Skip to main content

Short Courses

We are committed to advancing social science research through education and training. Each year we offer 100+ short courses on a range of topics in the social and behavioral sciences.

COVID-19 Updates

In light of the ongoing COVID-19 pandemic, all of our open courses will be held virtually via Zoom. Please visit our calendar to browse upcoming course offerings.

 


Short Course Categories


More than 2,250 participants, including UNC undergraduate and graduate students, faculty and staff, attend our short courses annually. A number of courses are also open to researchers from not-for-profit organizations, government agencies, and corporations.

Many of our courses are free for the UNC community – you just need to secure your spot with a small deposit that is refundable upon attendance for the full course. These courses fill up on a first-come, first-served basis. Courses that require non-refundable registration fees will indicate as much on the individual event pages.

Short course categories include:

 

Past course descriptions:

Advances in Mixed Methods Design (Two-day Workshop)
In this workshop I will discuss advances in mixed-method design involving the interface of qualitative and quantitative methods. First I will distinguish between multiple-methods and mixed-methods, and why mixed-method designs may present threats to validity. We will then discuss the notion of theoretical drive, and QUAL-quan and QUAN-qual simultaneous designs. Day 1 we will discuss Qualitative-quantitative mixed method designs (simultaneous and sequential designs: QUAL-quan [including single sample with data transformation] and QUAL-quan and QUAN-qual two sample designs]. We will also discuss the research process, common problems, and examining published articles. Day 2 we will discuss qual-qual and quan-quan mixed methods designs, issues of quality and common pitfalls. Finally, writing the mixed method proposal will be outlined.

Analyzing Multi-media Data in Atlas.ti
This hands-on short course will illustrate the capabilities of the PC version of ATLAS.ti 7, a software program for coding and interpreting qualitative text. It provides a network editor that allows you to graphically display and examine the hierarchical and relational connections among your codes. ATLAS.ti provides numerous options for attaching memos and comments to text segments, documents, and codes.

Atlas.ti Hands-On Workshop (Part 1)
This hands-on short course will illustrate the capabilities of the PC version of ATLAS.ti 7, a software program for coding and interpreting qualitative text. It provides a network editor that allows you to graphically display and examine the hierarchical and relational connections among your codes. ATLAS.ti provides numerous options for attaching memos and comments to text segments, documents, and codes.

ATLAS.ti Hands-on Workshop (Part 2)
This hands-on short course will cover analysis features (co-occurrence explorer, the query tool, the codes-primary-documents table) and using diagrams in your analysis.

ATLAS.ti Introductory Work Session
This hands-on short course will illustrate the capabilities of the PC version of ATLAS.ti 7, a software program for coding and interpreting qualitative text. It provides a network editor that allows you to graphically display and examine the hierarchical and relational connections among your codes. ATLAS.ti provides numerous options for attaching memos and comments to text segments, documents, and codes.

ATLAS.ti Workshop — Analysis & Special Tools
This course will cover advanced analytic tools in ATLAS.ti, such as assessing intercoder reliability, the redundant codings analyzer, and working with survey data. We will also review the co-occurrence explorer and query tool and introduce how to use multi-media data in ATLAS. Finally, we will discuss how to use diagrams to visualize emerging themes. Registration is not required.

Building a Codebook and Writing Memos
This course focuses on coding and memoing qualitative data within both the psychometric and heuristic traditions. Coding and memoing are presented as complementary tasks that occur while engaging textual data and identifying meaning units. The course will cover how memo writing can lead to efficient coding and will look at the nature of codebook evolution and the differences between deductive, inductive, and thematic codes. Students new to qualitative research are welcome to attend.

Case Selection in Qualitative Analysis
Qualitative researchers sometimes incorporate elements of quantitative methods into their research designs, such as selecting respondents “at random” for small, in-depth interview projects or identifying “representative” neighborhoods for ethnographic case studies, aiming to increase generalizability. This course presents alternatives to current practices and calls for greater clarity in the logic of design when producing ethnographic research in a multi-method intellectual environment.

Collecting Qualitative Data
This course focuses on in-depth discussion of, and practice with, the three primary qualitative data collection methods — participant observation, in-depth interviews, and focus groups. Participants have a chance to develop data collection guides and role-play while picking up tips for effective and rigorous data collection. Wherever possible, we illustrate concepts and techniques with concrete international and domestic examples. Additionally, this course covers the following topics: Selecting appropriate data collection and sampling strategies for qualitative research; Identifying and addressing ethics considerations specific to qualitative research; Describing a range of visual and other enhanced data collection techniques Where time permits, we also incorporate logistical considerations, such as remote data collection and data management issues and options, to provide guidance on implementing qualitative data collection techniques.

Dedoose
Dr. Lieber is one of the designers and developers of the Dedoose application. This session will introduce and demonstrate the key features of Dedoose. Dedoose is a Web 2.0 application for the management and analysis of qualitative and mixed methods data. Users can excerpt, code, and interpret qualitative data and link these data to quantitative data for truly integrated mixed method analysis—complete with automatically generated and presentation ready charts and graphs, memo and chat systems, and team security controls. Data can be imported from and exported to commonly used word processor and spreadsheet file formats. Dedoose is ideal for collaborative research, as multiple users can be working and communicating simultaneous and in real time, and the new Training center supports the development and maintenance of inter-rater reliability.

