Data Wrangling

Prerequisites

The Pipe Operator %>%

Three solutions to a single problem

What is the average of 4, 8, 16 approximately?

1.What is the average of 4, 8, 16 approximately?

2.What is the average of 4, 8, 16 approximately?

3.What is the average of 4, 8, 16 approximately?

Solution 1: Functions within Functions

c(4, 8, 16)
[1]  4  8 16

mean(c(4, 8, 16))
[1] 9.333333

round(mean(c(4, 8, 16)))
[1] 9

Problem with writing functions within functions

Things will get messy and more difficult to read and debug as we deal with more complex operations on data.

Solution 2: Creating Objects

numbers <- c(4, 8, 16)
numbers
[1]  4  8 16

avg_number <- mean(numbers)
avg_number
[1] 9.333333

round(avg_number)
[1] 9

Problem with creating many objects

We will end up with too many objects in Environment.

Solution 3: The (forward) Pipe Operator %>%


Shortcut:
Ctrl (Command) + Shift + M

c(4, 8, 16) %>% 
  mean() %>% 
  round()
[1] 9

Combine 4, 8, and 16 and then
Take the mean and then
Round the output

The output of the first function is the first argument of the second function.

Do you recall composite functions such as \(f \circ g(x)\)?

Now we have \(f \circ g \circ h (x)\) or round(mean(c(4, 8, 16)))

h(x) %>% 
  g() %>% 
  f()
c(4, 8, 16) %>% 
  mean() %>% 
  round()

Subsetting data frames

glimpse(lapd)
Rows: 68,564
Columns: 35
$ `Row ID`                       <chr> "3-1000027830ctFu", "3-1000155488ctFu",…
$ Year                           <dbl> 2013, 2013, 2013, 2013, 2013, 2013, 201…
$ `Department Title`             <chr> "Police (LAPD)", "Police (LAPD)", "Poli…
$ `Payroll Department`           <dbl> 4301, 4302, 4301, 4301, 4302, 4302, 430…
$ `Record Number`                <dbl> 1000027830, 1000155488, 1000194958, 100…
$ `Job Class Title`              <chr> "Police Detective II", "Clerk Typist", …
$ `Employment Type`              <chr> "Full Time", "Full Time", "Full Time", …
$ `Hourly or Event Rate`         <dbl> 53.16, 23.77, 60.80, 60.98, 45.06, 34.4…
$ `Projected Annual Salary`      <dbl> 110998.08, 49623.67, 126950.40, 127326.…
$ `Q1 Payments`                  <dbl> 24931.20, 11343.96, 24184.00, 29391.20,…
$ `Q2 Payments`                  <dbl> 29181.61, 13212.37, 28327.20, 36591.20,…
$ `Q3 Payments`                  <dbl> 26545.80, 11508.36, 28744.20, 32904.81,…
$ `Q4 Payments`                  <dbl> 29605.30, 13442.53, 33224.88, 37234.03,…
$ `Payments Over Base Pay`       <dbl> 4499.12, 1844.82, 13192.43, 18034.53, 1…
$ `% Over Base Pay`              <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
$ `Total Payments`               <dbl> 110263.91, 49507.22, 114480.28, 136121.…
$ `Base Pay`                     <dbl> 105764.79, 47662.40, 101287.85, 118086.…
$ `Permanent Bonus Pay`          <dbl> 3174.12, 0.00, 7363.95, 7086.67, 0.00, …
$ `Longevity Bonus Pay`          <dbl> 0.00, 1310.82, 0.00, 0.00, 1251.19, 172…
$ `Temporary Bonus Pay`          <dbl> 1325.00, 0.00, 1205.00, 1325.00, 125.00…
$ `Lump Sum Pay`                 <dbl> 0.00, 0.00, 2133.18, 0.00, 2068.80, 0.0…
$ `Overtime Pay`                 <dbl> 0.00, 0.00, 4424.32, 9839.33, 0.00, 0.0…
$ `Other Pay & Adjustments`      <dbl> 0.00, 534.00, -1934.02, -216.47, -2068.…
$ `Other Pay (Payroll Explorer)` <dbl> 4499.12, 1844.82, 8768.11, 8195.20, 137…
$ MOU                            <chr> "24", "3", "24", "24", "12", "3", "24",…
$ `MOU Title`                    <chr> "POLICE OFFICERS UNIT", "CLERICAL UNIT"…
$ `FMS Department`               <dbl> 70, 70, 70, 70, 70, 70, 70, 70, 70, 70,…
$ `Job Class`                    <chr> "2223", "1358", "2227", "2232", "1839",…
$ `Pay Grade`                    <chr> "2", "0", "1", "1", "0", "2", "3", "1",…
$ `Average Health Cost`          <dbl> 11651.40, 10710.24, 11651.40, 11651.40,…
$ `Average Dental Cost`          <dbl> 898.08, 405.24, 898.08, 898.08, 405.24,…
$ `Average Basic Life`           <dbl> 191.04, 11.40, 191.04, 191.04, 11.40, 1…
$ `Average Benefit Cost`         <dbl> 12740.52, 11126.88, 12740.52, 12740.52,…
$ `Benefits Plan`                <chr> "Police", "City", "Police", "Police", "…
$ `Job Class Link`               <chr> "http://per.lacity.org/perspecs/2223.pd…

