NCHS Research and Development Survey

Overview

The Research and Development Survey (RANDS) is an ongoing series of cross-sectional and longitudinal surveys conducted by the National Center for Health Statistics (NCHS). Cross-sectional surveys are studies that analyze data from many different people at a single point in time. Longitudinal surveys are studies that analyze data from the same group of people over time. NCHS has conducted 8 rounds of RANDS surveys and 3 rounds of RANDS during COVID-19 surveys since RANDS was created in 2015.

NCHS created RANDS to explore the possibility of using survey panels operated by private companies (commercial survey panels) to collect information about the nation's health. Questions on RANDS have been used for comparisons with:

• The National Health Interview Survey
• The National Intimate Partner and Sexual Violence Survey
• The National Survey of Family Growth

RANDS topics

The specific topics in each round of RANDS vary and have covered a wide range of health behaviors and conditions. RANDS surveys have included questions about—

• Access to and use of health care
• Chronic conditions
• Disability
• Opioid use
• Physical activity
• Psychological distress
• Smoking

RANDS panels

RANDS uses commercial survey panels that recruit large groups of participants for their surveys. A smaller group is then randomly chosen from the large group to participate in a RANDS survey. This is called probability sampling.

Since 2015, NCHS has had contracts with different companies to administer RANDS using their proprietary recruited probability panels. These contractors have included Gallup and NORC at the University of Chicago (NORC).

On several occasions, RANDS has also used a nonprobability sample for research purposes. People in a nonprobability sample are selected using methods that are not random. For example, people in a nonprobability sample may have volunteered for the study or they may have been chosen for specific characteristics.

Question evaluation and methodological research

RANDS surveys also supplement NCHS’s Collaborating Center for Questionnaire Design and Evaluation Research (CCQDER) question evaluation and development studies and methodological research.

Question evaluation

Question evaluation and development studies help researchers design better questions and data users interpret the findings from those questions. RANDS surveys use quantitative methods to evaluate how likely survey questions are to prompt an incorrect answer. These quantitative methods produce results that can be counted or measured.

Methodological research

Methodological research is the study of how research and evaluation are carried out to ensure this work is effective. RANDS methodological research has studied—

• Close-ended probe questions
• Split-panel experiments
• Weighting methods
• Methods for combining RANDS with other data sources

Close-ended probes

Close-ended probe questions explore thought processes by asking people to choose from a list of options to describe how and why they answered a previous question the way they did. Split-panel experiments involve randomly assigning people to different groups that are then asked different questions.

Probe questions embedded in RANDS surveys can expand on findings from CCQDER cognitive interviews because more people participate in the panel surveys than in cognitive interviews. Cognitive interviewing provides a detailed picture of the meanings and processes respondents use to answer questions.

Data and documentation

RANDS provides detailed reports and documentation to support analysis and use of RANDS data.

• Find reports and research articles using RANDS data.
• Access questionnaires, data and documentation from RANDS and RANDS during COVID-19.