Once families have decided to join the study, and have completed the informed consent process using the "Social Skills Club - Parent Consent", and once we obtain letters of agreement from participating elementary schools, children with ASD will be randomly assigned to one of two intervention conditions, SKILLS or ENGAGE, led by two trained doctoral students. Since children will be randomized to either the SKILLS or the ENGAGE group, we are hoping to determine the active ingredients for successfully integrating children with autism. For both the SKILLS and the ENGAGE groups, all interventions and observations of children will take place in the target child's school. Thus, upon admission to the project, research personnel will attend the participant's classroom at school, and distribute consent forms titled "Peer Relationships at School - Classroom Parent Consent" to all children in the class. The consenting process should be no more than ten minutes. For those children that return informed consent from their parents, as well as offer assent on the "Children's Classroom Assent" to join to the project, social network measures (including brief demographic information), as well as friendship surveys (Friendships Qualities Scale; FQS), and an emotion measure (loneliness questionnaire) will be distributed at pre-intervention (baseline), post-intervention, and follow-up time points. These measures will take approximately 30-40 minutes per each time point. Graduate students will administer the measures and will make every effort to do so at a convenient time for the teacher and for the classroom to avoid utilizing instructional time. In addition, the efficacy of the treatment interventions will be assessed with ongoing direct behavioral observations at entry, twice during treatment, at exit, and at the 2-month follow-up.