Description of Courses
Analysis of Survey Data I
3 credit hours
Instructor: William Yeaton, Evaluation Consultant
Research in the social sciences has increasingly come to rely on statistical concepts in the presentation and analysis of data. The application of a wide variety of research designs, including both experimental and non-experimental designs, requires real understanding of fundamental statistical concepts. The primary purpose of this course is to provide a rigorous introduction to sampling statistics in the context of these differing designs. Its main objective is to develop a deep, conceptual understanding of statistical reasoning rather than to foster rote application of statistical formulae. The course begins with a broad overview of research designs frequently used by survey researchers. It then focuses upon measures of sample variability, unbiased estimates of sampling error, kinds of sampling designs, and sampling distributions of sums, means, and percents for random samples. We will supplement our textbook with a chapter on stratified sampling from the Grove et al. text "Survey methodology."A short paper will be assigned in which students will be asked to analyze data from a previously conducted survey and to report their results using principles of sampling statistics learned in the course. In the last third of the course, data analytic techniques most commonly used in the context of these research designs are presented from the perspective of sampling statistics. Topics include z-tests and t-tests for one and two groups, correlation, and regression analysis. Additional course topics include normal approximations, measurement error, hypothesis testing, probability samples, and the calculation of sample size for specified precision levels.
Prerequisite: Mathematics through college algebra.
Analysis of Survey Data II
3 credit hours
Instructor: Mick Couper, University of Michigan
Survey data have features that differ from data generated from other types of data collection methods. This course provides participants with an overview of the nature of those features and an introduction to methods that properly handle the unique features of survey data. The course begins with a brief overview of survey design and its implications for analysis, and then covers the logic and methods of analysis, measurement theory and evaluation, scaling and index construction, contingency table analysis, and linear and logistic regression methods for bivariate and multivariate models. Logistic regression is extended to incorporate multinomial and ordered logit types of models. Homework and examination problems emphasize conceptual issues in each topic. The focus is on choosing appropriate statistical tools for analysis and on interpretation of results. Application of methods taught in this course using computer software is taught in the companion course, Computer Analysis of Survey Data II.
Prerequisites: (1) Completion of at least one graduate course in statistics or an instructor approved equivalent, (2) working familiarity with statistics through product moment correlation and analysis of variance, and (3) basic familiarity with survey methods. A self-diagnostic examination is available through the Summer Institute office and must be used to assess whether students have adequate skills for this course. Contact the Summer Institute with questions about the diagnostic examination.
Building and Testing Structural Equation Models
3 credit hours
Instructor: Amiram Vinokur, University of Michigan
Since the early 80's, Structural Equation Modeling (SEM) analyses, first using LISREL and later using EQS, and AMOS user friendly software packages, have gained prominence as they replace the older traditional analytic methods of factor analysis and path analysis. SEM merges confirmatory factor analysis with path analysis and provides means for constructing, testing, and comparing comprehensive structural path models as well as comparing the goodness of fit of models and their adequacy across multiple groups (samples). This course will cover the conceptual and technical issues of Structural Equation Modeling (SEM). Following the presentation of major conceptual issues, five basic structural models will be described in detail. The models vary from simple to more complex ones. They also cover a wide range of situations including longitudinal and mediational analyses, comparisons between groups, and analyses that include data from different sources such as from parents, teachers, and children. The description and discussion of the models will provide students with the knowledge and skills to apply SEM techniques using EQS software for analyzing, evaluating, and reporting results produced by this analytic method. This knowledge is easily transferable to the use of LISREL or AMOS software. Course work will include four structured assignments that will be completed at the lab and a final paper based on a structural model that the students will construct and test with their own data. The paper will provide a report of the model, analyses, results and conclusions.
Prerequisite: One or more courses in statistics that included in-depth treatment of linear regression analysis, basic knowledge of the concepts of item analysis and internal reliability, and some familiarity with factor analysis. At least some hands-on experience with data analysis using SPSS, SAS, or similar software for data analysis is also required.
