Survey Design

CSS works with the organizations we serve to design surveys and other studies to meet their needs.  We help organizations to specify the issues they want to learn about and to identify the most efficient ways to get valid, reliable answers.  If a client wishes to do a survey, we provide the full range of services, including drafting and testing of survey questions; sampling; selection of survey protocol; administration/fielding of the survey; analysis; and design and production of usable reports. 

In many cases, as with the various types of CAHPS surveys, clients are required to use specific questions and protocols.  In these cases, we work with clients to explore options for adding in custom questions, for analyzing the standardized questions to extract the most possible useful information, and for taking advantage of all allowable protocol options to achieve the best possible response rates and the lowest possible survey costs--as well as the highest feasible scores if results are going to be used by others to assess the client's performance.

Our expertise in questionnaire development includes conducting focus groups and other types of qualitative research as we and the client organization seek to frame study issues and conducting cognitive and psychometric testing.

To maximize response rates, we focus on our contact with sample members for each mode of data collection:

• Mail surveys: designing attractive and engaging mailing pieces that foster the highest possible response rates;
• Internet surveys: composing and designing compelling invitations to participate and easy-to-use online instruments, and
• Telephone surveys: composing the most effective scripting for use by our callers, who make calls in time periods that optimize the likelihood of successful contacts.

Our protocol designs are, at the very least, based on informed judgment born of extensive experience.  But in many cases our protocol recommendations are based on much more than that.  CSS regularly conducts rigorous, controlled experiments with different protocol options so that we can tell our clients with confidence the pros and cons of each option with regard to cost, response rate, and scores.