lapd <- clean_names(lapd)
glimpse(lapd)
Rows: 68,564
Columns: 35
$ row_id                     <chr> "3-1000027830ctFu", "3-1000155488ctFu", "3-…
$ year                       <dbl> 2013, 2013, 2013, 2013, 2013, 2013, 2013, 2…
$ department_title           <chr> "Police (LAPD)", "Police (LAPD)", "Police (…
$ payroll_department         <dbl> 4301, 4302, 4301, 4301, 4302, 4302, 4301, 4…
$ record_number              <dbl> 1000027830, 1000155488, 1000194958, 1000232…
$ job_class_title            <chr> "Police Detective II", "Clerk Typist", "Pol…
$ employment_type            <chr> "Full Time", "Full Time", "Full Time", "Ful…
$ hourly_or_event_rate       <dbl> 53.16, 23.77, 60.80, 60.98, 45.06, 34.42, 4…
$ projected_annual_salary    <dbl> 110998.08, 49623.67, 126950.40, 127326.24, …
$ q1_payments                <dbl> 24931.20, 11343.96, 24184.00, 29391.20, 208…
$ q2_payments                <dbl> 29181.61, 13212.37, 28327.20, 36591.20, 241…
$ q3_payments                <dbl> 26545.80, 11508.36, 28744.20, 32904.81, 215…
$ q4_payments                <dbl> 29605.30, 13442.53, 33224.88, 37234.03, 252…
$ payments_over_base_pay     <dbl> 4499.12, 1844.82, 13192.43, 18034.53, 1376.…
$ percent_over_base_pay      <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
$ total_payments             <dbl> 110263.91, 49507.22, 114480.28, 136121.24, …
$ base_pay                   <dbl> 105764.79, 47662.40, 101287.85, 118086.71, …
$ permanent_bonus_pay        <dbl> 3174.12, 0.00, 7363.95, 7086.67, 0.00, 0.00…
$ longevity_bonus_pay        <dbl> 0.00, 1310.82, 0.00, 0.00, 1251.19, 1726.16…
$ temporary_bonus_pay        <dbl> 1325.00, 0.00, 1205.00, 1325.00, 125.00, 68…
$ lump_sum_pay               <dbl> 0.00, 0.00, 2133.18, 0.00, 2068.80, 0.00, 0…
$ overtime_pay               <dbl> 0.00, 0.00, 4424.32, 9839.33, 0.00, 0.00, 4…
$ other_pay_adjustments      <dbl> 0.00, 534.00, -1934.02, -216.47, -2068.80, …
$ other_pay_payroll_explorer <dbl> 4499.12, 1844.82, 8768.11, 8195.20, 1376.19…
$ mou                        <chr> "24", "3", "24", "24", "12", "3", "24", "24…
$ mou_title                  <chr> "POLICE OFFICERS UNIT", "CLERICAL UNIT", "P…
$ fms_department             <dbl> 70, 70, 70, 70, 70, 70, 70, 70, 70, 70, 70,…
$ job_class                  <chr> "2223", "1358", "2227", "2232", "1839", "22…
$ pay_grade                  <chr> "2", "0", "1", "1", "0", "2", "3", "1", "B"…
$ average_health_cost        <dbl> 11651.40, 10710.24, 11651.40, 11651.40, 107…
$ average_dental_cost        <dbl> 898.08, 405.24, 898.08, 898.08, 405.24, 405…
$ average_basic_life         <dbl> 191.04, 11.40, 191.04, 191.04, 11.40, 11.40…
$ average_benefit_cost       <dbl> 12740.52, 11126.88, 12740.52, 12740.52, 111…
$ benefits_plan              <chr> "Police", "City", "Police", "Police", "City…
$ job_class_link             <chr> "http://per.lacity.org/perspecs/2223.pdf", …

subsetting variables/columns

select()

subsetting observations/rows

slice() and filter()

select is used to select certain variables in the data frame.

select(lapd, year, base_pay)
# A tibble: 68,564 × 2
    year base_pay
   <dbl>    <dbl>
 1  2013  105765.
 2  2013   47662.
 3  2013  101288.
 4  2013  118087.
 5  2013   90322.
 6  2013   62770.
 7  2013   93718.
 8  2013       0 
 9  2013   51246.
10  2013   74227.
# ℹ 68,554 more rows
lapd %>% 
  select(year, base_pay)
# A tibble: 68,564 × 2
    year base_pay
   <dbl>    <dbl>
 1  2013  105765.
 2  2013   47662.
 3  2013  101288.
 4  2013  118087.
 5  2013   90322.
 6  2013   62770.
 7  2013   93718.
 8  2013       0 
 9  2013   51246.
10  2013   74227.
# ℹ 68,554 more rows

select can also be used to drop certain variables if used with a negative sign.