Combining Qualitative and Quantitative Methods: Introduction and Overview
1.5 credit hours
Instructor:Lisa Pearce
In this course, participants will become familiar with multiple methods of data collection and how to combine them within a single research project. We will focus on collecting data using unstructured or in-depth interviews, focus groups, participant observation, archival research, survey interviews, and hybrid methods. We will discuss the strengths and weaknesses of each approach, and we will focus on how each different method can contribute to the research question in unique ways. This course is designed for those with a specific research question in mind, but who are new to collecting data (or new to multi-method approaches to collecting data). Throughout the course, participants will be asked to design and present multi-method data collection approaches for a research question of their choice. By the end of this module, participants will have an overview of a multi-method data collection project that will enable them to design, understand, and evaluate multi-method approaches within a single project.
Prerequisite: An introductory course in survey research methods or equivalent experience.
Computer Analysis of Survey Data II
1 credit hour
Instructor: Patricia Berglund, University of Michigan
Students enrolled in Computer Analysis of Survey Data II must also be enrolled in Analysis of Survey Data II.
This course is an optional computer laboratory designed to accompany Analysis of Survey Data II. It will emphasize the use of SAS to obtain results related to topics discussed in Analysis of Survey Data II. Particular attention will be paid to manipulating software in order to complete assignments from Analysis of Survey Data II. Secondarily, some attention to interpretation of results will also be included. The course will cover file preparation and manipulation, exploring data structure preparatory to index construction, index construction and evaluation, data exploration using descriptive and graphic techniques, bivariate and multivariate regression analyses, logistic regression analysis, and contingency table analysis. SAS will be used through the University of Michigan computing environment.
Prerequisite: (1) Completion of at least one graduate course in statistics, or an instructor approved equivalent level of experience in statistical methods, and (2) basic familiarity with survey methods. Enrollment in Analysis of Survey Data II is required.
Data Collection Methods
3 credit hours, Video Course
Instructor: Phillip Brenner, University of Michigan; Fred Conrad, University of Michigan
This course reviews alternative data collection methods used in surveys, focusing on interviewer-administered methods. It concentrates on the impact these techniques have on the quality of survey data, including measurement error properties, nonresponse, and coverage errors. The course reviews the literature on data collection methods, focusing on comparisons of major modes (face-to-face, telephone, and mail) and alternative methods of data collection (diaries, administrative records, direct observation, etc.) The implications of mode decisions for data quality and the data collection process are discussed. Special attention is paid to the statistical and social science literatures on interviewer effects and nonresponse. Current advances in computer-assisted survey information collection (including CATI, CAPI, TDE, and VRE) will be reviewed. This is not a how-to-do-it course on survey data collection, but rather focuses on the error properties of key aspects of the data collection process.
Prerequisite: An introductory course in survey research methods or equivalent experience.
Design and Analysis of Complex Sample Survey Data
3 credit hours (This class will meet for 8 weeks)
Instructor: Steven Heeringa, University of Michigan; Patricia Berglund, University of Michigan; Brady West, University of Michigan
This course provides an introduction to specialized software procedures that have been developed for the analysis of complex sample survey data. The course begins by considering the sampling designs of specific surveys: the National Comorbidity Survey (NCS), the National Health and Nutrition Examination Surveys (NHANES), and the 1984-1986 Longitudinal Survey of Aging (LSOA). Relevant design features of the NCS, NHANES and LSOA include weights that take into account differences in probability of selection into the sample and differences in response rates, as well as stratification and clustering in the multistage sampling procedures used in identifying the sampled households and individuals.
Prerequisite:Two graduate-level courses in statistical methods, familiarity with basic sample design concepts, and data analytic techniques such as linear and logistic regression.
Examining the Health and Retirement Study (HRS) Workshop-2012
not for credit
Instructors: Mary Beth Ofstedal, University of Michigan; Jinkook Lee, RAND Corporation
The HRS is a large-scale longitudinal study of the labor force participation and health transitions that individuals undergo toward the end of their work lives and in the years that follow. The survey collects information about income, work, assets, pension plans, health insurance, disability, physical health and functioning, cognitive functioning, psychosocial factors, family stuructre and transfers, and health care expenditures. The HRS Summer Workshop is intended to give participants an introduction to the HRS that will enable them to use the data for research. The format of the Summer Workshop is morning lectures on topics including basic survey content, sample design, weighting, and restricted data files and a hands-on data workshop each afternoon in which participants learn to work with the data under the guidance of HRS staff. In 2012, the workshop will feature a topical focus on the HRS international sister studies.