Ensuring That Different Audiences Can Learn from Qualitative Data and Findings
In this interactive presentation, Jolley Bruce Christman of Research for Action will discuss how qualitative researchers can get smart about how to communicate their findings so that they are useful. She will show how the strengths of the qualitative tradition — deep understanding of emic perspectives, the importance of context, and improvement of institutions as a process of cultural change — potentially offer us robust tools for communicating with the people and groups we seek to influence, but how we often forget our roots when it comes to communicating what we know and how we know it. Drawing on her 25 years of experience — the good and the disastrous — Jolley will share practical solutions and identify ongoing dilemmas that qualitative researchers need to be aware of and manage. She encourages session participants to bring their own problems of practice to serve as grist for the discussion.

Grounded Theory (2 Days)
This class introduces grounded theory methods from a social constructionist approach to new and experienced qualitative researchers. You will gain practical guidelines for handling data analysis, a deeper understanding of the logic of grounded theory, and strategies for increasing the theoretical power and reach of your work. Charmaz treats grounded theory as a set of flexible guidelines to adopt, alter, and fit particular research problems, not to apply mechanically. With these guidelines, you expedite and systematize your research. Moreover, using grounded theory sparks fresh ideas about your data. The sessions cover an overview of basic guidelines and hands-on exercises. Charmaz offers ideas about data gathering and recording to help you obtain nuanced, rich data. We discuss relationships between qualitative coding, developing analytic categories and generating theory and attend to specific grounded theory strategies of coding, memo-writing, theoretical sampling, and using comparative methods. You will receive guided practice in using each analytic step of the grounded theory method.

If you have collected some qualitative data, do bring a completed interview, set of field notes, or document to analyze. If you do not have data yet, the instructor will supply qualitative data for you. If you prefer to use a laptop for writing, bring one, but you can complete the exercises without a computer. Before concluding the class, the instructor will suggest how you can develop your analysis while writing the research report.

Grounded Theory Workshop
This is a 1.5-day course. This course will discuss the history and applications of grounded theory, the evolution of grounded theory as a method, and variations across disciplines. Other topics will include grounded theory design, theoretical sampling, and constant comparison. Day 1 will focus on optimal data collection approaches, the role of researcher in data production, open coding, and axial coding. Day 2 will cover reviewing codes, selective coding, and criteria for valid grounded theory findings.

Improving Your Interviewing Skills: Practicum in Designing Questions and Conducting Interviews
This workshop focuses on the challenges of constructing, ordering, and asking good in-depth interview questions and on being fully present during the data-collection experience. Workshop participants will have an opportunity to devise sample interview questions on a topic of their choice and to engage in practice interviews in which they learn from direct experience and participant feedback. Workshop participants should come with a topic of interest in mind for which in-depth qualitative interviewing is useful data collection tool.

Longitudinal Qualitative Research
Qualitative research that is conducted in the same sites or with the same research participants over a period of several weeks, months, or years may require somewhat different methodological and analysis approaches than a study based on one-time observations or interviews. This seminar will focus on why and how to do longitudinal qualitative research. Using an interactive format and concrete examples, we will discuss: What kinds of research questions are best answered with a longitudinal design; What conceptual and methodological questions are raised by longitudinal studies; How to adapt qualitative methods for longitudinal research; Issues involved in integrating qualitative and quantitative methods in longitudinal research; How to manage and analyze longitudinal data.

At the end of the seminar, participants should have a good idea of the issues involved in longitudinal qualitative research and an overview of how to go about doing it.

Making Choices in Qualitative Research
Beginning with the assumption that attendees know already the rationales for using a qualitative approach, this session will provide insights on practicalities and implications of choices — of numbers of cases, on the advantages and disadvantages of particular alternatives for gathering data (e.g. focus groups vs. interviews vs. film vs. participant observation). It will suggest ways to identify participants and sites, and alternative approaches to sampling. It will introduce strategies for assuring trustworthiness and credibility of one’s work. The session will be be useful to anyone in social sciences — to graduate students interested in writing qualitative dissertations as well as participants curious about qualitative research in general.

MAXQDA Hands-On Workshop
This course will cover the capabilities of MAXQDA, a software program that supports qualitative data analysis and helps users systematically code, evaluate, and interpret texts. It is also a powerful tool for developing theories as well as testing theoretical hunches. Its features include coding, memo writing, matrix building, and map building.

MAXQDA Hands-on Workshop — Part 2
This course will cover the advanced capabilities of MAXQDA, a software program that supports qualitative data analysis and helps users systematically code, evaluate, and interpret texts. Survey data, summary matrices, diagrams, merging, and other topics will be presented.

Mixed Methods WorkshopThis TWO-DAY course equips students to design, conduct, and critique mixed method research. From a pragmatic perspective, we will explore the strengths and weaknesses of a variety of data collection methods, and evaluate strategies for combining them. We will focus on mixed method research designs incorporating in-depth interviews, focus groups, participant observation, archival research, survey interviews, and/or hybrid methods. The majority of the course will center on research design and data collection issues. Some time will be spent on strategies for analyzing and presenting data from multiple sources. This course is designed for those who are relatively new to mixed method research, and interested in the principles that should guide it. Participants who come with a specific set of research questions in mind will have opportunities to apply topics of discussion to their own research ideas during the workshop.

