Valid and reliable statistics are essential to informing discussion, driving decisions, and shaping national, state and local policy. To meet the growing need for high-quality data to inform public policy, AIR supports the design, collection, analysis, and dissemination of research and statistics on significant social issues. Federal and state governments, private firms, universities, policy institutes, and nonprofit organizations across the country use AIR’s expertise in survey methodology, data acquisition, and data analysis to inform their work.
AIR's versatile team of survey, sampling, and data acquisition experts offers clients access to the latest in survey methodology, sampling methods, psychometrics, data storage and access, privacy protection, and data security expertise. We also provide deep expertise in statistical analysis and the presentation of statistical results, including proper representation of error and uncertainty in results based on sampled data.
Our experts provide clients with efficiencies and improve the data quality by guiding and supporting all aspects of survey research:
- Survey design and development
- Management of survey operations
- Statistical analysis and dissemination
- Data processing and diagnostics
- Quality assurance and control
- Analysis and reporting
In addition to our expertise in survey design and administration, our team has developed an expertise in the area of data acquisition. Data acquisition includes abstracting information from hard copy records, acquiring information from electronic records, acquiring data from files that reside on the internet or in other repositories, linking extant data files to other data repositories or to existing primary survey data. Data acquisition is growing in demand because it allows researchers to take advantage of a wide variety of existing data sources or “big data” as opposed to collecting new information via surveys. AIR’s survey methodologists and data scientists have expertise in identifying pertinent existing data sources, structuring and organizing these data specifically for analysis tasks and developing algorithms that link what would otherwise be stand-alone or independent data sources. This offering provides AIR’s clients an effective and cost efficient approach to acquiring high quality data to meet their research needs.
Identifying, assessing, and correcting for missing data is a key component of data processing and development. Failure to assess and, as needed, correct for item-level missing data, can degrade the quality of the data and limit its analytic utility. AIR’s methodological experts have extensive experience with data imputation to derive valid values for data items where the response is missing or unusable. We are experienced in designing, evaluating, and implementing imputation approaches, including logical, regression, hot deck, predictive mean matching, propensity score, and Markov Chain Monte Carlo imputation. Our experts also have a deep understanding of the software packages used to implement these methodologies.
