Content area
Full Text
ABSTRACT: Momentary time sampling, partial-interval recording, and event coding are observational coding methods commonly used to examine the social and challenging behaviors of children at risk for or with developmental delays or disabilities. Yet there is limited research comparing the accuracy of and relationship between these three coding methods. By coding the low-frequency social behaviors of 100 preschool-aged children with autism spectrum disorder, the current study examines the associations among the three methodologies. Results indicated a strong relationship among all three coding methodologies. The findings provide information to researchers and practitioners considering the benefits and drawbacks of each coding methodology.
Various observational coding methods have been employed across research studies in an effort to assess children's social behavior (Odom & Ogawa, 1992). Determining the best method of capturing and examining social behaviors is of particular interest in research involving children with autism spectrum disorder (ASD). Children with ASD characteristically display low incidences of social behaviors (American Psychiatric Association, 2000; Reszka, Odom, & Hume, 2012), making direct assessment of these behaviors a challenge (Cunningham, 2012; Tsao et al., 2008). Yet these behaviors are important to capture in order to best identify ecological features of classrooms and interventions that will promote the social development of children with ASD (Reszka et al., 2012). A variety of coding methods have been used to capture lowoccurring behaviors, such as the social behaviors of young children with ASD (Gardenier, MacDonald, & Green, 2004; Reszka et al., 2012; Tsao et al., 2008); however, no study has directly compared these coding methods in this group of children to determine the utility of each.
Three commonly used coding methods include momentary time sampling (MTS), partial-interval recording (PIR), and event coding. Event coding, also called frequency counting, is often viewed as the most precise observational method for assessing the quantity of behavior because the observer records each instance of the behavior (Suen & Ary, 1989). However, due to the intensity of data collection, event coding is sometimes difficult to collect in real time and in field settings. PIR and MTS are designed to estimate the quantity of behavior that occurs and were developed as alternatives to event coding because they may be less difficult to collect. However, these discontinuous methods of observational coding only...