NVivo Hands-on Workshop (Part 1)
This session will allow participants to work through a textual document in the PC version of NVivo, a software program for coding textual data such as interviews, focus groups, and field notes. It combines editable text and multimedia capabilities with searching and linking, as well as theory building. Text files can also be linked to graphics or audio files. The program provides an attribute system which can be used for coding demographic variables. It supports visual models, including the specification of types of links between objects in the model.

NVivo Hands-on Workshop (Part 2)
This course will cover analytical features in NVivo 11, including intercoder reliability, charts, and compound queries. We will also cover working with survey data.

NVivo Web Conference
This Web Conference will allow NVivo users to see the latest changes in the recently released NVivo 8.

QDA Miner
Provalis Research will hold this 2-day training workshop on QDA Miner 3.2, a qualitative analysis software package for coding and analyzing textual data. QDA Miner allows users to generate 2D and 3D multidimensional scaling plots. It creates hierarchical clustering of codes and proximity plots; it also computes correlation statistics, comparison statistics, barcharts, and line charts.

Qualitative Interviewing: Asking the Right Questions in the Right Way
Preparing a proper interview guide is only a first step to proper Q & A in a qualitative data collection episode. This session outlines key sections to include in an interview guide and offers suggestions for how to conduct a qualitative interview and/or focus group. The face-to-face interaction in this case is critical. The interviewer must balance attention to the questions designed for the interaction and the emergent topics in the interview. Core skills that focus attention on the audience for the study, the topics of the project, important lines of questioning and goals for ensuring quality interaction in this relationship improve the quality of data collection

Qualitative Research Design
This short course will provide an overview of different ways to design qualitative research projects and how to decide among them. We will address the centrality and interplay of research questions, theory, methods, and budgetary and time limitations for research design. The course will also cover how to craft and portray qualitative research design for grant proposals and articles worthy of publication in competitive journals. Participants are encouraged to bring their own ideas for qualitative research projects to develop further as time allows.

Qualitative Software Comparison Session
This course will briefly look at the differences between MAXQDA 2010, NVivo 9.0, Atlas.ti 6.0, and Dedoose. Other packages will be addressed, including HyperResearch and QDA Miner. No registration is required for this course.

Teamwork in Qualitative Analysis
This seminar, sponsored by the Faculty Working Group in Qualitative Research at UNC, will focus on executing a team approach to qualitative research. Research teams invite multiple perspectives, assist division of labor and provide motivation to continue our work. Each individual on a team brings a set of unique skills, ideas, approaches, strengths and weaknesses. Each individual also brings her or his own personality quirks. These aspects of your coworkers persona become visible over the course of a team project. How can you balance all of these nuanced factors to maintain a team that is interested, driven, and productive? A qualitative team leader must manage four major arenas to successfully navigate a team based project. The four arenas are: the practical, the methodological, the substantive and theoretical, and the interpersonal.

While challenges in any one of these arenas could potential disrupt the project, strengths in one of them can smooth disruption and redirect a team’s efforts. We consider each of these four arenas independently in this course and then discuss how they overlap and how maintaining attention to the status of each one increases the likelihood of a productive and content team. We will use real teamwork examples in this course to address key questions and issues. We ask that you bring your issues, stories and experiences to integrate into our discussion as well.

A Beginner’s Look at Multilevel Analysis
This five-session short course provides an introduction to multilevel modeling for those who have no prior knowledge of the topic. It will also be useful for those with some experience in multilevel data analysis but a desire to develop a stronger theoretical grounding and improved understanding of different multilevel models and computer output. Multilevel modeling is used throughout the social, medical and other sciences to understand how response variables at one level of analysis can be influenced by variables from other levels in a data nested hierarchy. The R programming suite will be used to demonstrate the specification of various multilevel models.

Agent-Based Modeling
This course offers an introduction to a new analytical method of agent-based modeling. It provides an easy way for the beginner to translate research goals into a dynamic model in simulation form. This course offers a step-by step, hands-on, interactive approach to conceptualizing, creating, implementing, and analyzing policy/political science simulation models. This analytical tool can be used in addition to traditional triangulation strategies to operationalize quantitative and qualitative variables (or a combination of both) into a simulation.

The goals of this session is to provide participants with a basic understanding of concepts like agents, emergence, self-organization, and non-linearity. NetLogo, a free downloadable modeling program, will be used to develop elementary skills for direct application of complex systems to policy research. Sample policy/political science problems will be presented and broken down into theory and assumptions in order to outline principles required for a successful simulation. Strategies behind identifying appropriate agent types, behaviors, and attributes, along with establishing system rules, will be presented for model inclusion. The role of energy, learning, feedback, information, scaling, and fitness will be discussed in order to add to modeling dynamics. Participants will have the opportunity to practice modeling policy in flexible and creative ways.

Upon completion of this course participants will be able to access NetLogo and navigate basic areas like the models library of social science, interface, information, and procedures. Additionally, participants will be capable of planning, creating, building, and reporting the basics of a policy simulation analysis. An in-depth course curriculum guide will be supplied for supplemental skill-building and modeling resources.

An Introduction to Conducting Experiments in the Social Sciences
This one-day course will provide a basic introduction to conducting experimental research in the social sciences. With an eye towards pragmatics, this course will teach participants about the experimental life cycle in social science, from research question to experimental design and implementation. Topics covered will include forming theory-grounded hypotheses, an overview of experimental methods, writing & submitting IRB applications, and using Qualtrics online survey software for experimental research.