select(lapd, -row_id, -department_title)
# A tibble: 68,564 × 33
    year payroll_department record_number job_class_title        employment_type
   <dbl>              <dbl>         <dbl> <chr>                  <chr>          
 1  2013               4301    1000027830 Police Detective II    Full Time      
 2  2013               4302    1000155488 Clerk Typist           Full Time      
 3  2013               4301    1000194958 Police Sergeant I      Full Time      
 4  2013               4301    1000232317 Police Lieutenant I    Full Time      
 5  2013               4302    1000329284 Principal Storekeeper  Full Time      
 6  2013               4302    1001124320 Police Service Repres… Full Time      
 7  2013               4301    1001221822 Police Officer III     Full Time      
 8  2013               4301    1001243583 Police Sergeant I      Full Time      
 9  2013               4301    1001317832 Police Officer II      Full Time      
10  2013               4301     100162910 Police Officer II      Full Time      
# ℹ 68,554 more rows
# ℹ 28 more variables: hourly_or_event_rate <dbl>,
#   projected_annual_salary <dbl>, q1_payments <dbl>, q2_payments <dbl>,
#   q3_payments <dbl>, q4_payments <dbl>, payments_over_base_pay <dbl>,
#   percent_over_base_pay <dbl>, total_payments <dbl>, base_pay <dbl>,
#   permanent_bonus_pay <dbl>, longevity_bonus_pay <dbl>,
#   temporary_bonus_pay <dbl>, lump_sum_pay <dbl>, overtime_pay <dbl>, …

Selection helpers

starts_with()
ends_with()
contains()

select(lapd, starts_with("q"))
# A tibble: 68,564 × 4
   q1_payments q2_payments q3_payments q4_payments
         <dbl>       <dbl>       <dbl>       <dbl>
 1      24931.      29182.      26546.      29605.
 2      11344.      13212.      11508.      13443.
 3      24184       28327.      28744.      33225.
 4      29391.      36591.      32905.      37234.
 5      20813       24136       21518.      25231.
 6      16057.      17927.      14150.      17052.
 7      22162.      25664.      23404.      24586.
 8          0           0         331.          0 
 9      11941.      14330.      13404.      14537.
10      17046.      20457.      18777.      21371.
# ℹ 68,554 more rows

select(lapd, ends_with("pay"))
# A tibble: 68,564 × 8
   payments_over_base_pay percent_over_base_pay base_pay permanent_bonus_pay
                    <dbl>                 <dbl>    <dbl>               <dbl>
 1                  4499.                     0  105765.               3174.
 2                  1845.                     0   47662.                  0 
 3                 13192.                     0  101288.               7364.
 4                 18035.                     0  118087.               7087.
 5                  1376.                     0   90322.                  0 
 6                  2415.                     0   62770.                  0 
 7                  2099.                     0   93718.                866.
 8                   331.                     0       0                   0 
 9                  2967.                     0   51246.               1540.
10                  3424.                     0   74227.               2233.
# ℹ 68,554 more rows
# ℹ 4 more variables: longevity_bonus_pay <dbl>, temporary_bonus_pay <dbl>,
#   lump_sum_pay <dbl>, overtime_pay <dbl>

select(lapd, contains("pay"))
# A tibble: 68,564 × 17
   payroll_department q1_payments q2_payments q3_payments q4_payments
                <dbl>       <dbl>       <dbl>       <dbl>       <dbl>
 1               4301      24931.      29182.      26546.      29605.
 2               4302      11344.      13212.      11508.      13443.
 3               4301      24184       28327.      28744.      33225.
 4               4301      29391.      36591.      32905.      37234.
 5               4302      20813       24136       21518.      25231.
 6               4302      16057.      17927.      14150.      17052.
 7               4301      22162.      25664.      23404.      24586.
 8               4301          0           0         331.          0 
 9               4301      11941.      14330.      13404.      14537.
10               4301      17046.      20457.      18777.      21371.
# ℹ 68,554 more rows
# ℹ 12 more variables: payments_over_base_pay <dbl>,
#   percent_over_base_pay <dbl>, total_payments <dbl>, base_pay <dbl>,
#   permanent_bonus_pay <dbl>, longevity_bonus_pay <dbl>,
#   temporary_bonus_pay <dbl>, lump_sum_pay <dbl>, overtime_pay <dbl>,
#   other_pay_adjustments <dbl>, other_pay_payroll_explorer <dbl>,
#   pay_grade <chr>

subsetting variables/columns

select()

subsetting observations/rows

slice() and filter()

slice() subsetting rows based on a row number.

