Selection of student subjects
All eight Oregon Regional Programs for students with autism spectrum disorder were invited to nominate students. Parents were sent letters by their child's early intervention program asking them to voluntarily include their child in the study. In 1998, sixty-seven parents returned permission forms and all of their children were included in the study. In 2001, parents of newly identified students with autism spectrum disorder, between the ages of 2 and 4, were sent letters by their early intervention programs asking them to voluntarily include their children in the study. Fifty-six new students were accepted during the 2001/2002 school year, increasing the number of participants to 123 students.

Training of the assessment team
Prior to the first baseline visits, assessment specialists were given extensive training by special education professionals on how to complete assessment visits. They were taught how to complete file assessments, to obtain and record scores for previously administered Battelle's and Vineland's, and how to interview the teacher and help them complete the student learning profile and the program implementation checklist.

In addition to learning how to complete file reviews and interview the teachers to obtain necessary information, the new assessment specialists were taught how to score the subtests of the ASIEP-2 assessment by two of the test's developers. They watched training videos and then practiced scoring by using prior videotapes of students who were given the ASIEP-2 subtests. They were then accompanied by previously trained special education professionals during all the baseline assessments.

Baseline assessments visits
Baseline assessment visits were scheduled for each subject participating in the study. A team of assessment specialists arrived at each site and completed file reviews. Information obtained from student files included the subject's date of birth, autism diagnosis date, and scores from previously administered Battelle Developmental Inventory and Vineland Adaptive Behavior Scales. They then assisted the teacher in completing the Student Learning Profile and Program Implementation Checklist. The assessment specialists also collected the ASIEP-2 Autism Behavior Checklist from the teacher. It had been sent ahead of time with instructions of how to fill it out, so it could be completed by the teacher prior to the visit.

After conducting the file review and meeting with the teacher to collect the Autism Behavior Checklist, Student Learning Profile, and Program Implementation Checklist, the assessment specialists administered the ASIEP-2 Sample of Vocal Behavior, Educational Assessment, and Interaction Assessment subtests to the student. Each subtest took between 10 to 30 minutes to complete. Specific administration instructions describe in the ASIEP-2 manual regarding materials, stimuli, procedures, room set up, instructions for adult present and scorers were followed.

To ensure the accuracy of the scoring, the Sample of Vocal Behavior and Interaction Assessment subtests were videotaped and then watched and re-scored by the assessment specialists back in their offices. All numbers on the assessments were then rechecked by another specialist before the data was inputted into the computer database. In addition, videotapes were randomly selected and scored by independent raters as an additional reliability check.

Quarterly assessments administered by the assessment team.
Each quarter (Fall, Winter, and Spring of each school year) assessment visits were scheduled with the student's teacher or specialist. The ASIEP-2 Interaction Assessment, Sample of Vocal Behavior, and Educational Assessment Subtests were administered at each quarterly assessment by trained assessment specialists. Procedures for the ASIEP-2 established during the baseline assessment were followed for each quarterly visit.

As the study progressed, 14 students reached the scoring ceilings on the ASIEP-2 Sample of Vocal Behavior and 11 students reached the scoring ceiling on the Educational Assessment Subtests. Two alternative assessments, the Expressive One Word and portions of the State of Oregon Basic Academic Skills Assessment System, were instead administered to these students at each quarterly assessment.

Annual assessments completed by the assessment team.
A trained assessment team member with a Ph.D. in Special Education, administered the Battelle Developmental Inventory (cognitive area) to each student once each school year. The assessment took place in the winter of each school year.

Annual assessments or other data gathered from the teacher.
Once a year, an ASIEP-2 Autism Behavior Checklist and a Vineland Adaptive Rating Scale (survey form) were completed by the teacher. Both of these assessments were completed approximately one year from the date of the last assessments completed the school year before.

In addition, toward the beginning of each school year, the teacher was asked to complete the program implementation checklist describing the total hours per week each subject received services, what type of services/teaching they received ( e.g., in a group or pull out one-to-one), what type of one-to-one pull out teaching they received (e.g., pivotal response training, discrete trial), and who provided the services (e.g., teacher, administrative assistant, related services staff). At the end of the school year, the information was again reviewed and verified with the teacher by an assessment specialist.

Annual assessments or other data gathered from the parents.
At the end of each school year, parents were sent surveys which gave them an opportunity to provide input on their child's communication skills, social interaction skills, and behavior. In addition, they could rate and comment on their child's program, report on any special diet they were using with their child, and describe any additional services they were paying for with their own family funds.

Annual observation data provided by consultants.
In the spring of each school year, special education professionals with expertise in behavioral instructional strategies and extensive knowledge in designing and setting-up early education and school-age programs for children with autism, visited each classroom to observe and give individual consulting advice to teachers of students in the study. The classroom observation form was completed during the visit and then a comment and suggestion sheet was given to the teacher at the end of the visit.

Behavioral instructional strategies training for teachers
To ensure fidelity of implementation, trainings in extensive behavioral instructional strategies for teaching children with autism were conducted each school year. Each participating region was given at least two trainings per year in pivotal response, discrete trial, and functional routines. All teachers and staff who taught students in the study were invited and encouraged to attend all trainings.