An Introduction to Designing Experiments
This short course will introduce participants to designing and conducting experiments. We will start with an introduction to experiments. What are the benefits and costs of experiments as an approach to research? What are the different types of experiments, and how can they be used to answer research questions in the social sciences? We will then discuss how to design a successful experiment. This will include discussion of deciding which type of experiment to use, treatment design, measurement concerns, sample selection, getting research permissions, and pre-analysis plans.

An Introduction to Implementing Experiments
This course will cover the basics of implementing an experimental design. It will include selecting a site and sample; piloting or pretesting the instrument; planning logistics for enumeration; training and supervising enumerators; and data management. The course will include time for participants to discuss how to apply these principles to their own experimental designs and contexts.

An Introductory Overview of Structural Equation Models
The purpose of the seminar, which is part of the CPC seminar series, is to give an overview of structural equation models (SEMs). It is not a introduction to SEM software, but focuses more on the conceptual aspects of SEMs. It provides a brief description of the major types of SEMs and the steps involved in modeling. A few hypothetical and empirical examples are included. Participants should have a thorough understanding of regression analysis and a basic understanding of matrices.

Bayesian Statistics: An Introduction for Social Scientists
This short course will be divided into 3 parts. The first part will discuss introductory principles in Bayesian inference, including the Bayesian paradigm, prior elicitation and computational methods. Also, Bayesian methods for linear models and generalized linear models will be discussed in detail.

The second part will examine models Bayesian methods for longitudinal data and survival models, and the third part will examine some special topics such as Bayesian model assessment and missing data. Several applications and case studies will be presented throughout the short course and SAS code as well as WinBUGS code will be provided throughout the course.

Bayesian Workshop
Bayesian statistical methods have become far more common in social science research, yet training in this approach has lagged behind. This two-day seminar provides an overview of Bayesian methods. It begins by reviewing the classical approach to model construction and parameter estimation using maximum likelihood methods. Next, it develops Bayes? Theorem and its extension to probability distributions and modeling. Following this exposition, the workshop covers MCMC methods for parameter estimation, especially the Gibbs sampler and the random walk Metropolis algorithm. Application of these methods to a variety of models including the OLS regression model, generalized linear models (especially the dichotomous and ordinal probit models), hierarchical models, and multivariate models follow. The seminar will be highly-applied with an emphasis on using R and WinBugs for conducting Bayesian analysis. A key focus will be on comparing the Bayesian approach with the classical approach and showing the advantages of using the Bayesian approach in model evaluation and comparison, and inference.

File Organization and Management for Social Science Researchers
Researchers are typically faced with the task of maintaining numerous files related to analyzes from various sources such as data, programs and output. Absent some management plan, file storage systems often result in confusion, replication and wasted time. This course offers strategies for efficient file management to reduce those problems.

Introduction to Structural Equation Models (SEM)
This four-hour short course, offered in two parts, provides an introduction to structural equation models (SEMs) for individuals who have little to no prior experience with the topic. Though no formal pre-requisites are needed, participants are expected to understand the basic tenets of linear regression analysis. This course will focus on the specification, identification, estimation, and evaluation of SEMs, with particular emphasis on the basic terminology, fit indices, and model parameters that allow researchers to test complex research questions. Basic examples of SEM estimation using Mplus will be provided. However, this course is not intended as a hands-on introduction to SEM software. An opportunity for class-wide and/or individualized Q&A will be offered.

Intro to LIMDEP
LIMDEP is statistical computing software developed by William Greene specifically for analysis of limited dependent variables: binary and multinomial choice variables, count variables, and panel data. LIMDEP traditionally has been one of the first packages to include advanced estimators such as multinomial probit and nested logit. I will give several examples of advanced estimation in LIMDEP and show how LIMDEP can be used as an alternative to Stata or SAS.

Logistic Regression
This short course provides an introduction to logistic regression. Model specification, identification, estimation, hypothesis-testing, and interpretation of results are covered. Software to estimate these models is discussed, but not demonstrated. This is not a course on software, but rather a course on the concepts and uses of logistic regression.

Matching, Propensity Scores and Causal Inference in Nonexperimental Settings (2-days, 9-3 p.m.)
In attempting to better estimate the causal impact of choices and behaviors, social scientists are increasingly turning to propensity scores as a means of matching individuals making different choices who are otherwise similar (e.g., teen mothers and other poor young women). This two-day course will begin by briefly reviewing matching as a means of controlling for observed differences between “treated” and untreated cases. It will compare the strengths and weaknesses of these methods relative to alternatives, such as regression. The course then will present the statistical justification for propensity scores and outline their advantages and disadvantages. The second half of the course will involve an application of propensity score methodology. Students will be introduced to R and the package -matchit-. R is comprehensive statistical analysis software available for free; -matchit- is a package that provides a flexible set of tools for matching cases based on the propensity score. Having preprocessed the data with -matchit-, one is free to analyze the data in any statistical package, including STATA and SAS.

Missing Data Using SAS PROC MI and MI Analyze
This course will first discuss the usual methods for dealing with missing values, such as listwise deletion and simple mean imputation and the consequences of each. We will move to regression model imputation without and with methods that compute variance estimates utilizing variance induced but the imputations. A lab session will be included to demonstrate software available for imputation and estimation using sythesized data.

