There is no prerequisite, but some knowledge of questionnaire design is of value. This course reviews a range of survey data collection methods that are both interview-based face-to-face and telephone and self-administered questionnaires that are mailed and online, i.
Mixed mode designs are also covered as well as several hybrid modes for collecting sensitive information e. The course also covers newer methods such as mobile web and SMS text message interviews, and examines alternative data sources such as social media. It concentrates on the impact these techniques have on the quality of survey data, including error from measurement, nonresponse, and coverage, and assesses the tradeoffs between these error sources when researchers choose a mode or survey design.
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. Students will view recorded lectures and complete reading assignments in preparation for class discussion sessions which will occur twice per week, one hour per session. Students are expected to attend all discussion sessions either in person or via BlueJeans. Successful discussion sessions will occur through preparation and active participation by all participants enrolled in the course.
Students should have questions or discussion topics in mind for the class sessions. It is not necessary to be physically in Ann Arbor to take the course. Once enrollment is confirmed via email, indicate if course attendance will be in person, in Ann Arbor or via BlueJeans.
An introductory course in survey research methods or equivalent experience. If joining remotely, participant must have computer, camera and headset available to join the class via BlueJeans https: Stephen Schilling , University of Michigan 2 credit hours. The last 50 years have seen development and use multilevel and mixed models, latent and structural equation models, generalized linear models, generalized linear mixed models, item response theory IRT models, and longitudinal models across a wide variety of disciplines.
Statisticians often note the overlap between these methods but lacked a unifying approach towards estimation, testing, and application. GLLAMMs allow estimation of multilevel models for binary, ordinal, and count data that include structural equation relationships among latent variables underlying observed data.
Diverse applications include 1 multilevel models of complex survey data with multistage sampling, unequal sampling probabilities, and stratification; 2 explanatory IRT Models: This class will consist of daily morning lecture and afternoon lab. Lectures begin with an introduction to GLAMMs, focusing on their structure and estimation and then move on to specific applications, concentrating on modeling of complex survey data, biometric data, and educational data.
Florian Keusch , University of Mannheim. This 2-day workshop will introduce students to different methods of collecting data in the social sciences. Surveys are the most common form of collecting primary data in many disciplines, and this course will provide students with an overview of interview-administered face-to-face and telephone and self-administered mail, web, mobile web, and SMS survey data collection as well as the combination of multiple modes mixed mode surveys.
The course will in particular discuss the implication of survey design decisions on data quality. In addition, students will also receive an overview on alternative data sources e. Amanda Sonnega , University of Michigan. The Health and Retirement Study hrsonline. HRS is a large-scale longitudinal study with more than 20 years of data on the labor force participation and health transitions that individuals undergo toward the end of their work lives and in the years that follow.
The HRS Summer Workshop features morning lectures on basic survey content, sample design, weighting, and restricted data files. Hands-on data workshops are held every afternoon in which participants learn to work with the data including the user-friendly RAND version of the HRS data under the guidance of HRS staff. At the end of the week, students have the opportunity to present their research ideas to the class and HRS research faculty and obtain feedback.
Topics include but are not limited to in depth information on HRS data about health insurance and medical care; biomarkers, physical measures, and genetic data; cognition; health and physical functioning; linkage to Medicare; employment, retirement, and pensions and linkage toe Social Security records; psychosocial and well-being; family data; and international comparison data. This course introduces the skills needed to conduct focus group interviews.
Students will learn about the critical components of successful focus group research. They will develop a plan for a focus group study and then practice key skills. Attention will be placed on moderating, recruiting, developing questions, and analysis of focus groups.
This course will be particularly applicable for those conducting focus group research in academic, non-profit, and government settings. The course format includes daily lectures along with opportunities to practice critical skills in small groups.
Focus groups are used to understand issues, pilot test ideas, and evaluate programs. They also provide great insight when used in combination with surveys. Focus groups have been used to help design surveys, to pilot test surveys, and to understand survey findings.
