Survey Research

  1. Survey Design
  2. Questionnaire Development
  3. Sampling
  4. Data Collection
  5. Data Analysis
  6. Weighting Scheme and Estimation
  7. OMB Clearance and Public Data File
  8. Telephone, Mail, Internet, Focus Group
  9. On-site Interviews
  10. CATI and RDD

Our staff has extensive experience with the design and implementation of face-to-face, telephone, and mail-in surveys with a variety of sampling unit including person, household, and establishments (or firms). Our exceptional expertise lies in large multi-stage national surveys with stratification and clustering being utilized at various stages and in development of public health related surveys with complex skipping and branching patterns.

For some complex surveys, we have experience in constructing sampling frames from a large number of sources and in extensive stratification of population units. For estimation purposes, we develop weight components based on selection probabilities, non-response rates, and adjustments to known or estimated universe totals. We also develop survey estimation and variance estimation formulas, applying standard textbook formulas, Taylor Series approximation methods, and pseudo-replication procedures such as balanced half-sample replication and the jackknife procedure.

For survey instrumentation, we focus on developing a questionnaire that evokes responses to gather reliable and valid information, to reduce respondent burden, and to maximize the response rate. We have hands-on experience in a computer-assisted telephone interview (CATI) that facilitates the interview process and entering, coding, and editing of data. The use of CATI is generally efficient in controlling skip patterns based on individual responses, reduce clerical error by eliminating the need for separate data entry from hand-recorded questionnaire, and ensure correct entry through pre-defined format mechanism. Quality control measures are extensively utilized during data collection in order to minimize interviewer variance and bias.

As a final product of a national survey, we generally produce public use files for government data. We focus on maintaining strict confidentiality requirement by consolidating or re-categorizing specific data fields, depressing or collapsing variables, or coding variables.

Our selected recent experiences are listed below.

Survey of Maryland Lawyers for Pro Bono Activities

State of Maryland / Administratice Office of the Courts

This survey intends to annually document the pro bono legal services rendered by 32,000 Maryland lawyers. This annual survey serves several purposes: to increase the level of awareness of the need for pro bono services, to inform lawyers with a list of pro bono opportunities with contact numbers, to allow lawyers a chance to review their pro bono participation, and thus to induce more participation in pro bono activities being rendered in Maryland. This survey adopts a combination of mail-in with telephone follow-up survey, fax-in survey, and web-based internet survey instruments.

Design and Conduct Survey Task Order (ADDSTO)

Centers for Medicare and Medicaid Services (CMS)

This project is a five year IDIQ Task Order Master Contract to design and conduct survey research projects relating to federal policy makings on Medicare, Medicaid, State Children's Health Insurance Program (SCHIP), beneficiary services, health care providers, and provider payment policy.

National Hospital Malpractice Insurance Survey (NHMIS)

Centers for Medicare and Medicaid Services (CMS)

The National Hospital Malpractice Insurance Survey (NHMIS) is a national stratified smaple survey of short-term hospitals. This study aimed to estimate hospital malpractice insurance costs in light of the increasingly prevalent self-insurance and develops a method for calculating the malpractice insurance cost change over time at the national level. The study results were used to rebase the Medicare hospital input price index and the annual percent change to update the Medicare prospective payment (PPS) operating reimbursement rates to hospitals.

Research Support Services

Centers for Medicare and Medicaid Services (CMS)

This project is a five year IDIQ Task Order Master Contract to provide survey research services in data analysis, data acquisition, data collection, and development of research files in support of research, analytic, and demonstration projects.

Survey of Commercial Insurers and State Insurance Offices

Centers for Medicare and Medicaid Services (CMS)

We conducted collection of rate schedules from commercial insurers and state insurance commissioners for the calculation of rate of change of hospital malpractice insurance costs, adopting an actuarial approach in converting different coverage limits into a constant coverage limit.

Survey of Montgomery County Physicians

State of Maryland / ENSR Corporation

To assess the health effects of a Maryland Montgomery county incinerator, the survey covered physicians who have practices in the study area. The frame was constructed from the database of the Maryland Board of Physician Quality Assurance, Association of State Medical Board Executive Directors (AIM). The questionnaire was developed to ask the physicians their opinions and general knowledge of their patient base with regard to respiratory illnesses, including chronic obstructive pulmonary disease (COPD) and asthma and their trends over time.

Survey of Internet-based Service Providers

Small Business Administration

We conducted a telephone interview survey with Internet-based service providers who adopted creative financing alternatives for their operation. We estimated the expected operating revenue under each alternative financing mechanism by applying industry standards and experiences of similar service providers. The survey results were used to design and evaluate the feasibility of transforming the PRO‑Net system, a nationwide Internet based database of information on small businesses, into a self-supporting entity.