| |
SRP maintains a highly experienced staff of survey professionals, trained
interviewers, and data management specialists. SRP services are available
to faculty, staff, and students throughout the University, and to local,
state, and federal agencies and others working in the public interest. SRP services are also available to faculty from other institutions.
SRP BROCHURE
SRP is able to carry out any or all phases of a survey research project,
including sample design, questionnaire design, data collection, data processing,
statistical analysis, and reporting. SRP staff has the training and experience
to design scientifically valid studies customized to client research needs.
SRP can help clarify study objectives, and provide advice on how to obtain
the highest quality data within available resource constraints. What makes
SRP unique at the University is our survey data collection infrastructure,
including a 32-station computer-assisted telephone interviewing (CATI)
facility, a pool of trained and experienced interviewers, and professional
supervisory staff to ensure continuous quality control. More detailed
descriptions of SRP services and experience are presented below.
Survey Design
Sample Design
SRP staff provides advice on sampling methodology and sample size to
meet study objectives and budget. Services include in-house development
of specialized sample frames, or consultation with commercial sampling
firms to obtain an appropriate sample. Examples of the types of samples
provided include:
- Random digit dialing (RDD) samples: particularly suited for
telephone surveys of the general population, randomly selected computer-generated
telephone numbers, including both listed and unlisted, are used to target
specific geographic areas. RDD sampling for any region of the United
States is available (e.g., county, state, MSA, national). SRP uses software
to draw RDD samples in-house, and also has the capability to target
areas with particular demographic characteristics (e.g., ethnic and racial minorities; low/high income households).
- Targeted/listed samples: samples to reach specific subgroups
or rare populations. These are useful for telephone recruitment of participants
for larger studies. They are also used for supplementing RDD samples
with over-samples of rare populations. SRP can also assist in developing
or updating study-specific population lists.
Data Collection Methodology
SRP staff can provide:
- Advice on the most productive and cost-effective method of data
collection (mail, telephone, in-person, internet/web-based, or mixed-mode).
- Specification of procedures for maximizing response rates within
the framework of available resources. For example, recommended number
of attempts to contact respondents, securing initial cooperation, use of incentives, design of effective cover letters, following-up incomplete interviews, converting refusals into acceptances,
and use of mixed-mode designs.
Questionnaire Development
SRP staff can provide assistance with all aspects of the development
of a reliable and valid instrument, including:
- Writing, formatting, and ordering questions.
- Asking sensitive questions.
- Instrument pre-testing (both traditional field pre-tests and “in-depth” cognitive interviewing techniques for initial item development).
Data Collection
SRP conducts local, regional, and national telephone and mail surveys
with both the general and special populations, as well as in-person surveys
in the Southwestern Pennsylvania region. SRP also conducts internet/web-based
surveys.
Interviewer Training and Quality Control
SRP relies on a pool of experienced interviewers for telephone and in-person
surveys. These include both undergraduate and graduate students, as well
as non-students under the management of senior field work supervisors.
Interviewers are trained in general and project-specific interviewing
techniques, such as maintaining objectivity, reading questions as written,
and securing respondent cooperation. Training involves a combination of
lecture, written hand-outs, video instruction, and role-playing. Interviewers
are continuously monitored during data collection for quality assurance,
and periodic re-training sessions occur as necessary. Fieldwork supervisors
and senior staff members assigned to a particular project meet on a regular
basis to discuss any problems or issues that may arise.
SRP can also provide interviewers and/or interviewer training for non-UCSUR studies.
Computer-Assisted Telephone Interviewing
A 32-two line phone bank at UCSUR is dedicated to a computer-assisted
telephone interviewing (CATI) system. Survey instruments are programmed
into the system, and interviewers read questions as they appear on the
screen and enter the respondent’s answers directly into the computer.
CATI allows complex question contingencies, sample management, and call
scheduling to be handled automatically. CATI also affords the opportunity
to incorporate randomized experiments (e.g., varying question order, question
wording, response options) into surveys. In order to maximize response
rates, calls are typically made on varying days of the week at different
times to maximize the probability of contacting respondents, and attempts
are routinely made to convert initial refusals into completed interviews.
The CATI lab is open 7 days a week to maximize probability of contact with respondents. Data are automatically exported to a statistical analysis package and
examined by programmer/data management specialists on a daily basis for
additional quality control purposes.
