A descriptive summary of the studies used in the meta-analysis.

Note: If a range of N was given, two numbers are presented, the smallest number of dogs used from one estimate and the largest number of dogs from another estimate. If authors concluded in their study that they had evidence for consistency for at least one trait, the study was given a ‘1’ for consistency; if authors concluded in their study that they lacked evidence for consistency of any trait, the study was given a ‘0’ for no consistency. Likewise, if authors concluded in their study that they had evidence for a lack of consistency for at least one trait, the study was given a ‘1’ for consistency; if authors concluded that there was at least one trait that was not consistent the study was given a ‘1’ for no consistency. In both cases (‘consistency’ or ‘no consistency’) a zero is given if the study did not report a consistent trait or a trait that was not consistent. A ‘1’ for Puppy test indicates that dogs were tested as puppies (<12 months of age) during the first test period or not (a ‘0’). A ‘1’ and a ‘0’ for Puppy test indicates studies that presented from results from both age categories of dogs. If a range for test interval was given, two numbers are presented, the shortest interval and the largest interval. The average study effect size was calculated by multiplying correlations for each subgroup by the inverse of its variance. The sum of these products was then divided by the sum of the inverses.