Hierarchical Linear Models
1 credit hour
Instructor: Joop Hox, Utrecht University
The purpose of this course is to help researchers and students who want to apply multilevel techniques in their research. The course considers multilevel regression models in detail, and introduces multilevel structural equation models. It starts with an introduction to the basic two-level regression model, estimation methods, and interpretation of results. Next, it discusses extensions and special applications, such as logistic regression models, longitudinal models, and multivariate models. At the end of the course, multilevel structural equation models are introduced briefly. Although this is not a computer course, references are made to multilevel software packages, and there are software demonstrations on example data sets. A basic understanding of statistical inference, and some experience with analysis of variance and multiple regression analysis are prerequisites. Some acquaintance with structural equation modeling is useful, but not required.
Improving Research Using Principles of Experimental and Quasi-experimental Design
3 credit hours
Instructor: William Yeaton, Evaluation Consultant
This course is tailored for two audiences: 1) Survey researchers who wish to increase their general methodological skills; 2) Researchers who seek a greater understanding of experimental and quasi-experimental design. Surveys are more than data collection. In fact, contemporary surveys have become increasingly complex in their makeup and purpose. The best surveys now utilize an ever-expanding variety of methodological principles to better manage the challenges posed by complexity. These principles can be used both to plan future research and to critique the existing literature. This course will expand the toolkit of researchers by applying principles of experimental and quasi-experimental research methods to enhance the quality of various kinds of studies, including those based on surveys, field and laboratory research, and archival databases. The multiple ways in which high quality surveys can be used to maximize study inference will also be discussed. Since experiments that require random assignment are typically not feasible due to real-world constraints, focus will be placed upon practical design and measurement strategies rather than on statistical analysis. Especially important are those design procedures aimed to strengthen causal inference and to generalize study results. Examples will be taken from a variety of disciplines including sociology, medicine, psychology, public health, education, political science, social work, and business. Recent developments in meta-analysis will be discussed in the context of inferences that cannot be made within single studies, including those based on results of multiple surveys.
Prerequisite: Knowledge of introductory statistics.
JOBS Training Workshop: Learning to Become a Trainer in the JOBS Program for the Unemployed
not for credit
Instructors: Steve Barnaby and Paula Wishart
This course prepares staff in human resources, social work, and community based service organizations to deliver the JOBS program to unemployed job seekers. In this two-week course, participants will acquire the skills to deliver the JOBS program toward receiving final certification to deliver the program in their organization. The course will provide participants with: 1] key skills for delivering JOBS, 2] principles underlying effective program delivery, and 3] details for effective administration of JOBS in the participants own organization. To assure a high quality learning experience participants will practice the delivery of the entire JOBS program during the course. Detailed description.
These two popular short courses are scheduled so that in a single week students can obtain experience in current recommendations for both the design of survey questions and the methods for testing those questions.
Introduction to Applied Questionnaire Design
1 credit hour
Instructor: Nora Cate Schaeffer, University of Wisconsin-Madison
This part of the course provides students with practice applying principles of question design, so that students leave the course with tools to use in diagnosing problems in survey questions and writing their own survey questions. Each day's session combines lecture with group discussion and analysis. The lecture provides guidelines for writing and revising survey questions and using troubled questions from surveys as examples for revision. For some hands-on portions, students, organized into small workgroups, use the guidelines to identify problems in the survey questions and propose solutions. Assignments require that students write new questions or revise problematic questions and administer them to fellow students. Sessions consider both questions about events and behaviors and questions about subjective phenomena (such as attitudes, evaluations, and internal states).
Question Testing Methods is a course that complements well with this class.
Question Testing Methods
1 credit hour
Instructor: Pamela Campanelli, UK Survey Methods Consultant
This course aims to introduce the broad range of techniques currently available to test and improve survey questionnaires. It has two strands: the first focusing on the theoretical and experimental literature related to question testing; the second, a "hands-on" approach, focusing on how to implement each method. Question testing methods covered include standard field pretesting, expert review, congitive forms appraisal, interviewer rating form, respondent debriefing and vignettes, classical behavior coding and sequence-based approaches, cognitive interviewing and the "3 Step Test Interview", focus groups, and split ballot experiments. Discussion will also focus on the strengths and weaknesses of each individual method as well as proposals for multi-method question evaluation strategies.
