High-Quality Individual and Group Reports for Early Childhood Development

• In Event: 81.028 - Supporting Children's Early Learning and Kindergarten Goals With a Standards-Aligned Assessment: Effective Practices in Statewide Aggregate Reporting

Abstract

1. Objectives
This paper will describe the tools for communicating results from the DRDP (2015) assessment, an authentic formative assessment of early childhood development. We will (a) show how the reports address concerns identified in the literature about reporting of assessment results (e.g., misinterpretation, valuation) and (b) describe how the reports serve the needs and interests of multiple stakeholders, including statewide aggregate reporting.

2. Perspectives or theoretical framework
Several principles guide the report development process and address common problems with test interpretation and use, six of which we will discuss in this presentation:

1. Developmental construct modeling: The DRDP measures a set of developmental constructs. Score reports situate a child’s knowledge and skills within developmental continua aligned to be on the same measurement scale.

2. Formative assessment: The reports support a strength-based approach to interpreting scores, focusing on what the child can do in each domain. The reports provide qualitative information about the child’s next steps in development, helping teachers plan instruction and parents plan supportive activities.

3. Scientifically-based performance levels. Test development included alignment to CA standards for kindergarten readiness with child development experts. Score interpretation is based on an informative system of criterion-referenced benchmarks for expected skills (CZBs).

4. Multidimensional modeling: The results disaggregate areas of development, providing diagnostic information for formative purposes.

5. Statistical precision. Score reports present error bars around point estimates.

6. Longitudinal analysis: The group reports provide information on change (in individuals or groups) in multiple domains, all of which are on the same scale, leading to representations that permit easy comparisons for statewide aggregate reporting

3. Methods and Data Sources
Researchers used an iterative approach to create the DRDP reports. Researchers collected requests from the field, drafted visual representations of domain-level information and presented the reports to local and regional stakeholders for feedback and revision. After several cycles, the researchers integrated beta versions of the reports through online software. At present, the DRDPtech software allows stakeholders to access various summaries of child or group information in standardized formats.


4. Results
The DRDPtech software provides two types of reports: child reports and group reports.
Child reports present current and prior scores for an individual. Group reports present scores for groups such as schools or districts. The reports are dynamically generated through the web interface. The reports indicate the continuous nature of the underlying attribute while also illustrating qualitatively distinct performance levels. Parents and teachers can understand a student’s developmental profile or monitor changes in a child’s development, whereas administrators can identify groups of students with the greatest needs and can generate data for program evaluation.

5. Scholarly Significance
This project is an example of the how researchers can use constrict modelling approaches, and the BEAR Assessment System (BAS) in particular, to develop high quality assessment systems that support teachers and children, and are useful for statewide aggregate reporting. The reports present accurate statistical information from multidimensional models while remaining simple enough for easy interpretation.

Authors

• Linda Morell, University of California - Berkeley

• Rebecca Freund, University of California - Berkeley