The data below include all the rows from third to seventh. Including third and seventh.

slice(lapd, 3:7)
# A tibble: 5 × 35
  row_id  year department_title payroll_department record_number job_class_title
  <chr>  <dbl> <chr>                         <dbl>         <dbl> <chr>          
1 3-100…  2013 Police (LAPD)                  4301    1000194958 Police Sergean…
2 3-100…  2013 Police (LAPD)                  4301    1000232317 Police Lieuten…
3 3-100…  2013 Police (LAPD)                  4302    1000329284 Principal Stor…
4 3-100…  2013 Police (LAPD)                  4302    1001124320 Police Service…
5 3-100…  2013 Police (LAPD)                  4301    1001221822 Police Officer…
# ℹ 29 more variables: employment_type <chr>, hourly_or_event_rate <dbl>,
#   projected_annual_salary <dbl>, q1_payments <dbl>, q2_payments <dbl>,
#   q3_payments <dbl>, q4_payments <dbl>, payments_over_base_pay <dbl>,
#   percent_over_base_pay <dbl>, total_payments <dbl>, base_pay <dbl>,
#   permanent_bonus_pay <dbl>, longevity_bonus_pay <dbl>,
#   temporary_bonus_pay <dbl>, lump_sum_pay <dbl>, overtime_pay <dbl>,
#   other_pay_adjustments <dbl>, other_pay_payroll_explorer <dbl>, mou <chr>, …

filter() subsetting rows based on a condition.

The data below includes rows when the recorded year is 2018.

filter(lapd, year == 2018)
# A tibble: 14,824 × 35
   row_id            year department_title payroll_department record_number
   <chr>            <dbl> <chr>                         <dbl>         <dbl>
 1 8-1000027830ctFu  2018 Police (LAPD)                  4301    1000027830
 2 8-1000194958ctFu  2018 Police (LAPD)                  4301    1000194958
 3 8-1000232317ctFu  2018 Police (LAPD)                  4301    1000232317
 4 8-1001124320ctFu  2018 Police (LAPD)                  4302    1001124320
 5 8-1001221822ctFu  2018 Police (LAPD)                  4301    1001221822
 6 8-1001317832ctFu  2018 Police (LAPD)                  4301    1001317832
 7 8-100162910ctFu   2018 Police (LAPD)                  4301     100162910
 8 8-1001675957ctFu  2018 Police (LAPD)                  4301    1001675957
 9 8-1001884819ctFu  2018 Police (LAPD)                  4302    1001884819
10 8-1001893163ctFu  2018 Police (LAPD)                  4302    1001893163
# ℹ 14,814 more rows
# ℹ 30 more variables: job_class_title <chr>, employment_type <chr>,
#   hourly_or_event_rate <dbl>, projected_annual_salary <dbl>,
#   q1_payments <dbl>, q2_payments <dbl>, q3_payments <dbl>, q4_payments <dbl>,
#   payments_over_base_pay <dbl>, percent_over_base_pay <dbl>,
#   total_payments <dbl>, base_pay <dbl>, permanent_bonus_pay <dbl>,
#   longevity_bonus_pay <dbl>, temporary_bonus_pay <dbl>, lump_sum_pay <dbl>, …

Relational Operators in R

Operator Description
< Less than
> Greater than
<= Less than or equal to
>= Greater than or equal to
== Equal to
!= Not equal to

Logical Operators in R

Operator Description
& and
| or

Q. According to datausa.io Los Angeles had a median household income of $62474 in 2018. How many LAPD staff members had a base pay higher than $62474 in year 2018 according to this data?

Median household income is not the same thing as median employee income. Our aim is data wrangling and not necessarily statistical analysis for now.

lapd %>% 
  filter(year == 2018 & base_pay > 62474)
# A tibble: 11,690 × 35
   row_id            year department_title payroll_department record_number
   <chr>            <dbl> <chr>                         <dbl>         <dbl>
 1 8-1000027830ctFu  2018 Police (LAPD)                  4301    1000027830
 2 8-1000194958ctFu  2018 Police (LAPD)                  4301    1000194958
 3 8-1000232317ctFu  2018 Police (LAPD)                  4301    1000232317
 4 8-1001124320ctFu  2018 Police (LAPD)                  4302    1001124320
 5 8-1001221822ctFu  2018 Police (LAPD)                  4301    1001221822
 6 8-1001317832ctFu  2018 Police (LAPD)                  4301    1001317832
 7 8-100162910ctFu   2018 Police (LAPD)                  4301     100162910
 8 8-1001675957ctFu  2018 Police (LAPD)                  4301    1001675957
 9 8-1001884819ctFu  2018 Police (LAPD)                  4302    1001884819
10 8-1001893163ctFu  2018 Police (LAPD)                  4302    1001893163
# ℹ 11,680 more rows
# ℹ 30 more variables: job_class_title <chr>, employment_type <chr>,
#   hourly_or_event_rate <dbl>, projected_annual_salary <dbl>,
#   q1_payments <dbl>, q2_payments <dbl>, q3_payments <dbl>, q4_payments <dbl>,
#   payments_over_base_pay <dbl>, percent_over_base_pay <dbl>,
#   total_payments <dbl>, base_pay <dbl>, permanent_bonus_pay <dbl>,
#   longevity_bonus_pay <dbl>, temporary_bonus_pay <dbl>, lump_sum_pay <dbl>, …

