Survey Research Operations (SRO) is the main data collection unit within the Survey Research Center. SRO constitutes over half of SRC, and provides a wide range of survey design, data collection and data processing services. SRO staff members work closely with members of the Survey Methodology Program and incorporate cutting-edge methodology and technical systems into their projects. SRO conducts national surveys as well as small-scale, regional, and methodological surveys. Face-to-face and telephone interviewing are conducted using computer-assisted techniques as well as paper-and-pencil mode. Large-scale mail, self-administered and web surveys are also conducted.
From proposal development through data delivery and report writing, SRO provides personalized, high quality, state-of-the-art service to their clients. SRO's team of experienced study directors assist the researcher in proposal development and all aspects of survey implementation, identifying innovative methods and techniques designed to improve quality and reduce survey costs.
Sample Design and Analysis
SRO carries out a wide range of SRC sampling activities including consulting on statistical problems, consulting on sample designs for a wide range of federal and nonfederal clients, and research to develop and apply new sampling methods.
Much of SRO’s work uses multistage area sampling of households. Probability selection methods are employed in all sample selection stages, including random selection of respondents from households. Many samples have also been designed to represent populations of special areas such as individual states, countries, metropolitan areas, or cities; national samples of work, school, and social organizations; research institutes and universities; members of professional societies and other groups; and a multipurpose national probability sample of nonhouseholds. Random digit dialing sampling of households is also used extensively for telephone surveys of household populations.
SRO’s National Sample SRO maintains a national area probability sample. The first stage units are a probability selection of counties or county groups which are permanently staffed by local interviewers employed by ISR. The sample is concentrated in geographic areas -- termed Primary Sampling Units (PSUs) -- which are counties or metropolitan statistical areas (MSAs). The national sample is readily adapted to smaller or larger sample size requirements or to studies of special populations, such as specific age groups, blacks, or Hispanics.
ATLAS*GIS Capabilities Since 1991, ISR has been using TIGER (Topographically Integrated Geographic Encoding and Reference) and GIS (Geographic Information Systems) files to produce sample segment maps, perform stratification, and conduct spatial analysis of attribute data. This reduces the amount of time it takes listers and interviewers to find sample segments. By using ATLAS*GIS to produce sample segment maps, a segment can be easily displayed at different scales, customized to show only certain geographic features, and supplemented with other pertinent segment information.
Software Development SRC has developed a package of computer programs (IVEware) which calculate sampling errors for survey statistics using either pseudo-replication techniques – Balanced Repeated Replication and Jackknife Repeated Replication – or Taylor Series approximation methods of estimation. By specifying appropriate options for these programs, SRO staff can calculate sampling errors for ratio means, totals, simple differences of ratio means, regression coefficients, simple or multiple correlation coefficients (standardized and unstandardized), and partial correlation coefficients. In a single run, these programs can calculate estimates for the total population, its subclasses and domains. Each sampling error program is designed to deal automatically with weighted estimates including post stratification adjustments.
Data Collection and Processing Services
SRO conducts both centralized and decentralized data collection activities. Highly experienced production managers organize the efforts of the decentralized and centralized interviewing staff. Research technicians carry out data processing activities (including editing, coding, and behavior coding). Staff from sampling and computing implement the sample design. These staff also produce reports which monitor interview production, quality control measures, data processing, and field costs. Study directors and survey managers work closely with the researcher and coordinate the overall implementation process.
Ann Arbor Staff SRO’s office in Ann Arbor includes more than 100 staff members who oversee and monitor all data collection activities. Their responsibilities include: questionnaire development, programming, testing questionnaire application software and related word processing, budgeting and scheduling projects, providing materials and supplies to the interviewing staff, and support services to the Ann Arbor staff. They carry out quality control activities including reviewing completed questionnaires and noninterview reports, conducting interview validations, and generating progress reports used in evaluating the production status of each project. The Ann Arbor office is also responsible for monitoring project costs and providing regular cost and production reports to clients.
Project Management SRO maintains a staff of highly skilled study directors and project managers who work closely with the researcher to plan and implement each project. SRO study directors and project managers are trained in survey methods as well as the latest project management techniques. They become active members of the research team for each project, and provide input on survey design, questionnaire development, field procedures, and study-specific interviewer training, to ensure that project goals are met as efficiently as possible. SRO has developed dynamic reporting systems to monitor key indicators of production, with are used by the project team in making adjustments to maximize quality and efficiency throughout the data collection period. Cost monitoring is also an important part of SRO project management, and SRO’s Cost Reporting System allows the study director to update cost projections on a regular basis, to provide the client with accurate estimates of final project cost.
Decentralized Data Collection The field staff includes production managers located in Ann Arbor and throughout SRO’s primary sampling units. On average, these employees have more than a decade of experience in field management, including: conducting interviews, hiring, training and supervising interviewers, and implementing a wide variety of survey designs. A staff of team leaders and team leader coordinators assists the production managers. These staff members are responsible for quality control activities in the field, as well as for validation activities and monitoring interviewers’ weekly progress and cost reports.
SRO maintains an active national field staff of more than 1000 interviewers. Each newly hired interviewer receives three days of training in general interviewing techniques, as well as study-specific training for each project on which he or she works (even if the interviewer has worked for another organization). SRO interviewers are mature, well-educated, experienced, and represent the diversity of the 120 primary areas in which they work. SRO also maintains a bilingual interviewing staff in areas of high concentrations of non-English speaking households.
