The following question is about visualisation.

The data shows calories of a selection of chocolate bars, 100g equivalents. Calories mapped to the vertical axis. If you are wanting the reader to compare the inter quantile range of calories of milk and dark chocolates, which part of the plot do you need to observe?

This question is about tidy temporal data. Below are the data on military expenditure for different countries

head(expenditure)

## # A tibble: 6 × 4
##   Entity      Code   Year military_expenditure
##   <chr>       <chr> <dbl>                <dbl>
## 1 Afghanistan AFG    1970              5373185
## 2 Afghanistan AFG    1973              6230685
## 3 Afghanistan AFG    1974              6056124
## 4 Afghanistan AFG    1975              6357396
## 5 Afghanistan AFG    1976              8108200
## 6 Afghanistan AFG    1977              8553767

To compare expenditure counts of Australia and Germany between 2015 and 2020, we are going to select the observations for those two countries and that time range. Then we want change the data format so that the expenditure for Australia and the expenditure for Germany across different years appears in two different columns.

Fill in the blanks for the following code to get our desired output:

exp2 <- expenditure %>% 
  # (a) which pivot function
  # (b) which id_cols
  # (c) which column forms names_from
  # (d) which colum forms values_from
  dplyr::---(Entity %in% c("Australia", "Germany",
                Year >= 2015) %>%
  pivot_---(id_cols = ---, 
              names_from = ---,
              values_from = ---)

1. select

   filter

   military_expenditure

   Germany
2. longer()

   wider()

   year
3. Military_expenditure

   Time

   Year
4. Country

   Entity

   Australia

The following question is about tidy data. The table below contains looks at crime occurrence in different locations across Victoria:

What proportion of the crimes were recorded in the Monash LGA?

Incorrect answers are penalised.

The following question is about tidy data. The table below contains looks at crime occurrence in different locations across Victoria:

What is the proportion of crimes that entered through the front door?

Hint: If a data set contains "NA" values, it means the entries are missing. You can ignore these missing values here.

The following question is about tidy data. The table below contains looks at crime data in different locations across Victoria:

As usually we need to first inspect the variables and observations in this data set. What is the dimension of the data set?

This question is about tidy temporal data. Below are total daily pedestrian traffic counts for the month of March for 2020 and 2019.

walk_daily_counts %>% 
  arrange(day, month)

## # A tibble: 62 × 6
##    Date       Count  year month   day wday 
##    <date>     <int> <dbl> <ord> <int> <ord>
##  1 2019-03-01 34485  2019 Mar       1 Fri  
##  2 2020-03-01 26840  2020 Mar       1 Sun  
##  3 2019-03-02 33896  2019 Mar       2 Sat  
##  4 2020-03-02 27900  2020 Mar       2 Mon  
##  5 2019-03-03 27036  2019 Mar       3 Sun  
##  6 2020-03-03 28003  2020 Mar       3 Tue  
##  7 2019-03-04 33865  2019 Mar       4 Mon  
##  8 2020-03-04 27949  2020 Mar       4 Wed  
##  9 2019-03-05 34463  2019 Mar       5 Tue  
## 10 2020-03-05 24936  2020 Mar       5 Thu  
## # … with 52 more rows

To compare daily counts of pedestrians in March for 2019 compared to 2020, we could use a scatterplot. But first we would need to pivot the data to make daily counts for each year as column.

Fill in the blanks for the following code to get our desired output:

walk_daily_counts_wide <- walk_daily_counts_wide %>% 
  # (a) which pivot function
  # (b) which id_cols
  # (c) which column forms names_from
  # (d) which column forms values_from
  pivot_---(id_cols = ---, names_from = ---, values = ---)

## # A tibble: 31 × 3
##      day `2019` `2020`
##    <int>  <int>  <int>
##  1     1  34485  26840
##  2     2  33896  27900
##  3     3  27036  28003
##  4     4  33865  27949
##  5     5  34463  24936
##  6     6  33763  33456
##  7     7  35403  30580
##  8     8  43030  27444
##  9     9  40673  25149
## 10    10  36208  26425
## # … with 21 more rows

1. pivot_longer()

   pivot_wider()
2. day

   month

   wday
3. day

   year

   Date
4. wday

   Count

   year

The following question is about tidy data. The table below contains looks at crime occurrence in different locations across Victoria:

If you would like to calculate the proportion of the different crime types by location which code do you need to use?

Hint: Missing values is typically denoted by "NA" in the dataset, we can ignore these values by passing the option "na.rm = TRUE" to the appropriate R command.

Incorrect answers will be penalised.

This question is about visualising temporal data.

The example data is on pedestrian counts in the city of Melbourne. The below plot looks at distribution of the pedestrian counts over weekdays in March across 24hrs, comparing 2019 to 2020.

ped %>% 
  ggplot(aes(x=Time, y=Count, group=Date, colour=as.factor(year))) +
    geom_boxplot() +
    facet_wrap(~ year, ncol= 1, scales = "free") + 
    scale_colour_brewer("", palette="Dark2") + 
    theme(legend.position="bottom", legend.title = element_blank())

Image failed to load

By looking at the above plots, select all statements that are TRUE.

