Since 1989, Search Institute has conducted numerous studies of 6th – 12th grade students in public and private schools across the United States using a survey titled Search Institute Profiles of Student Life: Attitudes and Behaviors. The Developmental Asset framework was originally configured as a 30 asset framework. In 1996, the framework was expanded to 40 Developmental Assets, based on analysis of aggregated data on 254,000 students who took the original 30-asset survey from 1989-1994, the additional synthesis of child and adolescent research, as well as dialogue with researchers and practitioners. The first large-scale dataset documenting the 40 asset framework came from over 99,000 6th – 12th graders in the 1996-1997 academic year. In 2001, Search Institute updated the dataset using surveys administered during the 1999-2000 academic school year. The main difference between the 1999-2000 and the 1996-1997 datasets was that the ’99-’00 dataset was weighted in order to adjust for over- and under-representation of youth, in particular minority youth and youth living in more urban settings.
With the financial assistance of Thrivent Financial for Lutherans, we have again updated the dataset on the 40 Developmental Assets framework. The current data, which you see described on subsequent pages, and on which all of our most recent publications will be based, come from over 148,000 6th – 12th graders in over 200 communities across the country from the 2003 calendar year. We decided to use the calendar year (vs. the academic year) for at least two reasons: 1) we wanted to maximize the N of the dataset (i.e., the actual number of surveys comprising the dataset); as well as 2) maximize the diverse nature of the dataset. Using surveys administered during the 2003 calendar year provided the best data by these criteria.
We again weighted the 2003 data to adjust for under-representation of groups of youth (minority and urban) by using the 2000 Census data for community size and for race/ethnicity. Weighting is an acceptable statistical procedure often used by researchers which adjusts the values of responses for certain groups, in this case non-Caucasian youth and urban youth. Simply, this technique corrects for over- and under-representation of certain groups. Like the previous two datasets, this current dataset was drawn from individual communities that chose to survey their own students, and is consequently not nationally representative.
Much as we would like to eyeball changes from these datasets, it is not appropriate
to compare across the three datasets for the following reasons:
