The two datasets are the ones used for our statistical analyses that allowed the conclusions of our article:
Charpentier et al. "Primate mothers promote proximity between their offspring and infants who look like them".

For more details, please contact me:
marie.charpentier@umontpellier.fr

-------------

The first file ("facial distances") contains information about the facial distances retrieved from pairs of mandrills (Mandrillus sphinx) 
involving 80 study infants along with their sirehood and motherhood.

The table has 8 columns and 2556 rows. There is no missing data otherwise indicated in the main text.

-the first two columns provide the names of the two infants of each study pair.
-the 3rd column provides their facial distance (ie, the higher the value the lower the resemblance between the two infants), averaged over several pictures (the total number of distances considered for the two infants of each pair is provided in column 8).
This is the dependent variable of the associated model.
-the 4th column indicates whether the two infants were sired by the same father ("samedad"), two different fathers ("diffdad"), conceived under the same alpha male's tenure ("samealpha") or under different alpha male's tenures not ("diffalpha").
-The 5th column provides the sexes of each pair of infants (3 modalities: "ff", "fm" or "mm").
-The 6th column indicates whether the two infants shared the same mother ("same") or not ("diff").
-The 7th column gives the average (absolute) distance in age between the two infants for each pair of pictures compared (rounded up to the next integer).
-The 8th column indicates the total number of faciale distances considered for each pair of infants.

-------------

The second file contains informations about spatial associations among three types of dyads and the associated studied variables.

The table has 11 columns and 1187 rows. There is no missing data otherwise indicated in the main text.

-the first two columns provide the names of the two individuals of each study pair (for which a spatial association has been retrieved).
-the 3rd column indicates which type of dyad the two individuals belong to, with three classes ("inf-mom", "mom-mom", "inf-inf"). 
Note that this variable allows to split the data set in three tables that fed three different statistical models (eg, one model was performed on "inf-mom" rows only; one model was done on "inf-inf" rows only, and one model was done on "mom-mom" rows only).
-the 4th column gives the total number of scans performed on both individuals of each pair. This variable (once log-transformed) is used as an offset in our models.
-the 5th column gives the number of scans for which both individuals of each pair were found in proximity (<= 5m). This is the dependent variable in our models.
-the 6th column gives the (standardized) residuals of the average facial distance retrieved for each infant-infant pair (eg, when the type is "mom-mom", the two infants are their own offspring).
-the 7th column indicates the (absolute, standardized) difference in age of both infants.
-the 8th column provides the (absolute, standardized) difference in mothers' rank. Here, the two mothers of the two infants are considered.
-the 9th column indicates the (standardized) relatedness between both mothers (the two mothers of the two infants are considered).
-the 10th column provides the sexes of each pair of infants (3 modalities: "FF", "MF" or "MM").
-the 11th column indicates whether the two infants belong ("o") or not ("n") to the same matriline.

-------------