Dynamic Assessment of Learning Potential: Inter-Rater Reliability of Deficient Cognitive Functions, Types of Mediation, and Non-Intellective Factors

The objective of this study was to investigate the reliability of three major domains of individual dynamic assessment (DA): (a) deficient cognitive functions (DCF), (b) types of mediation given during DA, and (c) non-intellective factors. A sample of 35 young adolescents was administered eight tests from the Learning Potential Assessment Device (LPAD) (Feuerstein, Rand, & Hoffman, 1979). The sample was composed of children diagnosed with learning disabilities and educable mental handicaps, and normally achieving children. The DA procedure for each case was videotaped for 8 to 15 hours and later rated for the three main domains. Results in general showed moderate reliability scores for DCF and mediational strategies and lower reliability scores for the non-intellective factors (NIF). Separate analyses were carried out for ratings which include a 0 category (examiners could not observe a behavior) and ratings without a 0 category. The results showed a general tendency for higher agreement among raters when the 0 category was removed. In type of mediation, ratings were similar with or without the 0 rating only in the training phase, when agreement was higher in approximately 10% of categories when 0 ratings were included than when not. These results were explained by referring to the interaction of type of task and phase of testing (situation) interaction.