Take this course if you want to learn more about how focus groups might add to your research toolbox. This course provides students with practice applying principles of question design. Students leave the course with tools to use in diagnosing problems in survey questions and writing their own survey questions. The lecture provides guidelines for writing and revising survey questions and using troubled questions from surveys as examples for revision.
Each day's session combines lecture with group discussion and analysis. For some class activities, students work in small groups to apply lecture material 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. This 2-day course will introduce participants to the basic principles of survey design, presented within the Total Survey Error framework. Tuba Suzer-Gurtekin , Univesity of Michigan. Surveys continue to play an important role in addressing many kinds of problems about many kinds of populations stand alone or as part of an integrated information system.
Application of the scientific principles underlying surveys depends on good understanding of theories and empirical research from disciplines such as psychology, sociology, statsitics and computer science. The principles include problem and hypothesis formulation, study design, sampling, questionnaire design, interviewing techniques, pretesting, modes of data collection and data cleaning, management, and analysis. Students will be trained to determine major steps in data collection design and implemetnation and to refer to literature to justify the steps.
The cousre will also discuss team and project management in the content of survey research, identifying skillsets anf technical language required. The course will also provided training in an important subset of skills needed to conduct a survey form beginning to end.
Jim Lepkowski, University of Michigan. This is a foundation course in sample survey methods and principles. The instructors will present, in a non-technical manner, basic sampling techniques such as simple random sampling, systematic sampling, stratification, and cluster sampling. The instructors will provide opportunities to implement sampling techniques in a series of exercises that accompany each topic. Participants should not expect to obtain sufficient background in this course to master survey sampling.
They can expect to become familiar with basic techniques well enough to converse with sampling statisticians more easily about sample design. Stephen Schilling , University of Michigan. Over the past half century Item Response Theory IRT has revolutionized test analysis and scoring in education, psychology, and medicine.
IRT modeling is now the standard for almost all educational assessments, college readiness exams, and patient reported outcomes measures. IRT offers substantial advantages for many technical problems that arise in creating and using tests, including test design, test equating, assessment of item and test bias, and test scoring.
IRT models have the advantage of invariance of person estimates to the collection or sample of items in a particular test, and the invariance of item parameter estimates to the sample of subjects used in test calibration. This course will begin by comparing Item Response Theory to Classical Test Theory, focusing on the assessment of measurement error.
There we will focus on the key components of IRT: We will then move to a survey of models for unidimensional sets of dichotomously scored items, including the 1-parameter or Rasch model and the 2 and 3-parameter IRT models. We will then look at extensions of IRT to ordinal and nominal data, including the partial credit model, the generalized partial credit model, the graded response model, and the nominal response model. Finally we will examine specific applications of IRT, including test design and equating, assessment of test and item bias differential item functioning , and test scoring including computerized adaptive testing.
Here we will work real world problems and applications of IRT in educational assessment and the assessment of patient reported outcomes.
Students will be provided with the knowledge and skills to perform IRT analyses using freely available software available in the R statistical environment.
Course work will include three assignments and a final project that requires the students to use IRT analyses on their own data. Class will consist of a morning lecture and an afternoon computing lab. One or more courses in statistical methods that covered regression analysis, notions of statistical inference, and probability, as well as some familiarity with statistical software such as SPSS and SAS.
This course reviews multiple methods of data collection and presents study designs for combining multiple methods within a single research project. The course focuses on the integration of survey methods with multiple alternative methods to achieve a single data collection approach using the strengths of some methods to compensate for weaknesses in other methods.
The methods examined include unstructured or in-depth interviews, semi-structured interviews, focus groups, survey interviews, observation, geographic information systems, archival research, social media analysis and hybrid methods. Emphasis will be placed on the specific contribution of each method, as well as the use of combined methods to advance specific research questions. This course is designed for those with a specific research question in mind.
