The Early Deception Survey (EDS): Its psychometric properties in children aged 10–47 months

We developed the Early Deception Survey (EDS) to create an early deception taxonomy and measure. Study 1, which was exploratory, found N = 130 parents reported children engaged in 16 deception types before 47 months, with the earliest report at 8 months. Deception was frequent, typically produced within 1 day of parents completing the survey, and understood within 1 day. Parents’ deceptions towards children positively correlated with children’s deception understanding; and parents’ deception encouragement positively correlated with children’s deception production and understanding (although most parents did not report encouraging deception). Studies 2 ( N = 167) and 3 ( N = 382) found the 16-item EDS was unidimensional with good internal reliability for 10- to 47-month-olds. While Study 4 ( N = 85) found the EDS was unrelated to deception lab tasks, Study 5 found convergent validity ( N = 610), but not predictive ( N = 203) validity with the Early Social Cognition Inventory, and good longitudinal stability ( N = 203). While parent agreement ( N = 28) was strong, parent-Early Years Educator agreement ( N = 10) was poor. Furthermore, based on our sample, 25 % of children were predicted to engage in at least one deception type by 10 months, 50 % by 16 months, 75 % by 24 months; and 97.5 % by 38 months. We found only one demographic difference in how parents answered individual items, and found less educated and younger parents reported higher EDS scores. • -We developed the Early Deception Survey (EDS) for 10–47 months. • -The EDS has good internal reliability. • -The EDS has good longitudinal stability. • -The EDS has good convergent validity with the Early Social Cognitive Inventory. • -The EDS has good inter-observer agreement between parents.