lapd %>% 
  filter(year == 2018 & base_pay > 62474) %>% 
  nrow()
[1] 11690

Q. How many observations are available between 2013 and 2015 including 2013 and 2015?

lapd %>% 
  filter(year >= 2013 & year <= 2015)
# A tibble: 40,227 × 35
   row_id            year department_title payroll_department record_number
   <chr>            <dbl> <chr>                         <dbl>         <dbl>
 1 3-1000027830ctFu  2013 Police (LAPD)                  4301    1000027830
 2 3-1000155488ctFu  2013 Police (LAPD)                  4302    1000155488
 3 3-1000194958ctFu  2013 Police (LAPD)                  4301    1000194958
 4 3-1000232317ctFu  2013 Police (LAPD)                  4301    1000232317
 5 3-1000329284ctFu  2013 Police (LAPD)                  4302    1000329284
 6 3-1001124320ctFu  2013 Police (LAPD)                  4302    1001124320
 7 3-1001221822ctFu  2013 Police (LAPD)                  4301    1001221822
 8 3-1001243583ctFu  2013 Police (LAPD)                  4301    1001243583
 9 3-1001317832ctFu  2013 Police (LAPD)                  4301    1001317832
10 3-100162910ctFu   2013 Police (LAPD)                  4301     100162910
# ℹ 40,217 more rows
# ℹ 30 more variables: job_class_title <chr>, employment_type <chr>,
#   hourly_or_event_rate <dbl>, projected_annual_salary <dbl>,
#   q1_payments <dbl>, q2_payments <dbl>, q3_payments <dbl>, q4_payments <dbl>,
#   payments_over_base_pay <dbl>, percent_over_base_pay <dbl>,
#   total_payments <dbl>, base_pay <dbl>, permanent_bonus_pay <dbl>,
#   longevity_bonus_pay <dbl>, temporary_bonus_pay <dbl>, lump_sum_pay <dbl>, …

Q. How many observations are available between 2013 and 2015 including 2013 and 2015?

lapd %>% 
  filter(year >= 2013 & year <= 2015) %>% 
  nrow()
[1] 40227

Q. How many LAPD staff were employed full time in 2018?

lapd %>% 
  filter(employment_type == "Full Time" & year == 2018) %>% 
  nrow()
[1] 14664

We have done all sorts of selections, slicing, filtering on lapd but it has not changed at all. Why do you think so?

glimpse(lapd)
Rows: 68,564
Columns: 35
$ row_id                     <chr> "3-1000027830ctFu", "3-1000155488ctFu", "3-…
$ year                       <dbl> 2013, 2013, 2013, 2013, 2013, 2013, 2013, 2…
$ department_title           <chr> "Police (LAPD)", "Police (LAPD)", "Police (…
$ payroll_department         <dbl> 4301, 4302, 4301, 4301, 4302, 4302, 4301, 4…
$ record_number              <dbl> 1000027830, 1000155488, 1000194958, 1000232…
$ job_class_title            <chr> "Police Detective II", "Clerk Typist", "Pol…
$ employment_type            <chr> "Full Time", "Full Time", "Full Time", "Ful…
$ hourly_or_event_rate       <dbl> 53.16, 23.77, 60.80, 60.98, 45.06, 34.42, 4…
$ projected_annual_salary    <dbl> 110998.08, 49623.67, 126950.40, 127326.24, …
$ q1_payments                <dbl> 24931.20, 11343.96, 24184.00, 29391.20, 208…
$ q2_payments                <dbl> 29181.61, 13212.37, 28327.20, 36591.20, 241…
$ q3_payments                <dbl> 26545.80, 11508.36, 28744.20, 32904.81, 215…
$ q4_payments                <dbl> 29605.30, 13442.53, 33224.88, 37234.03, 252…
$ payments_over_base_pay     <dbl> 4499.12, 1844.82, 13192.43, 18034.53, 1376.…
$ percent_over_base_pay      <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
$ total_payments             <dbl> 110263.91, 49507.22, 114480.28, 136121.24, …
$ base_pay                   <dbl> 105764.79, 47662.40, 101287.85, 118086.71, …
$ permanent_bonus_pay        <dbl> 3174.12, 0.00, 7363.95, 7086.67, 0.00, 0.00…
$ longevity_bonus_pay        <dbl> 0.00, 1310.82, 0.00, 0.00, 1251.19, 1726.16…
$ temporary_bonus_pay        <dbl> 1325.00, 0.00, 1205.00, 1325.00, 125.00, 68…
$ lump_sum_pay               <dbl> 0.00, 0.00, 2133.18, 0.00, 2068.80, 0.00, 0…
$ overtime_pay               <dbl> 0.00, 0.00, 4424.32, 9839.33, 0.00, 0.00, 4…
$ other_pay_adjustments      <dbl> 0.00, 534.00, -1934.02, -216.47, -2068.80, …
$ other_pay_payroll_explorer <dbl> 4499.12, 1844.82, 8768.11, 8195.20, 1376.19…
$ mou                        <chr> "24", "3", "24", "24", "12", "3", "24", "24…
$ mou_title                  <chr> "POLICE OFFICERS UNIT", "CLERICAL UNIT", "P…
$ fms_department             <dbl> 70, 70, 70, 70, 70, 70, 70, 70, 70, 70, 70,…
$ job_class                  <chr> "2223", "1358", "2227", "2232", "1839", "22…
$ pay_grade                  <chr> "2", "0", "1", "1", "0", "2", "3", "1", "B"…
$ average_health_cost        <dbl> 11651.40, 10710.24, 11651.40, 11651.40, 107…
$ average_dental_cost        <dbl> 898.08, 405.24, 898.08, 898.08, 405.24, 405…
$ average_basic_life         <dbl> 191.04, 11.40, 191.04, 191.04, 11.40, 11.40…
$ average_benefit_cost       <dbl> 12740.52, 11126.88, 12740.52, 12740.52, 111…
$ benefits_plan              <chr> "Police", "City", "Police", "Police", "City…
$ job_class_link             <chr> "http://per.lacity.org/perspecs/2223.pdf", …

