Skip to main content

Table 3 Cross classification by tertiles of food outlets between Google data and DENUE, 2021

From: Agreement between a web collaborative dataset and an administrative dataset to assess the retail food environment in Mexico

Density of food outlets (counts/km2)

n

Correctly classified (%)a

Misclassified (Google data underestimates) (%)b

Misclassified (Google data overestimates) (%)c

Weighted kappa (SE)

Total census tracts

1693

70.1

14.4

15.5

0.64 (0.02)

Hermosillo, Sonora

662

67.5

16.2

16.3

0.60 (0.03)

Leon, Guanajuato

687

74.5

12.2

13.2

0.70 (0.03)

Tlalpan, Mexico City

203

71.4

14.3

14.3

0.68 (0.06)

Oaxaca de Juarez, Oaxaca

141

61.7

19.1

19.1

0.54 (0.07)

  1. DENUE National Statistical Directory of Economic Units, SE Standard Error
  2. aCorrectly classified = % of census tracts with density of food outlets of DENUE and Google data in the same tertile
  3. bMisclassified (Google data underestimates) = % of census tracts adjacently or extreme opposite tertile classified when Google data underestimates
  4. cMisclassified (Google data overestimates) = % of census tracts adjacently or extreme opposite tertile classified when Google data overestimates