Web-Based Data Collection
SRP has the capability to conduct web-based surveys and other online
data collection and management. These are ideal for collecting data from
networked communities with lists of e-mail addresses, and are often more
cost effective than phone or traditional mail surveys. UCSUR operates
secure servers that host such activities. Our web surveys are similar to our CATI surveys in that they incorporate complex question contingencies and randomized experiments.
Additional Data Collection Services
- Telephone surveys using paper/pencil methodology and mail surveys are
also available. In-person surveys in the Southwestern Pennsylvania region
can also be carried out.
- Foreign language interviews - SPR has conducted Spanish, French, and Chinese interviews and can accommodate a variety of languages through our liaisons with various University schools and departments.
Data Management/Database Development/
Statistical Analysis/Reporting
Data Management
Using a variety of computer programs, languages, data base management
software, and operating systems, SRP staff are available to perform a
number of data management functions, including
editing, coding, and survey data processing, including data entry. SRP
staff also has the capability to integrate survey respondent data with
GIS-based spatial data, and with other secondary data sources
(e.g., U.S. Census data).
Database Development
SRP develops custom database systems for recruiting studies, investigational studies, publication tracking, journal management, special University projects, and other research initiatives. These database systems provide enterprise level data management capabilities for University researchers.
Statistical Analysis of Survey Data
SRP staff can perform statistical analyses of survey data, ranging from
simple tabulations to relatively complex statistical modeling using a
variety of statistical packages. Analytic capabilities include:
- Descriptive statistical analyses of survey responses, including
calculations of sampling error, and statistical adjustments for unequal
selection probabilities and non-response.
- Cross-classification analyses with demo-graphic or other predictive
factors.
- Multivariate analyses, including ANOVA, factor analysis, linear and
non-linear regression, logistic regression, categorical methods, and
structural equation modeling.
Report Preparation and Presentation
Technical reports detailing specific methodologies are provided for each
survey project. SRP staff also has experience in the production of a variety
of substantive reports, including:
- In-depth narrative reports.
- Oral presentation of results to interested parties.
Experience
SRP has conducted a wide range of survey research projects over the past
three years. The surveys have involved both general and special populations,
have focused on a wide variety of topics, and have incorporated various
modes of data collection (i.e., telephone, mail, web, face-to-face, mixed-mode).
Our clients have included University faculty from a variety of schools
and departments, faculty from other universities, state and local governments,
and other outside agencies. The major types of projects SRP conducts and
specific examples of recent or current SRP studies are presented next.
General Population/RDD Surveys
These are surveys that attempt to make generalizations about the general
population, and typically involve the use of RDD sample telephone surveys.
- RDD telephone survey in the 11-county Southwestern PA region focused on public opinion concerning water quality and management issues. (n = 1,500)
- RDD telephone survey in the Tri-state area (PA, OH, WV) focused on participation and education in the arts, as part of the Pittsburgh Regional Indicators Project. (n = 1,200)
- National RDD surveys in the U.S. and Canada exploring public conceptions of and attitudes toward democracy and government institutions, to provide comparison data for the Latin American Public Opinion Project. (n = 600 U.S., n = 600 Canada)
- National RDD telephone surveys focusing on a variety of issues related to needs for and use of various sources of information, including the internet, libraries, and museums. (n = 5,000)
- Statewide RDD telephone survey of adult Pennsylvanians focusing on use of public libraries to determine the overall economic benefits of public libraries in PA (n = 1,000)
- Telephone survey to assist the City of Pittsburgh with an evaluation of the crime prevention and neighborhood revitalization Weed & Seed program in three city neighborhoods. (n=1,200)
- National RDD telephone survey using randomized experiments to explore interracial differences and evaluations and perceptions of the criminal justice system. (n=1,200)
Medical/Clinical Follow-Up Studies Involving Survey Components
These involve survey data collection conducted as part of larger medical/clinical
studies, and typically involve client-supplied list-based samples.
- Telephone surveys of stem cell donors to determine the role of ethnic and cultural factors on donor-related outcomes and quality of life as part of a five-year, multi-site study. (n = 600 per year, n = 3,000 total)
- Telephone interviews focusing on relationships and interactions with healthcare providers with veterans who have been diagnosed with Hepatitis C as part of a two and half year study. (n = 1,200)
- Telephone interviews focusing on quality of life among end-stage renal disease patients as part of a multi-site, randomized clinical trial comparing different methods of delivery of hemodialysis. (n = 500 total over 4 years)
- Telephone follow-up surveys focusing on quality of life issues as part of a randomized clinical trial comparing peripheral blood stem cell vs. bone marrow donation (n = 5,000 total interviews over 5 years)
- Telephone interviews with HIV+ and HIV- veterans to explore the role of alcohol use in adherence to medication treatment. (n = 3,000)
- Thirty-day follow-up telephone interviews regarding vital status, functional status, and follow-up treatment among pneumonia patients in a multi-site clinical trial of treatment protocols. (n = 2,000)
Special Population Surveys
These are surveys of special populations, including professional groups,
specific sub-populations, and those utilizing particular services. These
also typically involve client-supplied list-based samples.