Multidimensional Scaling
This short course will provide a basic introduction to multidimensional scaling. Specific topics to be covered include: The basic idea of MDS; types of data that might be input to MDS; the general estimation procedure; interpretation of results; different varieties of MDS; and software options for performing MDS analyses. The course will meet for two sessions. The first meeting will occur in the classroom, to introduce the basic concepts and ideas. The second meeting will occur in the computer laboratory, to give course participants hands-on experience with the MDS routines in one or more of the major statistical software packages.

This short course is intended for a general audience. It does not assume any prior experience with MDS or familiarity with advanced statistical methods (i.e., beyond basic regression analysis). Participants should be able to perform basic data processing tasks with a statistical software package (e.g., Stata, SAS, SPSS), but no special knowledge of MDS software is assumed or necessary.
Regression Diagnostics (parts 1 and 2)

Regression Diagnostics part 1 of 2: Outliers and Influential Cases. Outliers and influential cases are terms used to describe observations in a data set that may have undue influence upon the results of an analysis. This course will describe techniques to detect those cases, assess their influence, and suggest methods for dealing with them. The course will primarily focus on the practical aspect of performing these tests using SAS, but some statistical material is included. Familiarity with linear regression and its assumptions is required. Regression Diagnostics part 2 of 2: Regression Assumptions. This session on regression diagnostics will cover regression assumptions, introduce heteroscedasticity and specification tests to assess whether these assumptions hold, and recommend techniques for correcting assumption violations. The course will primarily focus on the practical aspect of performing these tests, but some statistical material is included. Familiarity with linear regression and its assumptions is required.

Social Network Analysis
Network analysis focuses on relationships between or among social entities. It is used widely in the social and behavioral sciences, as well as in political science, economics, organizational studies, behavioral biology, and industrial engineering. The social network perspective, which will be taught in this workshop, has been developed over the last sixty years by researchers in psychology, sociology, and anthropology. The social network paradigm is gaining recognition in the social and behavioral sciences as the theoretical basis for examining social structures. This basis has been clearly defined and the paradigm convincingly applied to important substantive problems. However, the paradigm requires concepts and analytic tools beyond those provided by standard quantitative (particularly, statistical) methods. This five day workshop covers those concepts and tools. The course will present an introduction to concepts, methods, and applications of social network analysis drawn from the social and behavioral sciences. The primary focus of these methods is the analysis of relational data measured on groups of social actors.

Topics include an introduction to graph theory and the use of directed graphs to study actor interrelations; structural and locational properties of actors, such as centrality, prestige, and prominence; subgroups and cliques; equivalence of actors, including structural equivalence, blockmodels, and an introduction to relational algebras; an introduction to local analyses, including dyadic and triadic analyses; and an introduction to statistical analyses, using models such as p1 and exponential random graph models. The workshop will use several common software packages for network analysis: UCINET, Pajek, NetDraw, and STOCNET.

Structural Equation Models and Latent Variables: An Introduction
This course will introduce participants to Structural Equation Models (SEMs) with and without latent variables. It provides an overview of the statistical theory underlying SEMs and practice with SEM computer software. The topics include path analysis, confirmatory factor analysis, simultaneous equation models (recursive and nonrecursive), incorporating multiple indicators and measurement error into models, alternative estimators, model identification, and assessing model fit. The course does not require prior experience with LISREL or other SEM software. Participants should have strong backgrounds in matrix algebra, regression analysis, and factor analysis.

The Statistical Analysis of Cost-Effectiveness Data
This mini-course will discuss and illustrate appropriate statistical methods for the analyses of cost-effectiveness data. Traditional cost-effectiveness analyses have focused on the incremental cost-effectiveness ratio. As is well known, however, ratios have strange statistical properties. For that reason, economists have turned to an alternative, the cost-effectiveness acceptability curve (CEAC). This course will introduce the conceptual basis for CEAC and will illustrate their use with STATA. Students will learn the basics of bootstrapping and graphing within STATA.

Address-Based Sampling as an Alternative to RDDDeveloping cost effective methods for reaching households which no longer have a landline but do have access to a cell phone, so called cell phone only households (CPOs), is a critical item on the agenda of most survey and market researchers. Concerns about sample coverage and data biases resulting from the exclusion of CPOs have increased as the penetration of CPO households continues to climb, exceeding 50% in some subgroups. To date, two methodologies have emerged as potential means for addressing this issue. The first involves sampling telephone numbers from known cell phone exchanges and calling these numbers or combining these with a sample of landline numbers in a dual frame design. An alternative approach involves sampling of addresses rather than telephone numbers. Address based sampling (ABS) is a new technique built upon the relatively recent availability of large scale address databases, which can provide as much as 98% coverage of U.S. households. Further, these addresses can be reverse-matched to commercially available databases to identify a relatively large proportion of telephone numbers, facilitating the use of mixed-mode approaches. In this course, we delineate and compare the advantages and limitations of these two approaches, including discussion of sampling and weighting approaches, operational considerations, timeliness, and cost. We draw examples from several recently conducted studies employing each of these approaches. The findings from this research help bring into sharper focus the potential alternatives to traditional random digit dialed (RDD) telephone surveys for conducting cost effective data collection of the general public.

Adjusting for Item Nonresponse
Analysis of data that is not complete due to non-response or refusal to respond to selected questions is a common problem. Inferences based on the analysis that restricts the sample to those with no missing data on relevant analytical variables may be biased. Over the last few years a variety of methods and software have become available to analyze incomplete data.