Introduction to Applied Questionnaire Design is a course that complements well with this class.
Prerequisite: A course in questionnaire design or equivalent experience.
Introduction to Focus Groups as Qualitative Research
1.5 credit hours
Instructor: David Morgan, Portland State University
This course will cover the design and execution of research projects using focus groups. The course will emphasize four basic topics: 1) how to design projects using focus groups, including issues involved in the selection and recruitment of participants; 2) how to write interview guides; 3) how to moderate focus groups; and 4) how to analyze the data from focus groups. For each of these four topics, the variety of options that are available will be presented, followed by a discussion on how to evaluate these options for your particular research purpose.
Prerequisite: An introductory course in research methods or equivalent experience.
Introduction to Survey Nonresponse
1 credit hour Video Course
Instructor: Edith de Leeuw, Utrecht University
The purpose of this course is to provide participants with an efficient and general tool-bag to fight both unit-and item nonresponse. The theme of the course is Integrated Management of Missing Data, that is, planning for nonresponse in the design phase of a survey. Reduction of unit and item nonresponse requires a careful survey design and planned fieldwork procedures. In addition it also requires the collection of auxiliary data to inform decisions about nonresponse follow up, and the use of such information to enhance adequate post survey adjustment for nonresponse. This implies careful consideration of the likely impact of nonresponse on key statistical estimates and a thorough theoretical understanding of unit and item nonresponse. The course starts with a basic introduction in nonresponse, discussing different types of nonresponse and its consequences. This is followed by a thorough discussion of methods intended to reduce unit-nonresponse. At the end of the course the emphasis is on understanding item nonresponse, and we will finish with procedures to diagnose nonresponse mechanisms in data already collected.
Prerequisite: Basic coursework in social science research methods and statistics. Advanced coursework in survey methods would be helpful, but is not essential.
Introduction to Survey Research Techniques: Introduction to Survey Methods and their Application in International Settings
4 credit hours
Instructors: Zeina Mneimneh, University of Michigan, Beth-Ellen Pennell, University of Michigan, and guest lecturers
This course is designed to provide an overview of the theory and practice of survey research methods. Guidelines regarding each phase of the survey lifecycle from planning and preparation to implementation, documentation, and dissemination will be discussed. Topics covered include: questionnaire design, sampling design, interviewer training, data collection methods, data processing, quality control, and ethics. Please refer to the syllabus for a complete listing of topics. For each phase of the survey lifecycle, the challenges faced when designing and implementing surveys in international and multicultural settings will also be discussed. This will include challenges faced in translation and adaptation of questionnaires, implementing proper sampling techniques, quality assurance in multi-national studies, cultural variations in response styles, and data harmonization. Topics related to local infrastructure challenges, capacity building, and consequences of deviations from specifications, standardization versus localization, and differing research traditions will also be presented. Various strategies to deal with typically encountered challenges will be addressed. Examples will be drawn from several large-scale international programs. In addition to exercises and assignments that will be geared to each topic, course participants are required to prepare and present a final class project that involves responding to a mock request for proposal (RFP). The final project will provide an opportunity to apply the course material in an integrated fashion. To simulate real-life survey contexts, course participants will work in teams for the final class project.
This course is intended for students and professional who are interested in conducting surveys or who would like to learn more about the survey process and the field of survey research.
Prerequisite: Some familiarity with survey research methods is helpful, but not required.
Introduction to Survey Sampling
1 credit hours course
Instructor: Jim Lepkowski, University of Michigan
Introduction to Survey Sampling will cover the main techniques used in sampling practice: simple random sampling, cluster sampling, stratification, systematic selection, and probability proportional to size sampling. The course will also cover sampling frames, cost models, sampling error estimation techniques, and compensating for nonresponse. It is a survey research course rather than a statistics course. It focuses on design of survey samples and estimation of descriptive statistics rather than the analysis of collected data. It also focuses on sampling human populations.