Moving forward we are only going to focus on year 2018, and use job_class_title, employment_type, and base_pay. Let’s clean our data accordingly and move on with the smaller lapd data that we need.

lapd %>% 
  filter(year == 2018) %>% 
  select(job_class_title, 
         employment_type, 
         base_pay)
# A tibble: 14,824 × 3
   job_class_title                  employment_type base_pay
   <chr>                            <chr>              <dbl>
 1 Police Detective II              Full Time        119322.
 2 Police Sergeant I                Full Time        113271.
 3 Police Lieutenant II             Full Time        148116 
 4 Police Service Representative II Full Time         78677.
 5 Police Officer III               Full Time        109374.
 6 Police Officer II                Full Time         95002.
 7 Police Officer II                Full Time         95379.
 8 Police Officer II                Full Time         95388.
 9 Equipment Mechanic               Full Time         80496 
10 Detention Officer                Full Time         69640 
# ℹ 14,814 more rows

lapd <- 
  lapd %>% 
  filter(year == 2018) %>% 
  select(job_class_title, 
         employment_type, 
         base_pay)

glimpse(lapd)
Rows: 14,824
Columns: 3
$ job_class_title <chr> "Police Detective II", "Police Sergeant I", "Police Li…
$ employment_type <chr> "Full Time", "Full Time", "Full Time", "Full Time", "F…
$ base_pay        <dbl> 119321.60, 113270.70, 148116.00, 78676.87, 109373.63, …

Goal:

Create a new variable called base_pay_k that represents base_pay in thousand dollars.

lapd %>% 
  mutate(base_pay_k = base_pay/1000)
# A tibble: 14,824 × 4
   job_class_title                  employment_type base_pay base_pay_k
   <chr>                            <chr>              <dbl>      <dbl>
 1 Police Detective II              Full Time        119322.      119. 
 2 Police Sergeant I                Full Time        113271.      113. 
 3 Police Lieutenant II             Full Time        148116       148. 
 4 Police Service Representative II Full Time         78677.       78.7
 5 Police Officer III               Full Time        109374.      109. 
 6 Police Officer II                Full Time         95002.       95.0
 7 Police Officer II                Full Time         95379.       95.4
 8 Police Officer II                Full Time         95388.       95.4
 9 Equipment Mechanic               Full Time         80496        80.5
10 Detention Officer                Full Time         69640        69.6
# ℹ 14,814 more rows

glimpse(lapd)
Rows: 14,824
Columns: 3
$ job_class_title <chr> "Police Detective II", "Police Sergeant I", "Police Li…
$ employment_type <chr> "Full Time", "Full Time", "Full Time", "Full Time", "F…
$ base_pay        <dbl> 119321.60, 113270.70, 148116.00, 78676.87, 109373.63, …

Goal:

Create a new variable called base_pay_level which has Less Than 0, No Income, Less than Median and Greater than 0 and Greater than Median. We will consider $62474 as the median (from previous lecture).