Obviously, since the data are missing all these approaches involve assumptions about the relationship between the observed and missing portions of the potential complete data. This course will provide an overview of issues needing attention when choosing methods for analyzing incomplete data. The primary focus will be on weighting and imputation based approaches as they are more widely applicable and software for implementing them has become available in recent years. This course will emphasize applications and heuristic theory. Illustrative examples using actual and simulated data sets will be provided along with the computer code. As a prerequisite, the students should have taken quantitative course and should be familiar with linear and logistic regression analysis.

Advanced Topics in Qualtrics
This course will cover topics which are beyond the scope of the Introductory course, including complex routing (“survey flow”), smart text (“piping”), randomization, authentication, calculations, and various types of customization using html and javascript.

Analysis Procedures for Cognitive Interviews
This half-day course is intended for individuals who have attended “Cognitive Interviewing: A Hands on Approach” (on October 8, 2015 or previously) or have otherwise gained knowledge/experience in cognitive interviewing. The course will focus on analysis of the data obtained from cognitive interviews, an important but undeveloped area in cognitive testing. Dr. Willis will excerpt from his recent book “Analysis of the Cognitive Interview in Questionnaire Design,” to cover the following topics: (a) defining the unit of analysis; (b) aggregating results across interviews, interviewers, and testing organizations; (c) the use of text summaries versus coding schemes; (d) the degree of quantification appropriate for qualitative testing results; and (e) the use of specialized software for the various analysis steps. Finally, he will describe the use of the Cognitive Interviewing Reporting Format (CIRF) to guide the write-up of cognitive interviewing projects for clients and for publication, and the use of the online Q-Bank database for accessing and storing reports.

An Overview of Topics in “Big Data”: Unpacking Data Science for Beginners
This course is a one-day introduction to “Big Data” as method of conducting research. The course will cover a range of issues, including: • Characteristics of data that is collected through these techniques. For example, when is scale of data important, vs. the nonreactive nature of the data. • Common methods for obtaining datasets for “Big Data” • Epistemological approaches for using data, including the inductive nature of many data analytic techniques. • Comparison of data analytic techniques with other forms of research. • Exploration of a variety of tools that are commonly used in Big Data research. • Common analytical techniques in data science. People who take this course will be able to define the pros and cons of data science as a research method, understand common terms related to Big Data techniques, and identify research questions that are appropriate to these techniques. It’s impossible to give a very technical training in a one day class, so while we’ll cover where one can go to learn more, this class will not delve deeply into technical aspects of big data. Given the nature of the instructor’s research, the class will focus on data mined from social media sites, which is one of the most common sources for data analytic approaches. Any person with a solid background in research methods will benefit from this course.
Balancing Data Confidentiality & Data Quality
Statistical summary data such as tabulations are built from data pertaining to individual entities (persons, households, businesses, organizations or groups). Statistical microdata are unit-record data containing multiple item responses pertaining to individual entities. Statistical data base query systems, once only a possibility, are becoming a reality. The need for data products that combine information across data bases and organizations is increasing. The data from which all of these statistical data products are built typically is reported at the individual entity level and often is confidential. Standard ethical survey practice demands that confidential data pertaining to individuals or entities not be revealed through released statistical data. Confidentiality concerns have been addressed by researchers and government statisticians over several decades, resulting in a suite of increasingly sophisticated and effective survey methods for statistical disclosure limitation (SDL), several of which have been implemented in software and incorporated within the survey practices of government statistical agencies in the U.S. and abroad. Until very recently, however, the effects of disclosure limitation methods on data quality and usability have been largely ignored. The interplay between data confidentiality and data quality is a central subject of this course. This course has three objectives: (1) to familiarize the student with statistical disclosure limitation and SDL methods; (2) to examine potential effects of SDL methods on data usability and quality; and, (3) to present SDL methods that, in addition to protecting confidentiality effectively, control deterioration in the usability, quality and completeness of the final data product. Practical data quality questions include: What effect does the SDL method have on key statistics? What effect does the SDL method have on the distribution of the original data?

Cognitive Interviewing: A Hands-On Approach (Two day)
Cognitive interviewing has become a very popular method for pretesting and evaluating survey questionnaires. The current approach favored by Federal laboratories and private research institutions mainly emphasizes the use of intensive verbal probes that are administered by specially trained interviewers to volunteer respondents, often in a laboratory environment, to delve into the cognitive and socio-cultural processes associated with answering survey questions. Based on this information, the evaluator makes judgments about where questions may produce difficulties in a number of subtle ways, due to cognitive demands they impose, cultural mismatches, or other shortcomings. The short-course will focus on the specifics of how to conduct verbal probing, and how to process and communicate the results obtained. Although an introduction to theory and background perspective is included, the course will focus on the application and practice of cognitive interviewing techniques, as these are targeted toward both interviewer-administered (face-to-face or telephone) and self-administered (paper and computer) surveys. Participants will practice the conduct of cognitive interviews across modes, and will evaluate their results by judging where questions have failed, and what one might do to revise them. The course aims to provide a working familiarity with cognitive techniques, so that students will be able to begin conducting cognitive interviews on their own.

