STAA57 - WorkSheet 4

Goal: Master data transformation and summarization.

Use the dinesafe data.

1. Find the average number of distinct daily inspections.

2. Create a new day-of-week variable, call it DOW, using the wday() function in the lubridate package. Find the average number of distict daily inspections by day of week.

3. For each distinct inspection, find the time since the last inspection of the same establishment.
   (Hint: use the lag() window function )

4. Find the average number of days between inspections, grouped by the minimum number of inspections per year. Remove missing values from the average calculation, use na.rm = TRUE in the mean() function.
   (Hint: when you subtract dates, you get the number of days between them)

5. The results from the previous part suggest that Toronto Public Health actually meets its minimum inspections per year. This is the opposite conclusion to what you get by comparing the actual and expected number of inspections over the two-year period (code shown below). Explain which calculation you think is correct and why.

dinesafe  %>% 
  group_by(MINIMUM_INSPECTIONS_PERYEAR, ESTABLISHMENT_ID) %>% 
  summarise(N_INSP_ACT = n_distinct(INSPECTION_ID)) %>% 
  mutate(N_INSP_EXP = MINIMUM_INSPECTIONS_PERYEAR * 2, 
         N_INSP_DIF = N_INSP_ACT - N_INSP_EXP) %>%
  summarise(AVG_INSP_DIF = mean(N_INSP_DIF))