AISP Best PracticesResourcesPg_BestPractices

AISP commissioned 5 papers from experts in the areas of data quality, data integration, benefit cost analysis, ethics, and legal issues. These papers aim to further discern best practices when it comes to building, growing, using, and sustaining integrated data systems.  AISP coordinates a professional network of mature integrated data system (IDS) sites across the United States, regularly convening representatives from each to discover and explore common challenges and solutions in IDS work. Through these meetings, the network experts identified five important areas of best practice with respect to an IDS, namely legal considerations, ethical considerations, database infrastructure, data quality, and reflecting social benefits and costs through analyses. The network sought out authorities in each of these areas to write on key considerations and outline ‘best practices’ as they applied to the field of IDS.  Undoubtedly, the field of IDS will continue to grow and develop. These papers provide a guide and foundation.

Benefit cost analysis (BCA) is an accounting framework used to evaluate the financial consequences of decisions. Its primary objective is to improve public welfare, and often the data used to perform BCA cannot be easily quantified since they are non-market goods. BCA of administrative data can be used to determine whether social policy initiatives warrant continued investment or if they are wasteful. Currently, BCA on social policy is lacking; however, the use of integrated data systems for BCA has the potential to change this, as integrated data provide a rich source of administrative data on social programs that can be used relatively cheaply to perform BCA, which would result in more efficient policy-making.

Best Practice Paper“A Primer for Understanding Benefit-Cost Analysis”

All financial analyses entail a tradeoff between the cost of the analysis and the value of information gained. A comprehensive BCA can be resource intensive, especially for social policy initiatives, where benefits (and costs) typically accrue in non-market arenas. In the social services arena, collection and generation of data can be time-consuming and expensive. This serves as a barrier to broader implementation of BCA, as proper data collection can make BCA appear cost-prohibitive. Where available, integrated data systems can reduce the funding needed to conduct benefit-cost analysis and discover worthwhile projects that otherwise might not have been found.

This paper demonstrates how data integration impacts BCA in three ways:

  1. BCA Usage
  2. Presentation of Basic Methodological Framework
  3. Provide a simplified example of a BCA that demonstrates relevant concepts and speaks to the challenges faced in seeking to expand BCA usage.

There are no shortage of data at government agencies. The issue is that these data are typically housed at the agency responsible for collecting it, and these agencies don’t automatically permit their data to ‘talk’ to each other. For example, in a given city, foster care, juvenile justice, and homeless services regularly collect data on their clients, but these data aren’t integrated. So, while the same clients may engage in multiple services, there is no way for the agency to know this. However, there are a variety of methods available to integrate these data to obtain a more complete understanding of services outcomes.

Best Practice Paper: “An Overview of Architectures and Techniques for Integrated Data Systems Implementation”

This paper begins with a discussion of potential applications of linked administrative data in policy-level and case-level decision-making. It then presents primary data integration approaches and options that are available to health and human services enterprises based on today’s technologies and know-how. The paper also addresses the data architecture options and business process implications of embarking on a data integration program.

The value of an integrated data system is directly proportional to the quality of data integrated. Established systems’ audit records and collaborate with partners to ensure quality data. Because the collection of administrative data is not originally done with research in mind, issues regarding data quality arise. This can be offset through data linkage.

Best Practice Paper – “Administrative Record Quality and Integrated Data Systems”

*suggested citation: Boruch, Robert F. (2011). “Administrative Record Quality and Integrated Data Systems”. Actionable Intelligence for Social Policy (AISP), University of Pennsylvania.

Furthermore, given the demand for further information on IDS and data quality, we commissioned Aileen Rothbard to write an additional paper addressing this topic:

Quality Issues in the Use of Administrative Data Records

*suggested citation: Rothbard, Aileen (2013). Quality Issues in the Use of Administrative Data RecordsActionable Intelligence for Social Policy, University of Pennsylvania.

The ever-growing digital world allows for quicker analysis and understanding of data. With this comes a variety of ethical issues and considerations associated with the maintenance, integration, and use of administrative data for research purposes. Before conducting research or even establishing an integrated data system, the following four principles must be considered:

  • Security of the data
  • Confidentiality of information contained in the data
  • Permission to use data for research
  • Appropriate/ethical use of the data by the researchers

Best Practice Paper: “Ethical Use of Administrative Data for Research Purposes”

The goal of the paper is to provide an overview of the ethical issues and considerations associated with the maintenance, integration, and use of administrative data for research purposes. The paper is intended as a guide for data custodians (a.k.a., data owners, data stewards) as well as for other individuals who may be granted permission to use these data for research purposes (i.e., data users/researchers).

Large, electronic datasets are vital tool in social science research. They can contain information from a variety of sources such as, healthcare records, criminal justice and juvenile justice records, education records, child welfare records, or judicial records. These datasets do raise concerns around individual privacy and confidentiality issues as they contain sensitive information about individuals, and care must be taken to ensure that all privacy laws are followed.

Best Practice Paper“Legal Issues in the Use of Electronic Data Systems for Social Science Research”

This paper provides an overview of legal issues in using and linking large datasets for social science research. The paper is based on three assumptions. First, linked datasets are essential in conducting services research and policy analyses. Second, it is usually legally possible to collect information and create and link data, though the legal rules for different categories of information may vary. Third, while privacy and confidentiality laws are critical in thinking about these issues, the legal rules governing the security of data are as important.