Conducting Cross Cultural Surveys – Challenges and Best Practices
This course will provide an introduction to survey research methods for designing multinational and multicultural surveys. It begins with an overview of the field of comparative surveys. This will summarize their history and discuss some unique design features and implementation challenges inherent in their design and implementation. The second section discusses quality and risk management frameworks for comparative surveys. It will present some tools for monitoring quality processes and outcomes, and will reference the new Guidelines for Comparative Surveys (ccsg.isr.umich.edu). The third section of the course focuses on issues in study design, considering organizational structure, data collection infrastructure and management, and cost and quality tradeoffs. The second half of the course addresses instrument design for comparative surveys. It opens with a discussion of issues in defining objectives, identifying constructs, developing questions, and monitoring design process quality that are particular to the field of comparative surveys. It will also cover some technical challenges in crafting the questions into a survey instrument; visual display of text in various languages, placement of response categories and instructions, use of color, screen density, and other features of contemporary survey instruments will be reviewed from a multilingual and multicultural context. The links between design and mode considerations are also covered. The course concludes with a module on question adaptation and translation focusing on the critical role that version production often plays. Examples will be drawn from demographic and social indicator surveys, attitudinal surveys, health and education surveys, and quality of life surveys.

Conducting Surveys in Developing Countries
Cross-border and multi-country survey research is used by a wide variety of government and non-governmental organizations around the world. Conducting surveys across cultures, languages and polities presents an array of unique challenges. These challenges, and as well as simple processes of survey logistics and data quality control, take on new levels of difficulty as the educational and technical sophistication of the target populations and their environments become less developed and stable. This short course will expose attendees to some normative approaches to the challenges of conducting successful survey research in these places amongst the majority of the world’s people.

Cross Cultural Survey Methods (Half day)
The course will focus on measurement error in cross-cultural surveys. It will begin with a theoretical background for cross-cultural differences drawing on cross-cultural psychology, present relevant psycholinguistic theories to demonstrate how language can influence the response formation process and discuss issues related to translation. The course will briefly touch on other sources of error.

Data Collection Methods
This course will present research work which attempts to understand the effect of data collection decisions on survey errors. This is not a “how -to-do-it” course on data collection, but instead presents material that reviews effects of survey design decisions on data quality. This course is designed to sensitize students to alternative design decisions and their impact on the data obtained from surveys. The course will review alternative modes and methods of data collection used in surveys. It concentrates on the impact modes of data collection have on the quality of survey data, including measurement error properties, levels of nonresponse, and coverage error. Methods of data collection will focus on advances in computer assisted methodology and comparisons among various methods (e.g. telephone versus face to face, paper versus computer assisted, interviewer administered versus self-administered). The statistical and social science literature on interviewer effects will also be examined, including literature related to the training and evaluation of interviewers. With respect to nonresponse, we will review current literature on the reduction of nonresponse and the impact of nonresponse on survey estimates.

Data Collection Using Mobile Phones in Developing Countries: New Approaches with SMS, IVR, and CATI
The rapid growth of mobile phones in developing countries opens up new possibilities for data collection. Short message service (SMS), interactive voice response (IVR), online surveys (web) and computer-assisted telephone interviewing (CATI) can produce data faster and less expensively than face-to-face surveys. This course will introduce students to the design and implementation of SMS, IVR, Web, and CATI surveys in low- and middle-income countries. In this course, Dr. Lau will draw from real world examples to illustrate how these modes work. We will also discuss basic survey design principles in each mode, focusing on sampling, questionnaire development, and survey design. New for this year’s course: Students will work in small groups to design questionnaires and collect data in real time via SMS and IVR using a free, open source survey tool.

Designing and Conducting Surveys of Businesses and Organizations
Surveys of businesses and organizations differ in important ways from surveys of individual persons and households. In particular, they rely on one or more employees or representatives of the organization to report data about the entity on its behalf. As a result, a respondent’s approach to the survey is influenced by organizational characteristics, which provide a context within which the response process occurs, affecting survey participation decisions, data quality, and response burden. Practical issues emerge that have implications for the effectiveness of data collection instruments, procedures and strategies in the organizational setting, such as who decides whether to participate in the survey, who is the “right” respondent, are the desired data in records, are the data accessible, and so on. This course provides an overview of methodological issues associated with the use of surveys to collect data from organizations. We will identify key differences between household surveys and organizational surveys, emphasizing organizational behaviors and attributes that affect survey response. We will demonstrate an approach to survey design that utilizes understanding and consideration of this organizational context when developing, adapting, and implementing data collection instruments and procedures. This course will include topics related to survey planning, questionnaire design and pretesting, data collection modes, and communication and response improvement strategies.

Designing Effective Web Surveys
This course will illustrate the appropriate use of web tools (such as radio buttons, check boxes, slider bars), and the use of images, screen layout and other aspects of the user interface which affect accuracy of survey results. The course will not address web survey software or programming; the principles to be discussed are independent of any single software package.

Designing Good Survey Questions
This basic course provides practical guidance in writing closed-ended items for standardized survey research. Using many examples, we discuss what makes a good survey question and how to avoid common pitfalls. We focus on developing unbiased questions with appropriate response scales which can be understood in consistent ways by all respondents. We also discuss various methods for pretesting survey questions.

Designing Multi-Item Scales
This course provides in introduction to developing instruments with multiple items to measure a single construct. Examples include measures of various social and psychological variables that might be assessed in health, marketing, journalism, or other research areas. Participants will also be encouraged to suggest content areas for discussion. After a brief theoretical introduction, we will turn to practical issues such as when a multi-item scale is (or isn’t) appropriate, determining the number and content of items in the scale, what type and how many response options should be offered, whether scales should include both “negative” and “positive” items, whether the parts of a subscale should be grouped or scattered, and other common concerns in scale development. Dr. DeVellis will use real-life examples to demonstrate the scale development process. There will be ample opportunity for questions and discussion.