The following question is about wrangling data. Here is the table flights from the nycflights13 package that we have wrangled previously in class.

glimpse(flights)

## Rows: 336,776
## Columns: 19
## $ year           <int> 2013, 2013, 2013, 2013, 2013, 2013, 2013,…
## $ month          <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,…
## $ day            <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,…
## $ dep_time       <int> 517, 533, 542, 544, 554, 554, 555, 557, 5…
## $ sched_dep_time <int> 515, 529, 540, 545, 600, 558, 600, 600, 6…
## $ dep_delay      <dbl> 2, 4, 2, -1, -6, -4, -5, -3, -3, -2, -2, …
## $ arr_time       <int> 830, 850, 923, 1004, 812, 740, 913, 709, …
## $ sched_arr_time <int> 819, 830, 850, 1022, 837, 728, 854, 723, …
## $ arr_delay      <dbl> 11, 20, 33, -18, -25, 12, 19, -14, -8, 8,…
## $ carrier        <chr> "UA", "UA", "AA", "B6", "DL", "UA", "B6",…
## $ flight         <int> 1545, 1714, 1141, 725, 461, 1696, 507, 57…
## $ tailnum        <chr> "N14228", "N24211", "N619AA", "N804JB", "…
## $ origin         <chr> "EWR", "LGA", "JFK", "JFK", "LGA", "EWR",…
## $ dest           <chr> "IAH", "IAH", "MIA", "BQN", "ATL", "ORD",…
## $ air_time       <dbl> 227, 227, 160, 183, 116, 150, 158, 53, 14…
## $ distance       <dbl> 1400, 1416, 1089, 1576, 762, 719, 1065, 2…
## $ hour           <dbl> 5, 5, 5, 5, 6, 5, 6, 6, 6, 6, 6, 6, 6, 6,…
## $ minute         <dbl> 15, 29, 40, 45, 0, 58, 0, 0, 0, 0, 0, 0, …
## $ time_hour      <dttm> 2013-01-01 05:00:00, 2013-01-01 05:00:00…

For the following questions, write down the verbs and columns that you would need to use to do the calculations to answer it from the flights table. We provide the code structure and a list of possible verbs for you to select from:

What is the typical daily number of flights that American Airlines flies out of LGA between 7am and 8am?

flights %>% 
  # step 1
  ___(carrier == "AA", origin == "LGA", between(hour, 7, 8)) %>%
  # step 2
  ___(year, month, day) %>% 
  summarise(n =  mean(n))

1. The verb for the first step is mutate()

   The verb for the first step is filter()
2. The verb for the second step is mutate()

   The verb for the second step is count()

   The verb for the second step is group_by()

The following question is about wrangling data. Here is the table flights from the nycflights13 package that we have wrangled previously in class.

glimpse(flights)

## Rows: 336,776
## Columns: 19
## $ year           <int> 2013, 2013, 2013, 2013, 2013, 2013, 2013,…
## $ month          <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,…
## $ day            <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,…
## $ dep_time       <int> 517, 533, 542, 544, 554, 554, 555, 557, 5…
## $ sched_dep_time <int> 515, 529, 540, 545, 600, 558, 600, 600, 6…
## $ dep_delay      <dbl> 2, 4, 2, -1, -6, -4, -5, -3, -3, -2, -2, …
## $ arr_time       <int> 830, 850, 923, 1004, 812, 740, 913, 709, …
## $ sched_arr_time <int> 819, 830, 850, 1022, 837, 728, 854, 723, …
## $ arr_delay      <dbl> 11, 20, 33, -18, -25, 12, 19, -14, -8, 8,…
## $ carrier        <chr> "UA", "UA", "AA", "B6", "DL", "UA", "B6",…
## $ flight         <int> 1545, 1714, 1141, 725, 461, 1696, 507, 57…
## $ tailnum        <chr> "N14228", "N24211", "N619AA", "N804JB", "…
## $ origin         <chr> "EWR", "LGA", "JFK", "JFK", "LGA", "EWR",…
## $ dest           <chr> "IAH", "IAH", "MIA", "BQN", "ATL", "ORD",…
## $ air_time       <dbl> 227, 227, 160, 183, 116, 150, 158, 53, 14…
## $ distance       <dbl> 1400, 1416, 1089, 1576, 762, 719, 1065, 2…
## $ hour           <dbl> 5, 5, 5, 5, 6, 5, 6, 6, 6, 6, 6, 6, 6, 6,…
## $ minute         <dbl> 15, 29, 40, 45, 0, 58, 0, 0, 0, 0, 0, 0, …
## $ time_hour      <dttm> 2013-01-01 05:00:00, 2013-01-01 05:00:00…

For the following questions, write down the verbs and columns that you would need to use to do the calculations to answer it from the flights table. We provide the code structure and a list of possible verbs for you to select from:

What hour of day should you plan to fly if you want to avoid departure delays as much as possible?

flights %>% 
  # step 1
  ___(hour) %>%
  # step 2
  ___(avg_delay = mean(___, na.rm = TRUE)) %>% 
  # step 3
  ___(avg_delay)

1. The verb for the first step is group_by()

   The verb for the first step is mutate()

   The verb for the first step is count()
2. The verb for the second step is mutate()

   The verb for the second step is summarise()
3. The appropriate column to compute the mean from is arr_delay

   The appropriate column to compute the mean from is dep_delay
4. The verb for the third step is sorty()

   The verb for the third step is arrange()