Let’s first check to see there is anyone earning exactly the median value.

lapd %>% 
  filter(base_pay == 62474)
# A tibble: 0 × 3
# ℹ 3 variables: job_class_title <chr>, employment_type <chr>, base_pay <dbl>

lapd %>% 
  mutate(base_pay_level = case_when(
    base_pay < 0 ~ "Less than 0", 
    base_pay == 0 ~ "No Income",
    base_pay < 62474 & base_pay > 0 ~ "Less than Median, Greater than 0",
    base_pay > 62474 ~ "Greater than Median")) 
# A tibble: 14,824 × 4
   job_class_title                  employment_type base_pay base_pay_level     
   <chr>                            <chr>              <dbl> <chr>              
 1 Police Detective II              Full Time        119322. Greater than Median
 2 Police Sergeant I                Full Time        113271. Greater than Median
 3 Police Lieutenant II             Full Time        148116  Greater than Median
 4 Police Service Representative II Full Time         78677. Greater than Median
 5 Police Officer III               Full Time        109374. Greater than Median
 6 Police Officer II                Full Time         95002. Greater than Median
 7 Police Officer II                Full Time         95379. Greater than Median
 8 Police Officer II                Full Time         95388. Greater than Median
 9 Equipment Mechanic               Full Time         80496  Greater than Median
10 Detention Officer                Full Time         69640  Greater than Median
# ℹ 14,814 more rows

We can’t really see what we have created

lapd %>% 
  mutate(base_pay_level = case_when(
    base_pay < 0 ~ "Less than 0", 
    base_pay == 0 ~ "No Income",
    base_pay < 62474 & base_pay > 0 ~ "Less than Median, Greater than 0",
    base_pay > 62474 ~ "Greater than Median")) %>% 
  select(base_pay_level)
# A tibble: 14,824 × 1
   base_pay_level     
   <chr>              
 1 Greater than Median
 2 Greater than Median
 3 Greater than Median
 4 Greater than Median
 5 Greater than Median
 6 Greater than Median
 7 Greater than Median
 8 Greater than Median
 9 Greater than Median
10 Greater than Median
# ℹ 14,814 more rows

We can use pipes with ggplot too!

lapd %>% 
  mutate(base_pay_level = case_when(
    base_pay < 0 ~ "Less than 0", 
    base_pay == 0 ~ "No Income",
    base_pay < 62474 & base_pay > 0 ~ "Less than Median, Greater than 0",
    base_pay > 62474 ~ "Greater than Median")) %>% 
  select(base_pay_level) %>% 
  ggplot(aes(x = base_pay_level)) +
  geom_bar()

glimpse(lapd)
Rows: 14,824
Columns: 3
$ job_class_title <chr> "Police Detective II", "Police Sergeant I", "Police Li…
$ employment_type <chr> "Full Time", "Full Time", "Full Time", "Full Time", "F…
$ base_pay        <dbl> 119321.60, 113270.70, 148116.00, 78676.87, 109373.63, …

Goal:

Make job_class_title and employment_type factor variables.

lapd %>% 
  mutate(employment_type = as.factor(employment_type),
         job_class_title = as.factor(job_class_title)) 
# A tibble: 14,824 × 3
   job_class_title                  employment_type base_pay
   <fct>                            <fct>              <dbl>
 1 Police Detective II              Full Time        119322.
 2 Police Sergeant I                Full Time        113271.
 3 Police Lieutenant II             Full Time        148116 
 4 Police Service Representative II Full Time         78677.
 5 Police Officer III               Full Time        109374.
 6 Police Officer II                Full Time         95002.
 7 Police Officer II                Full Time         95379.
 8 Police Officer II                Full Time         95388.
 9 Equipment Mechanic               Full Time         80496 
10 Detention Officer                Full Time         69640 
# ℹ 14,814 more rows

as.factor() - makes a vector factor
as.numeric() - makes a vector numeric
as.integer() - makes a vector integer
as.double() - makes a vector double
as.character() - makes a vector character

Once again we did not “save” anything into lapd. As we work on data cleaning it makes sense not to “save” the data frames. Once we see the final data frame we want then we can “save” (i.e. overwrite) it.

In your lecture notes, you can do all the changes in this lecture in one long set of piped code. That’s the beauty of piping!

lapd <- 
  lapd %>% 
  clean_names() %>% 
  filter(year == 2018) %>% 
  select(job_class_title, 
         employment_type, 
         base_pay) %>% 
    mutate(employment_type = as.factor(employment_type),
           job_class_title = as.factor(job_class_title),
           base_pay_level = case_when(
             base_pay < 0 ~ "Less than 0", 
             base_pay == 0 ~ "No Income",
             base_pay < 62474 & base_pay > 0 ~ "Less than Median, Greater than 0",
             base_pay > 62474 ~ "Greater than Median")) 

Word of caution

The functions clean_names(), select(), filter(), mutate() all take a data frame as the first argument. Even though we do not see it, the data frame is piped through from the previous step of code at each step. When we use these functions without the %>% we have to include the data frame explicitly.