Interviewer Training SRO maintains a training department that develops and coordinates interviewer training programs and staff development workshops and seminars. The interviewer-training regimen developed at ISR has been adopted by organizations throughout the world. The SRO training program includes sets of interactive CD-ROM materials for home study and refresher training, as well as specialized modules on topics such as refusal aversion or sampling.
Centralized Data Collection SRO operates a state-of-the-art telephone facility which has been conducting Computer_Assisted Telephone Interview (CATI) surveys since 1975. This is one cornerstone of SRO’s extensive program in computer assisted interviewing (CAI). The CAI software and physical facilities are regularly updated to ensure the highest possible quality for all national, regional and local projects conducted in the facility.
Recently, SRO redesigned and updated its telephone facility, further increasing the capacity for researchers to collect high-quality national, regional, and local survey data. The telephone facility now contains 60 interviewer carrels, reserved supervisor's stations, office space for project managers, project-specific storage areas, and a monitoring office where research technicians can carry out quality control activities. The facility’s state-of-the-art telephone system has full monitoring capabilities. Samples of interviews are monitored, and interviewers are evaluated on interviewing technique (question reading, probing, data recording, etc.) as well as on qualitative factors (pace, clarity, and tone, etc.). The telephone system allows monitoring of interviews from multiple locations in the facility, as well as remote monitoring by clients.
In addition to conducting telephone surveys, the facility also carries out numerous other activities in support of survey efforts. These support activities include: questionnaire pre-testing, respondent-interviewer behavior coding, questionnaire checking and editing, mail survey administration, focus group moderation, the recruitment of focus group participants, and data entry. The facility also maintains several toll-free phone lines for respondents, including one specifically for Spanish-speaking respondents.
Coding and Data Entry ISR performs data entry for paper and pencil studies, as well as coding for open-ended questions from both paper and pencil and CAI surveys. An experienced staff of coders, supervisors, and data entry and support personnel design and build codes. Lab staff are responsible for converting the response data into machine-usable form. They also oversee the final checking function during which errors are detected and corrected by the coders or brought to the attention of the project staff.
Coding staff do precoding of complex open-ended questions (e.g., occupation) or any portion of a questionnaire that would present problems to production coding. Production coding and data entry are done using CAI software, which facilitates working with interview data and performs a series of automatic checks designed to eliminate wild or inconsistent codes.
ISR’s Computer-Assisted Interviewing (CAI)
SRO was one of the early adopters of centralized computer assisted interviewing (CAI). Beginning in the mid 1970s SRO developed its own CAI system. Currently, SRO uses Blaise software (developed by the Central Bureau of Statistics in the Netherlands) for most projects. However, in order to maintain maximum flexibility SRO does not commit all of its resources to a single CAI system. SRO also makes available CASES and several web-based software application options. This commitment to multiple CAI systems is testament to SRO’s dedication to pushing the frontiers of the field while remaining responsive to a broad range of clients’ survey needs.
Methodological Research on Computer-Assisted Interviewing Since the time SRO conducted its first CAI surveys, SRO has been engaged in methodological research to evaluate the effects of the new method of data collection and to improve the quality of data collected using CAI. The tradition of this research program has grown to include computer-assisted interviewing more broadly and expanded to cover a greater range of issues in CAI, including Audio Computer Assisted Self Administered Interviews (A-CASI), as well as Video-CASI. We use a variety of methods to explore interviewer and respondent behavior as well as the production of errors in entering responses in CAI. More recently, we have focused attention on the design of CAI instruments and the impact of such design features on interviewer performance and data quality. SRO is also a leader in the collection and analysis of process data that are available as part of the computer-assisted data collection process.
These research efforts led to improvements in interviewer training, the use of statistical process control methods in ongoing production, changes in the design of CAI instruments, and enhancements of the interviewing software.
CAI Sample Management SRC's sample management system is yet another example of our commitment to technological excellence and innovation. The SurveyTrak system facilitates effective sample management, enables daily communication between field and central office staff, and ensures reliable, secure transmission of interview data. SurveyTrak capabilities include: uploading of case and management data (e.g., interview data, status of sample cases, updated mailing address information) from the field; downloading of case assignments, software upgrades or additions from the central office to the field; assignment, transfer and reassignment of cases by the supervisor or central office staff; recording of requests for mailing of respondent payments or letters; production of a variety of reports documenting data collection progress; communication between interviewers, supervisors and central office staff using email and message boards; and remote control of laptops for maintenance or problem solving needed by central office staff.
Other SRO technical developments include an automated codebook for Blaise questionnaires, and a web-based survey documentation system. SRO has also customized the application of an automated call scheduler for the Blaise case management system.
SRO Clients SRO serves a wide range of clients, ranging from large federal agencies to individual University researchers and small not-for-profit groups. SRO’s study directors provide customized cost estimates and study designs for each project, and work closely with each researcher and his/her team to ensure that the data collection effort meets the research goals set forth in the proposal. SRO is committed to providing the highest quality service at a reasonable cost to all clients. Please complete the Proposed Survey Information Form to inquire about capability and costs for conducting a survey with SRO.