gistfile1.txt

Bicycle data missing time or location, as percentages

year: time location
2006: 1.59 0.26
2007: 1.74 0.37
2008: 2.97 0.42
2009: 2.86 0.68
2010: 4.60 1.49

From
http://www.montrealgazette.com/news/bike-accidents/index.html
http://blogs.montrealgazette.com/2012/03/26/mapping-bicycle-collisions-in-montreal/

Data: https://docs.google.com/spreadsheet/ccc?key=0AkHTOK71aJ7qdG0zeV9xRVkyUURkR2dKLVR6bi1XWFE#gid=0

# ruby 1.9

require 'CSV'
data = CSV.read("data.tsv", {:col_sep => "\t"})

totals = Hash.new {|k,v| k[v] = 0}
missing_time = Hash.new {|k,v| k[v] = 0}
missing_location = Hash.new {|k,v| k[v] = 0}

data.each do |year, date, time, civic, street1, street2|
  totals[year] += 1
  missing_time[year] += 1 if time.nil?
  missing_location[year] += 1 if  civic.nil? & street1.nil? & street2.nil?
end

totals.keys.sort.each do |year|
  puts "%4d: %.2f %.2f" % [year, (missing_time[year].to_f / totals[year].to_f) * 100, (missing_location[year].to_f / totals[year].to_f) * 100]
end