Data frame is used as the first argument

clean_names(lapd)
# A tibble: 14,824 × 4
   job_class_title                  employment_type base_pay base_pay_level     
   <fct>                            <fct>              <dbl> <chr>              
 1 Police Detective II              Full Time        119322. Greater than Median
 2 Police Sergeant I                Full Time        113271. Greater than Median
 3 Police Lieutenant II             Full Time        148116  Greater than Median
 4 Police Service Representative II Full Time         78677. Greater than Median
 5 Police Officer III               Full Time        109374. Greater than Median
 6 Police Officer II                Full Time         95002. Greater than Median
 7 Police Officer II                Full Time         95379. Greater than Median
 8 Police Officer II                Full Time         95388. Greater than Median
 9 Equipment Mechanic               Full Time         80496  Greater than Median
10 Detention Officer                Full Time         69640  Greater than Median
# ℹ 14,814 more rows

Data frame is piped

lapd %>% 
  clean_names()
# A tibble: 14,824 × 4
   job_class_title                  employment_type base_pay base_pay_level     
   <fct>                            <fct>              <dbl> <chr>              
 1 Police Detective II              Full Time        119322. Greater than Median
 2 Police Sergeant I                Full Time        113271. Greater than Median
 3 Police Lieutenant II             Full Time        148116  Greater than Median
 4 Police Service Representative II Full Time         78677. Greater than Median
 5 Police Officer III               Full Time        109374. Greater than Median
 6 Police Officer II                Full Time         95002. Greater than Median
 7 Police Officer II                Full Time         95379. Greater than Median
 8 Police Officer II                Full Time         95388. Greater than Median
 9 Equipment Mechanic               Full Time         80496  Greater than Median
10 Detention Officer                Full Time         69640  Greater than Median
# ℹ 14,814 more rows

Data

Observations

Aggregate Data

Summaries of observations

Aggregating Categorical Data

Categorical data are summarized with counts or proportions

lapd %>% 
  count(employment_type)
# A tibble: 3 × 2
  employment_type     n
  <fct>           <int>
1 Full Time       14664
2 Part Time         132
3 Per Event          28

lapd %>% 
  count(employment_type) %>% 
  mutate(prop = n/sum(n))
# A tibble: 3 × 3
  employment_type     n    prop
  <fct>           <int>   <dbl>
1 Full Time       14664 0.989  
2 Part Time         132 0.00890
3 Per Event          28 0.00189

Mean

summarize(lapd, 
          mean_base_pay = mean(base_pay))
# A tibble: 1 × 1
  mean_base_pay
          <dbl>
1        85149.

Mean is not a good measure when the data are skewed

Median

summarize(lapd, 
          median(base_pay))
# A tibble: 1 × 1
  `median(base_pay)`
               <dbl>
1             97601.
median(lapd$base_pay)
[1] 97600.66

Quantiles

summarize(lapd, quantile(base_pay, c(0.25, 0.50, 0.75)))
# A tibble: 3 × 1
  `quantile(base_pay, c(0.25, 0.5, 0.75))`
                                     <dbl>
1                                   67266.
2                                   97601.
3                                  109368.

Recall that we use multiple functions such as mean() and median within the summarize function

summarize(lapd,
          mean_base_pay = mean(base_pay),
          med_base_pay = median(base_pay))
# A tibble: 1 × 2
  mean_base_pay med_base_pay
          <dbl>        <dbl>
1        85149.       97601.

Aggregating Data by Groups

group_by()

group_by() separates the data frame by the groups. Any action following group_by() will be completed for each group separately.

Q. What is the median salary for each employment type?

lapd %>% 
  group_by(employment_type)
# A tibble: 14,824 × 4
# Groups:   employment_type [3]
   job_class_title                  employment_type base_pay base_pay_level     
   <fct>                            <fct>              <dbl> <chr>              
 1 Police Detective II              Full Time        119322. Greater than Median
 2 Police Sergeant I                Full Time        113271. Greater than Median
 3 Police Lieutenant II             Full Time        148116  Greater than Median
 4 Police Service Representative II Full Time         78677. Greater than Median
 5 Police Officer III               Full Time        109374. Greater than Median
 6 Police Officer II                Full Time         95002. Greater than Median
 7 Police Officer II                Full Time         95379. Greater than Median
 8 Police Officer II                Full Time         95388. Greater than Median
 9 Equipment Mechanic               Full Time         80496  Greater than Median
10 Detention Officer                Full Time         69640  Greater than Median
# ℹ 14,814 more rows

lapd %>% 
  group_by(employment_type) %>% 
  summarize(med_base_pay = median(base_pay))
# A tibble: 3 × 2
  employment_type med_base_pay
  <fct>                  <dbl>
1 Full Time             97996.
2 Part Time             14474.
3 Per Event              4275 

We can also remind ourselves how many staff members there were in each group.

lapd %>% 
  group_by(employment_type) %>% 
  summarize(med_base_pay = median(base_pay),
            count = n())
# A tibble: 3 × 3
  employment_type med_base_pay count
  <fct>                  <dbl> <int>
1 Full Time             97996. 14664
2 Part Time             14474.   132
3 Per Event              4275     28

Note that n() does not take any arguments.