Latent Class Analysis of Survey Error
1 credit hour Video Course-cancelled for 2012
Instructor: Paul Biemer, RTI International and the University of North Carolina at Chapel Hill
This one-week course will span a range of topics dealing with the quality of data collected through the survey process. The course begins with discussion of total survey error, as measured by the mean squared error, and its relationship to survey costs and general quality dimensions such as timeliness, coherence, and accessibility. Then the major sources of error in surveys are discussed in some detail, including (a) the origins of each error source (i.e., its root causes), (b) the most successful methods known for reducing the errors emanating from these error sources, and (c) methods that are most often used in practice for evaluating the effects of the sources on total survey error. The course will introduce participants to concepts and ideas for understanding the nature of survey error, techniques for improving survey quality, and, where possible, their cost implications, and methods for evaluating data quality in ongoing survey programs. The course is not designed to provide an in-depth study of any topic but rather as an introduction to the field of survey quality.
Prerequisite: Some prior research experience is helpful, but not required.
Methods of Survey Sampling
3 credit hours Video Course (This class will meet for 8 weeks)
Instructor: James Wagner, University of Michigan; Jim Lepkowski, University of Michigan
A fundamental feature of many sample surveys is a probability sample of subjects. Probability sampling requires rigorous application of mathematical principles to the selection process. Methods of Survey Sampling is a moderately advanced course in applied statistics, with an emphasis on the practical problems of sample design, which provides students with an understanding of principles and practice in skills required to select subjects and analyze sample data. Topics covered include stratified, clustered, systematic, and multi-stage sample designs, unequal probabilities and probabilities proportional to size, area and telephone sampling, ratio means, sampling errors, frame problems, cost factors, and practical designs and procedures. Emphasis is on practical considerations rather than on theoretical derivations, although understanding of principles requires review of statistical results for sample surveys. The course includes an exercise that integrates the different techniques into a comprehensive sample design.
Prerequisite: Two graduate-level courses in statistical methods.
Multi-Level Analysis of Survey Data
3 credit hours Video Course
Instructors: Valerie Lee, University of Michigan; Robert Croninger, University of Maryland
Although many surveys gather data on multiple units of analysis, most statistical procedures cannot make full use of these data and their nested structures: for example, individuals nested within groups, measures nested within individuals, and other nesting levels that may be of analytic interest. In this course, students are introduced to an increasingly common statistical technique used to address both the methodological and conceptual challenges posed by nested data structures -- hierarchical linear modeling (HLM). The course demonstrates multiple uses of the HLM software, including growth-curve modeling, but the major focus is on the basic logic of multi-level models and the investigation of organizational effects on individual-level outcomes. Although we use data drawn from a nationally representative sample of U.S. elementary schools, students, and teachers for instructional exercises, students should feel free to use their own data provided the data have a multi-level structure and are suitable for course goals (developing and interpreting a two-level model with a random intercept and a random slope). The multi-level analysis skills taught in this course are equally applicable in many social science fields: sociology, public health, psychology, demography, political science, and in the general field of organizational theory. Typically the course enrolls students from all these fields. Students will learn to conceptualize, conduct, interpret, and write up their own multi-level analyses, as well as to understand relevant statistical and practical issues.
Prerequisite: At least one graduate-level course in statistics or quantitative methods, and experience with multivariate regression models, including both analysis of data and interpretation of results. School of Education students must have successfully completed ED-795 or equivalent. If you can not meet this criterion, you must speak directly to the instructor prior to being given permission to enroll.
Qualitative Data Analysis: With and Without the Use of Computers
1.5 credit hours
Instructor: Eben Weitzman, University of Massachusetts-Boston
This course builds upon the topics taught in the qualitative methods courses, An Introduction to Focus Groups as Qualitative Research, Combining Qualitative and Quantitative Methods: Introduction and Overview, and Qualitative Methods: Semistructured Interviewing. Once qualitative data have been collected, the researcher is faced with the (often daunting) task of making sense of it all. In this two-week course, participants will learn methods for organizing, interpreting, and drawing and verifying conclusions from qualitative data. Our approach throughout will be active, participatory, and engaged with real data. As there is a wide variety of software available to assist the researcher in managing and analyzing qualitative data, we will become familiar with some of the options and, more importantly, learn how to make intelligent, individualized selections of software that best meet the needs of a particular researcher faced with a particular project. We will apply what we learn to the analysis of real data, as we use selected software to enter, summarize, and code data collected in the previous qualitative methods courses, ending in a research report. Students who have qualitative research projects of their own, such as dissertations, may bring a sample of their data on diskette. There will be an opportunity for students in this situation to choose software for their own projects, and take some early steps in analysis. During the second week of the course, there will be a mandatory lab session held 1:00-5:00 p.m. every weekday for all participants to become familiar with relevant software.
