Skip to main content

Table 2 Similarity between contact matrices

From: Can co-location be used as a proxy for face-to-face contacts?

  InVS13 InVS15 LH10 LyonSchool Thiers13
Co-presence 0.790 0.710 0.968 0.706 0.681
Sampling 1 0.946 0.829 0.960 0.845 0.857
Sampling 2 0.958 0.901 0.894 0.945 0.937
Sampling 3 0.888 0.816 0.958 0.738 0.691
  1. For each data set we compute the cosine similarity between the average daily contact matrix and the co-presence matrix, as well as for the contact matrices obtained for each sampling method of the co-presence data, averaged over 100 realisations for each sampling method. To compute the cosine similarities, each matrix is first transformed into a vector by concatenating its rows.