Data from: The importance of partner inclusion criteria for understanding drivers of social variation among individuals: Data from blue monkeys

# Data from: The importance of partner inclusion criteria for understanding drivers of social variation among individuals: Data from blue monkeys [https://doi.org/10.5061/dryad.b2rbnzsns](https://doi.org/10.5061/dryad.b2rbnzsns) ## Description of the data and file structure There are three data files in this dataset. Each file summarizes the social ties and sociodemographic, demographic, and individual characteristics of adult female blue monkeys (*Cercopithecus mitis stuhlmanni*) living in multiple groups in the Kakamega Forest, western Kenya, collected over 13 years through focal follow behavioral observations (see Methods in accompanying manuscript). Two individuals are considered to have a social tie (and thus be social partners) if they lived in the same group for at least three-quarters of a year, and were observed to interact with friendly social contact (i.e., grooming each other in either direction or sitting in contact). The strength of ties is defined as the proportion of dyadic observation time (i.e., the proportion of beeps when *either* of the two individuals was under focal observation) that the two individuals were observed in friendly social contact. In each file, a row corresponds to data for one female in one study year. Each of the three files has the same columns. The three files were prepared by considering social interactions with different sets of social partners. For each of the three files, a different set of partner selection criteria was applied, and the potential partners who met those criteria are the “eligible partners”. The data files contain raw values; in our analysis, we standardized (Z-score) values in the columns **observation_time**, **relatives**, **age**, and **infant_days**. The field **available_partners** was increased by 1e-6 (to avoid negative infinities) and then log-transformed before being standardized. See the associated paper for more details on the collection and preparation of these data. The three files are named as follows: 1. annual_social_contact_all_partners – In this file, ties to all partners are included. 2. annual_social_contact_peers_only – In this file, only ties to peers (other adult females) are included. 3. annual_social_contact_non_peers_only – In this file, only ties to non-peers (non-adults or males) are included. In each file, the columns are as follows: **subj** – A unique code for each individual adult female **year** – The “study year”, or year of observations summarized in this row. Study years started on September 1 of the preceding calendar year, and ran through August 31 of the corresponding calendar year. For example, study year 2018 started on September 1, 2017, and continued through August 31, 2018. **group** – A unique code identifying the group that the subject belonged to in the specified study year. **observation_time** – The total number of minutes of focal follow observation for each subject/study year combination. **available_partners** – The number of eligible social partners available to the subject in the specified group and study year **relatives** – The number of close relatives present in the group and available as social partners for the subject. Close relatives are defined as those whose matrilineal coefficient of relationship to the subject is at least 1/16 (0.0625). **age** – The subject’s age in years at the end of the study year. **dominance_scaled** – The subject’s dominance position in the given study year, scaled to group size. Ranks were computed with DomiCalc, and are expressed as a value between 0 and 1 (lowest to highest rank). Represents the proportion of peers who are lower-ranked than the subject. **infant_days** – The proportion of days in the study year that the subject had her own infant (<1 year old) present with her in the group. **number_of_ties** – The total number of partners with whom the subject was observed to interact (grooming and sitting in contact) in the specified group and study year. **strong_ties** – The number of partners with whom the subject had a “strong tie” in the specified study year. A strong tie is defined to be a tie with a strength that is more than 1 standard deviation above the mean for a given group in a given study year. **weak_ties** – The number of partners with whom the subject had a “weak tie” in the specified study year. A weak tie is defined to be a tie that is not a strong tie, i.e., the strength of the tie is not at least 1 standard deviation above the group-study year mean. **total_tie_strength** – The sum of the strengths of all ties the subject had in the specified study year. **evenness** – A measure of how evenly the subject’s social activity was distributed among all eligible partners in a given group and study year. Calculated as the Shannon diversity index taken across all tie strengths (including 0-strength ties), divided by the maximum possible diversity index, which is the natural log of the number of eligible partners.