Throughout the course, participants will be asked to design multi-method approaches to a research question of their choice. By the end of this course, participants will have an overview of multi-method research that will enable them to design, understand, and evaluate multi-method approaches within a single project. Standardized multi-item scales are more common in some disciplines than others. This 2-day course is designed to inspire participants from all disciplines that it is possible to develop your own high quality multi-item scales or correctly adapt existing multi-item scales and offers an introduction on how to do this.
It covers the psychometric principles of question development while adding in principles of general questionnaire design. Focusing first on Classical Measurement Theory, participants design their own multi-item scales. This is followed by a group discussion of existing multi-item scales. The course then introduces some basic statistical tools for assessing the reliability and dimensionality of multi-item scales and participants get to practice evaluating some existing scales in a computer lab session.
The course finishes with an introduction to Item Response Theory. There is no prerequisite, but a little knowledge about questionnaire design, multi-item scales and SPSS would be of value. Robert Henson , University of North Carolina. Although many surveys gather data on multiple units of analysis, most statistical procedures cannot make full use of these data and their nested structures: 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 also known as mixed models or random effect models 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.
The multi-level analysis skills taught in this course are equally applicable in many social science fields: 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. This course will be taught over a week and will include both classes that provide the basic concepts of these models in addition to labs where participants will get hands on experience and practice with respect to determining the appropriate model, running the analysis, evaluating the reasonableness of the model and interpreting the results.
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 or equivalent.
If you cannot meet this criterion, you must speak directly to the instructor prior to being given permission to enroll.
It is not necessary to be physically in Ann Arbor to participate in these workshops. Probability and Non-probability Sampling Methods is a sampling course that differs from traditional sampling classes. First, this class gives an equal amount of attention to both probability and non-probability sampling methods as non-probability sampling cannot be discussed meaningfully without understanding probability sampling and these two methods offer distinctive advantages and disadvantages.
The course will start with examining probability sampling techniques and their properties, including simple random selection, systematic selection, cluster sampling, stratified sampling, and probability proportionate to size selection.
Issues of weighting to compensate for unequal chances of selection and variance estimation for calculating confidence intervals are also examined. Then the wide variety of non-probability sampling methods are examined, from panel-based convenience samples, to river samples, quota samples, respondent-driven samples, and other techniques.
The properties of these samples are discussed, and assumptions needed to obtain estimates are examined. We will also examine these two approaches from the total survey error perspectives. The lab sessions to be held after each class will combine R programming and group discussions on the topics that need to be considered when implementing various sampling approaches.
Hands-on examples of frame preparation, sample draws, post-survey adjustments and analysis specific to design will be provided and discussed. The course is not designed to provide the mastery of survey sampling. Rather, it provides materials that will accommodate participants to become familiar with advantages and disadvantages of the two methods and their implementation which will allow them to make informed design decisions.
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.
Jessica Broome , Jessica Broome Research. This course provides an overview of the art and science of questionnaire design. Topics will include basic principles of questionnaire design; factual and non-factual questions; techniques for asking about sensitive topics; designing scales and response options; survey mode considerations; and an introduction to pre-testing surveys. The course will consist of both lectures and hands-on activities. RSD has financial support available to those who qualify.
Responsive survey design RSD refers to a method for designing surveys that has been demonstrated to increase the quality and efficiency of survey data collection. RSD uses evidence from early phases of data collection to make design decisions for later phases. Beginning in the Summer Institute, we will offer a series of eleven one-day short courses in RSD techniques. Survey Methodology for Randomized Controlled Trails does not have the remote participation option.
Survey Methodology for Randomized Controlled Trials half-day workshop. Randomized Controlled Trials RCTs are an important tool for tests of internal validity of causal claims in both health and social sciences.
Experimental research is commonly used in sciences such as sociology and psychology, physics, chemistry, biology and medicine etc. It is a collection of research designs which use manipulation and controlled testing to understand causal processes. Generally, one or more variables are manipulated to determine their effect on a dependent variable.
