One of Tufte's most enduring ideas is that of Small Multiples, which he describes as frames of a movie. It's creating a series of graphics that show the same thing, but iterated over a value.
A few examples:
Think of it like this: What if we made a chart for each of the 5 largest crimes on campus, showing a count by month for each over 5 years?
In [1]:
library(lubridate)
library(dplyr)
library(ggplot2)
unlcrime <- read.csv("../../Data/unlcrime.csv", na.string = "NA")
head(unlcrime)
Attaching package: ‘lubridate’
The following object is masked from ‘package:base’:
date
Attaching package: ‘dplyr’
The following objects are masked from ‘package:lubridate’:
intersect, setdiff, union
The following objects are masked from ‘package:stats’:
filter, lag
The following objects are masked from ‘package:base’:
intersect, setdiff, setequal, union
CaseNo Incident.Code Reported Case.Status Start.Occurred End.Occurred Building Location Stolen Damaged Description 11000001 NARCOTICS - POSSESSION 2011-01-01T00:10:00 Cleared by Arrest 2011-01-01T00:10:00 515 N 19th Parking Lot Vine to 515 North 19th St. (GOV'T PARKING LOT (UNL)) NA NA UNL student cited and released for Poss of Marijuana < 1 oz and Poss of Drug Paraphernalia after a CSO reported suspicious drug-related activity from a vehicle. 1.2 g of marijuana seized. 11000011 NARCOTICS - POSSESSION 2011-01-01T23:16:00 Cleared by Arrest 2011-01-01T23:16:00 N 10TH ST & CHARLESTON ST (STREET) NA NA Traffic Stop. Vehicle stopped for a headlight out. Upon contact Officer smelled the strong odor of burnt marijuana. Driver cited/released for Possession of Drug Paraphernalia after a search yielded a pipe with burnt marijuana residue. 11000013 ALCOHOL - MINOR IN POSSESSION 2011-01-02T00:38:00 Cleared by Arrest 2011-01-02T00:38:00 N 17TH ST & N ANTELOPE VALLEY PKWY (STREET) NA NA Three persons cited for MIPC after being stopped for no headlights at N. Antelope valley parkway and N. 17rh Street. 11000019 TRAFFIC - SUSPENDED DRIVER 2011-01-02T23:24:00 Cleared by Arrest 2011-01-02T23:24:00 N 17TH ST & VINE ST (STREET) NA NA Non-student cited and lodged in jail for Driving Under Suspension after being stopped for speeding. 11000020 NARCOTICS - POSSESSION 2011-01-03T04:46:00 Cleared by Arrest 2011-01-03T04:46:00 N 10TH ST & CHARLESTON ST (STREET) NA NA Non-student stopped for speeding. Officer smelled burnt marijuana from within the vehicle and upon searching located burnt marijuana cigarettes, a pipe, and .6 g of marijuana. Cited and released for Speeding, Poss of Marijuana < 1 oz, and Poss of Drug Paraphernalia. 11000022 MEDICAL EMERGENCY 2011-01-03T10:34:00 Inactive 2011-01-03T10:34:00 Campus Recreation Center 841 North 14th St (SPORT FACILITIES) NA NA UNL affiliated female disoriented in need of medical attention at the Rec Center. Transported to the hospital by LFR.
Let's first figure out what the top five UNLPD calls are.
In [23]:
crimes <- unlcrime %>% group_by(Incident.Code) %>%
summarize(
count = n()
) %>%
arrange(desc(count)) %>%
top_n(5)
head(crimes)
Selecting by count
Incident.Code count ALCOHOL - MINOR IN POSSESSION 1138 MEDICAL EMERGENCY 951 TRAFFIC - SUSPENDED DRIVER 945 NARCOTICS - POSSESSION 877 ALCOHOL - DWI 651
Now we can filter those out to create our data.
In [4]:
crimesbymonth <- unlcrime %>%
filter(Incident.Code == "ALCOHOL - MINOR IN POSSESSION" | Incident.Code == "MEDICAL EMERGENCY" | Incident.Code == "TRAFFIC - SUSPENDED DRIVER" | Incident.Code == "NARCOTICS - POSSESSION" | Incident.Code == "ALCOHOL - DWI") %>%
mutate(
shortdate = format(as.POSIXct(Reported, format="%Y-%m-%d")),
month = floor_date(ymd(shortdate), "month"),
) %>%
group_by(Incident.Code, month) %>%
summarize(
count = n()
)
The new bits we're going to learn today is the facet_wrap where we tell ggplot, essentially, make a new chart for each one of these things. Note: Your facet cannot be a date. To use years or months, you'll have to create one in your data.
In [10]:
ggplot(crimesbymonth, aes(x=month, y=count, group=1)) +
geom_line() + labs(x="Month", y="Count") + facet_wrap(~Incident.Code)
What else could you do with this? Look at the previous data we've worked with. What could you do with your existing data and existing charts? What could be improved through small multiples? Examples: The prison data, the majors by race and sex data, the parking tickets data.
In [11]:
tickets <- read.csv("../../Data/tickets.csv", stringsAsFactors = FALSE)
In [12]:
head(tickets)
Citation Date Location Violation 15078429 2012-04-02 07:15:00 North Stadium Expired Meter 24048318 2012-04-02 07:22:00 Housing No Valid Permit Displayed 24048320 2012-04-02 07:26:00 14th & W Street No Valid Permit Displayed 15078430 2012-04-02 07:36:00 Champions Club Parking in Unauthorized Area 18074937 2012-04-02 07:39:00 Sandoz Expired Meter 18074938 2012-04-02 07:40:00 Sandoz Expired Meter
In [21]:
ticketscount <- tickets %>% mutate(
CleanDate = ymd_hms(Date),
month = floor_date(CleanDate, "month"),
) %>%
group_by(Location, month) %>%
summarize(
count = n()
) %>%
mutate(
year = format(as.Date(month, format="%Y/%m/%d"),"%Y")
)
In [22]:
ggplot(ticketscount, aes(x=month, y=count, group=1)) +
geom_line() + labs(x="Month", y="Count") + facet_wrap(~Location)
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
geom_path: Each group consists of only one observation. Do you need to adjust
the group aesthetic?
In [ ]:
Content source: mattwaite/JOUR491-Data-Visualization
Similar notebooks: