search

NYCPlanning / db-acs

American Community Survey data processing for Population Fact Finder
4 stars 1 forks source link

demo - AsnEast : for a lot of demographic variables, we are seeing lots of 11.0 for MOE fields. Should we exclude them??? #16

Closed SPTKL closed 4 years ago

SPTKL commented 4 years ago

e.g.

geoid GEO_ID B02015_015M B02015_017M B02015_011M B02015_007M B02015_012M B02015_020M
2034800 1400000US36005034800 16.0 16.0 16.0 113.0 16.0 16.0
2036000 1400000US36005036000 11.0 11.0 11.0 34.0 11.0 11.0
2032600 1400000US36005032600 11.0 11.0 11.0 188.0 11.0 11.0
2031000 1400000US36005031000 11.0 11.0 11.0 50.0 11.0 11.0
2031600 1400000US36005031600 11.0 11.0 6.0 20.0 11.0 11.0
2031800 1400000US36005031800 11.0 11.0 11.0 20.0 11.0 11.0
2031200 1400000US36005031200 11.0 11.0 11.0 47.0 11.0 11.0
2031400 1400000US36005031400 11.0 11.0 11.0 27.0 11.0 11.0
2034200 1400000US36005034200 11.0 11.0 11.0 11.0 11.0 11.0
2034400 1400000US36005034400 11.0 11.0 11.0 11.0 11.0 11.0
2035000 1400000US36005035000 11.0 11.0 11.0 11.0 11.0 11.0

If the 11s are taken into calculation, then we get MOE for BX21 => 251, If not, then we get 238, which is closer to what population got (236)

SPTKL commented 4 years ago

similar case in economic table for variable Pv150t174, it seems like when calculating MOE, population ignored all M value that are equal to 16. e.g. for geoid 1000700 m_pop: 57 m_mine: 75

data: 

[{'B17024_034M': 16.0,
  'B17024_112M': 16.0,
  'B17024_021M': 16.0,
  'B17024_060M': 57.0,
  'B17024_099M': 16.0,
  'B17024_008M': 16.0,
  'B17024_073M': 16.0,
  'B17024_086M': 16.0,
  'B17024_125M': 16.0,
  'B17024_047M': 16.0}]

sum([i**2 for i in info[0].values() if i != 16])**0.5 ==> 57 sum([i**2 for i in info[0].values()])**0.5 ==> 75

EricaMaurer commented 4 years ago

if the estimate is 0, then MOE should be 0 so the variables used as inputs to this calculation should follow this rule and then the MOE should match

SPTKL commented 4 years ago

Addressed