Increasingly, educational opportunity programs are asked to incorporate strong theory and empirically-grounded activities into their service delivery models. Additionally, programs are tasked with demonstrating the effectiveness of their services. To aid clients in these endeavors we offer a variety of research services. The RED department has more than 10 years of experience working with programs to select evidenced-based practices and conduct in-depth research. We are well versed in a variety of research methods and offer quantitative, qualitative, and mixed methods studies. All team members have at least master’s-level training in research methods, and are certified in Group Design Standards by the Institute of Education Sciences’ What Works Clearinghouse (IES WWC). We take multiple steps to ensure that all data is handled with care, and confidentiality is maintained throughout the research process. We adhere to strict ethical standards of practice and use the Federal Educational Rights and Privacy Act (FERPA) and Human Subjects Protection protocols to guide our work. All research requests will be reviewed by the University of Kansas’ Institutional Review Board for approval.

Literature Reviews

A literature review is a systemic investigation of the empirical research literature. Through this process current knowledge on substantive findings related to service implementation and effectiveness can be gleaned. Empirically-grounded guidance on effective methodologies and theoretical frameworks can also be extrapolated through this process. The information produced from a literature review can inform clients’ program service model, development of a best practice field guide, or serve as pertinent information in the needs and plan of operation sections of a grant proposal.

Survey Design

Valid and reliable tools are essential for assessing program delivery and impact. Surveys are a powerful tool used to collect meaningful feedback from participants. These instruments can be used to assess change in participants’ knowledge, attitudes, behaviors, and perceptions. To ensure survey results are valid and reliable we follow rigorous design standards and employ a variety of development techniques, such as pilot testing, cognitive interviewing, and expert review.

The RED team has designed several tools specifically for educational opportunity programs, including a college knowledge survey for students and parents, a financial literacy assessment for students, a summer melt and postsecondary follow-up survey. We offer customized survey development to fit your program’s needs, as well as guidance on selecting pre-existing measures with sound psychometric properties. Surveys can be formatted as a paper/pencil or online assessment using Qualtrics.

Predictive Modeling

Predictive modeling is a multivariate statistical technique that can provide an understanding of the utility of key program components in predicting future desired outcomes. Using historical data collected by educational organizations (e.g., academic records, program participation records), staff from the RED department can build statistical models to examine the likelihood that a desired outcome will occur (e.g., enrollment in a postsecondary institutions). The information generated from these models can help program staff determine the optimal allocation of the program’s time and resources.

Quasi-experimental Research

When random assignment is not an option for your program, quasi-experimental designs (QED) offer a robust research method that can be used to test effectiveness. When participants are not randomly assigned it is likely that the two resulting groups will be different. To reduce the impact of these differences, and create statistically equivalent groups several QED methodologies can be applied (i.e., nonequivalent group designs, regression discontinuity design, propensity score matching). The RED department specializes in propensity score matching techniques. Through this technique a unidimensional index is created and equivalent groups are formed based on a set of pre-identified match variables (e.g., baseline performance, demographics).