Prerequisite: An introductory course in qualitative research methods (e.g. the previous courses in this sequence), or permission of instructor.
Qualitative Methods: Overview and Semi-Structured Interviewing
1.5 credit hours
Instructor: Nancy Riley, Bowdoin College
This course will focus on semi-structured, or in-depth, interviewing. This methodology is often most helpful in understanding complex social processes. The course will examine the goals, assumptions, process, and uses of interviewing and compare these methods to other related qualitative and quantitative methods in order to develop research designs appropriate to research goals. The course will cover all aspects of interviewing, including how to decide who to interview, how to ask good interview questions, and how to conduct successful interviews. Students will conduct interviews, and discuss the process and outcome of those interviews. We will examine the strengths and weaknesses of this methodology, particularly through discussion of some of the critiques of these methods.
Prerequisite: An introductory course in survey research methods or equivalent experience.
Questionnaire Design
3 credit hours Video Course
Instructors: Pamela Campanelli, UK Survey Methods Consultant; Emilia Peytcheva, RTI International
This course focuses on the design of questions and questionnaires used in survey research. The course will explore the theoretical and experimental literature related to question and questionnaire design as well as focusing on practical issues in the design, critique, and interpretation of survey questions that are often not taught in formal courses. There will be exercises both in and outside of class to reinforce both theory and practice, including the construction and testing of a class questionnaire.
Discussion will focus on the measurement of both factual and non-factual material. Topics include general principles of writing questions to ensure respondent understanding; techniques for measuring the occurrence of past behaviors and events; the effects of question wording, response formats, and question sequence on responses; an introduction to the psychometric perspectives in multi-item scale design; combining individual questions into a meaningful questionnaire; special guidelines for self-completion surveys (including web surveys) versus interview surveys; strategies for obtaining sensitive or personal information; and an introduction to techniques for testing survey questions.
Prerequisite: An introductory course in survey research methods or equivalent experience.
1 credit hour
Instructor: Mick Couper, University of Michigan
The course focuses on the design of web survey instruments and procedures, based on theories of human-computer interaction, interface design, and research on self-administered questionnaires and computer-assisted interviewing. The course begins with a brief review of web or Internet surveys in the general context of sources of survey error (sampling, coverage, nonresponse, measurement error) and costs. The course then discusses different approaches to web survey design and effective use of HTML tools for developing web surveys. The course draws on empirical results from experiments on alternative design approaches as well as practical experience in the design and implementation of web surveys.
Prerequisite: Basic coursework in social science research methods, including survey research. A working knowledge of survey research methods will be assumed.
Workshop in Survey Sampling Techniques
6 credit hours
Instructors: Steve Heeringa and Jim Lepkowski, University of Michigan
The Workshop in Sampling Techniques is a component of the Sampling Program for Survey Statisticians. The workshop can only be taken in conjunction with the sampling methods courses, Methods of Survey Sampling and Analysis of Complex Sample Survey Data. The workshop allows students the opportunity to implement methods studied in the companion methods courses such as segmenting and listing in area sampling; selection of a national sample of the U.S.; stratification; controlled selection; telephone sampling; national samples for developing countries; and sampling with microcomputers.
The workshop is a required class for the Sampling Program for Survey Statisticians (SPSS). The SPSS is an eight-week program. It consists of three courses: a methods course (SurvMeth 612), a course on the analysis of complex sample survey data (SurvMeth 614), and a hands-on daily workshop (SurvMeth 616). Students enrolled in these three courses are considered Fellows in the Program. The methods and the analysis courses may be taken without being a Fellow. However, the workshop cannot be taken alone. Fellows receive a certificate upon successful completion of the program.