The experimental method is a systematic and scientific approach to research in which the researcher manipulates one or more variables, and controls and measures any change in other variables.
Experimental Research is often used where:. The word experimental research has a range of definitions. In the strict sense, experimental research is what we call a true experiment. It has a control group , the subjects have been randomly assigned between the groups, and the researcher only tests one effect at a time. It is also important to know what variable s you want to test and measure.
A very wide definition of experimental research, or a quasi experiment , is research where the scientist actively influences something to observe the consequences. Most experiments tend to fall in between the strict and the wide definition.
A rule of thumb is that physical sciences, such as physics, chemistry and geology tend to define experiments more narrowly than social sciences, such as sociology and psychology, which conduct experiments closer to the wider definition.
Experiments are conducted to be able to predict phenomenons. Typically, an experiment is constructed to be able to explain some kind of causation. Experimental research is important to society - it helps us to improve our everyday lives. After deciding the topic of interest, the researcher tries to define the research problem. This helps the researcher to focus on a more narrow research area to be able to study it appropriately.
The research problem is often operationalizationed , to define how to measure the research problem. The results will depend on the exact measurements that the researcher chooses and may be operationalized differently in another study to test the main conclusions of the study. An ad hoc analysis is a hypothesis invented after testing is done, to try to explain why the contrary evidence.
There are various aspects to remember when constructing an experiment. Planning ahead ensures that the experiment is carried out properly and that the results reflect the real world, in the best possible way.
Sampling groups correctly is especially important when we have more than one condition in the experiment. One sample group often serves as a control group , whilst others are tested under the experimental conditions. Deciding the sample groups can be done in using many different sampling techniques. Population sampling may chosen by a number of methods, such as randomization , "quasi-randomization" and pairing. Reducing sampling errors is vital for getting valid results from experiments.
Researchers often adjust the sample size to minimize chances of random errors. Here are some common sampling techniques:. The research design is chosen based on a range of factors. Important factors when choosing the design are feasibility, time, cost, ethics, measurement problems and what you would like to test. The design of the experiment is critical for the validity of the results. It may be wise to first conduct a pilot-study or two before you do the real experiment.
This ensures that the experiment measures what it should, and that everything is set up right. Minor errors, which could potentially destroy the experiment, are often found during this process. With a pilot study, you can get information about errors and problems, and improve the design, before putting a lot of effort into the real experiment.
If the experiments involve humans, a common strategy is to first have a pilot study with someone involved in the research, but not too closely, and then arrange a pilot with a person who resembles the subject s. Those two different pilots are likely to give the researcher good information about any problems in the experiment. An experiment is typically carried out by manipulating a variable, called the independent variable , affecting the experimental group. The effect that the researcher is interested in, the dependent variable s , is measured.
Identifying and controlling non-experimental factors which the researcher does not want to influence the effects, is crucial to drawing a valid conclusion.
In statistics, quality assurance, and survey methodology, sampling is the selection of a subset (a statistical sample) of individuals from within a statistical population to estimate characteristics of the whole population. Two advantages of sampling are that the cost is lower and data collection is faster than measuring the entire population. Each .
In sociology and statistics research, snowball sampling (or chain sampling, chain-referral sampling, referral sampling) is a nonprobability sampling technique where existing study subjects recruit future subjects from among their acquaintances. Thus the sample group is said to grow like a rolling snowball. As the sample builds up, enough data are gathered to be useful for research.
SAMPLING TECHNIQUES INTRODUCTION Many professions (business, government, engineering, science, social research, agriculture, etc.) seek the broadest possible factual basis for decision-making. It is incumbent on the researcher to clearly define the target population. There are no strict rules to follow, and the researcher must rely on logic and judgment.
If you work in sampling, survey design, even experimental design, this book remains essential. Don't think that because this book is decades old . Buy Sampling Rare or Elusive Species: Concepts, Designs, and Techniques for Estimating Population Parameters on mihtorg.ga FREE .