- Web-based surveys of all public and academic libraries and samples of school and special libraries focusing on staffing, budget, and other workforce issues. (n = 20,000 total surveys)
- Web-based survey of local manufacturing firms focusing on their performance and busines practices. (n = 150)
- Web-based survey of local steel-supplier firms regarding their ties to the Pittsburgh region. (n = 100)
- Telephone surveys of former participants of the Youth Works jobs program for adolescents focusing on their education, career plans, and employment history. (n = 300)
- Mailed survey to independent pharmacists in PA regarding their opinions about a new law allowing them to administer immunizations (n = 700)
- Telephone survey of human resource personnel at local firms regarding experiences with local student interns (n = 500)
- Web-based survey of biological scientists focusing on varying definitions and uses of the gene concept in contemporary science. (n = 500)
- Annual mail surveys with telephone follow-up of medical assistance patient satisfaction with behavioral health treatment in Allegheny, Berks, York, and Adams counties. (n = 3,000 per year)
- Statewide telephone surveys with child care provider centers/home-based child care providers and web-based surveys of colleges and universities who offer training and degrees in early childhood education. (n = 750 telephone surveys; n = 100 web surveys)
Institutional Surveys
These are surveys conducted for administrators and faculty at the University
of Pittsburgh that focus on institutional issues with policy implications.
- Web-based survey of a representative sample of Pitt undergraduates focused on their experiences with writing as part of their undergraduate education. (n = 1,000)
- Web-based survey of Law School faculty focusing on several areas relevant to strategic planning. (n = 40)
- Annual telephone surveys of Pitt undergraduate and withdrawing students to determine factors related to satisfaction and retention. (n = 2,000 per year)
- Web-based survey to develop a searchable database with information about faculty community service activities. (n = 750)
Participant Recruitment Efforts
UCSUR has conducted several recruitment studies over the past five years.
These have involved locating members of the target population and administering
brief screener instruments to determine eligibility and describe the larger
study. Several of these have been aimed at the recruitment of community
control samples in case-control studies. The Center also has extensive
experience in using RDD telephone surveys with screening questions to
obtain representative samples of sub-populations (e.g., particular age
groups; racial groups; people residing in specific zip-codes; those reporting
a primary care physician that they see regularly).
- RDD telephone recruitment of healthy control adults for a study of genetic factors in schizophrenia. (n = 200)
- RDD telephone recruitment of control families with 14–18 year old adolescents for a study focusing on how adolescent substance use changes over time. (n = 200)
- RDD telephone recruitment of control females between the ages of 45 and 70 in PA, NY and OH for a study of ovarian cancer. (n = 1,850 over 4 years)
- RDD telephone recruitment of older adults (age 50–75) for participation in an experiment focused on perceptions of risk for cancer. (n = 300)
- RDD telephone recruitment of community control families with 7-17 year olds for comparison to families with parents with bipolar disorder. (n = 300)
Survey Design/Data Management/Statistical
Analysis
These are studies, not involving primary data collection, for which SRP
provides survey design assistance, data management services, or statistical
analysis of survey data.
- Data entry and data processing for statewide surveys of K-12 teachers
and principals focused on evaluations and perceptions of newly developed
student performance standards (n = 1,500).
- Statistical analyses of the results of surveys assessing the impact
of electronic medical records systems on physicians and patients in
the UPMC health system.
Additional UCSUR research capabilities include:
- Modeling of complex systems, including regional economic
analysis, econometric forecasting and impact modeling;
- Policy analysis and program evaluation, including
design, data management, data analysis, and report writing for both
process and outcome evaluations. In addition, training and technical
assistance on such issues is provided to agencies that want to conduct
their own evaluations;
- Geographic information systems and mapping capabilities;
- Acquisition and management of secondary data, including
U.S. census and other government data.
For more information, contact:
Scott R. Beach, PhD
Survey Research Program
Director
University of Pittsburgh
121 University Place
Pittsburgh, PA 15260
Telephone: 412-624-7785
Fax: 412-624-4810
email: scottb@pitt.edu
|
|
