_fun = lubridate::mdy),
cert = dplyr::coalesce(certification, certification_2)) %>%
select(-certification, -certification_2) # drop unwanted columns
roster
```
### Examining dirty data
#### Finding duplicates
Use `get_dupes()` to identify and examine duplicate records during data cleaning. Let's see if any teachers are listed more than once:
```{r}
roster %>% get_dupes(contains("name"))
```
Yes, some teachers appear twice. We ought to address this before counting employees.
#### Tabulating tools
A variable (or combinations of two or three variables) can be tabulated with `tabyl()`. The resulting data.frame can be tweaked and formatted
with the suite of `adorn_` functions for quick analysis and printing of pretty results in a report. `adorn_` functions can be helpful with non-tabyls, too.
#### tabyl()
Like `table()`, but pipe-able, data.frame-based, and fully featured.
`tabyl` can be called two ways:
* On a vector, when tabulating a single variable: `tabyl(roster$subject)`
* On a data.frame, specifying 1, 2, or 3 variable names to tabulate: `roster %>% tabyl(subject, employee_status)`.
* Here the data.frame is passed in with the `%>%` pipe; this allows `tabyl` to be used in an analysis pipeline
One variable:
```{r}
roster %>%
tabyl(subject)
```
Two variables:
```{r}
roster %>%
filter(hire_date > as.Date("1950-01-01")) %>%
tabyl(employee_status, full_time)
```
Three variables:
```{r}
roster %>%
tabyl(full_time, subject, employee_status, show_missing_levels = FALSE)
```
**Adorning tabyls**
The `adorn_` functions dress up the results of these tabulation calls for fast, basic reporting. Here are some of the functions that augment a summary table for reporting:
```{r}
roster %>%
tabyl(employee_status, full_time) %>%
adorn_totals("row") %>%
adorn_percentages("row") %>%
adorn_pct_formatting() %>%
adorn_ns() %>%
adorn_title("combined")
```
Pipe that right into `knitr::kable()` in your RMarkdown report.
These modular adornments can be layered to reduce R's deficit against Excel and SPSS when it comes to quick, informative counts. Learn more about `tabyl()` and the `adorn_` functions from the [tabyls vignette](https://sfirke.github.io/janitor/articles/tabyls.html).
## <i class="fa fa-bullhorn" aria-hidden="true"></i> Contact Me
You are welcome to:
* submit suggestions and report bugs: https://github.com/sfirke/janitor/issues
* let me know what you think on Mastodon: [@samfirke@a2mi.social](https://a2mi.social/@samfirke)
* compose a friendly e-mail to: <img src = "https://samfirke.com/wp-content/uploads/2016/07/email_address_whitespace_top.png" alt = "samuel.firke AT gmail" width = "210"/>