Designing Self-Administered Questionnaires
Do you have missing, incomplete, or inconsistent data from your self-administered paper surveys? Are respondents reluctant or unwilling to fill out your forms? A few design principles for visual appeal and ease of completion can improve your data quality and your response rate. Attend this workshop to learn guidelines for creating effective paper-and-pencil data collection instruments.

Designing Web Surveys
The focus of this course is on the design of Web survey instruments. The course will focus on the appropriate choice and design of input tools (e.g., radio buttons, check boxes, drop boxes, text fields), including new features in HTML 5 and additional tools such as sliders. The course will also address layout and formatting of the instrument, including alignment of questions and response options, typeface, background color, and the design of grids or matrix questions. The design implications of browser-based mobile Web surveys will also be addressed. The course will draw on empirical results from experiments on alternative design approaches as well as on practical experience in the design and implementation of Web surveys. The course will not address the technical aspects of Web survey implementation (such as hardware, software, or programming) and will also not focus on question wording, sampling, or recruitment issues. The course will equip participants with the knowledge needed to make appropriate Web survey instrument design choices.

Evaluation of Survey Data
This course will span a range of topics dealing with the evaluation of survey data quality. The topics that will be discussed include: survey pretest methods (expert review, cognitive interviewing, non-interview methods and debriefing methods), methods used concurrently with data collection (monitoring, debriefing, and behavior coding) and post-survey methods (reinterview or followback methods, record check studies, external and internal validation studies and non-randomized observational studies). Utility of each method and examples of their use will be given.

Executing Your Survey Research Project
This workshop series will provide guidance to participants conducting survey research for their dissertation, thesis, or other project. Each week the workshop will focus on a topic and provide instruction, group discussion and an opportunity for participants to complete a worksheet or review handouts. The worksheets and handouts are tangible products that will help guide participants to execute their survey research. The series will provide information that can be applied to web or paper surveys. There are no prerequisites to this workshop series and participants are encouraged to bring any materials they have already developed for their project.

  • Workshop 1, Sept. 14: Creating a Timeline for Success and a Data Analysis Plan A comprehensive timeline helps the researcher set short and long-term goals to keep the project on track. The data analysis plan drives key decisions in survey design and implementation and should be developed at the beginning of the project. In this session participants will learn about important elements of these items and draft, review, and/or receive feedback on their own timeline and data analysis plan.
  • Workshop 2, Sept. 21: Participant Engagement and Data Management Plans This session focuses on ways to engage your participants to yield higher response rates and procedures to collect and manage survey data. How do the words and materials you use to recruit respondents affect cooperation rate and data quality? What options are available to recruit participants? What are the pros and cons of different contact and data collection modes (e.g. email, postal letter, phone calls, text messages. How will you store and manage your data once they are collected? In this session participants will draft, review, and/or receive feedback on their data collection protocol and data management plan.
  • Workshop 3, Sept. 28: Questionnaire Development
Good question development is the heart of a quality survey. We will review principles for question development, what should be considered when moving from a paper to a web survey, and tips for using previously designed scales. Participants will draft, review, and/or receive feedback on their survey questions.
  • Workshop 4, Oct. 5: Qualtrics Overview Qualtrics is an online survey software program available free of charge to UNC students, faculty, and staff. A free trial version is available to the public. This session covers programming surveys, assigning variable names and code numbers, distributing survey invitations and reminders, and exporting data for analysis. Participants will also receive an overview of advanced Qualtrics functionalities.
  • Workshop 5, Oct. 12: Paper Surveys and Testing Surveys (web and paper) There is more to designing an effective paper survey than typing up your questions and selecting ‘Print.’ This session provides a brief tutorial on designing and distributing paper surveys, followed by discussion of the important steps of testing surveys. We will review testing methods including cognitive interviewing, usability testing, observation, data review, and piloting. Participants will draft, review, and/or receive feedback on their plans for testing their surveys.
  • Workshop 6, Oct. 19: Institutional Review Board and Protecting Human Subjects
Most survey research in an academic setting requires approval from an Institutional Review Board (IRB) whose purpose is to protect the rights and well-being of research subjects. This session quickly reviews the purpose and function of an IRB and the UNC ethics training requirements for researchers before turning to the nuts and bolts of completing the UNC IRB Application for a survey project. Participants will receive detailed guidance on completing their own IRB application in ways that maximize efficiency and minimize processing delays.
  • Workshop 7, Oct. 26: Preparing Data for Analysis and Archiving This session focuses on data cleaning, preparation for analysis and archiving survey data for long-term preservation. Participants will learn how to keep a log for data that need to be cleaned, how to prepare data for commonly used analysis packages, and options for archiving data.

Experimental Design for Surveys
A key tool of methodological research is the split-ballot experiment, in which randomly selected subgroups of a sample receive different questions, different response formats, or different modes of data collection. In theory, such experiments can combine the clarity of experimental designs with the inferential power of representative samples. All too often, though, such experiments use flawed designs that leave serious doubts about the meaning or generalizability of the findings. The purpose of this course is to consider the issues involved in the design and analysis of data from experiments embedded in surveys. It covers the purposes of experiments in surveys, examines several classic survey experiments in detail, and takes a close look at some of the pitfalls and issues in the design of such studies. These pitfalls include problems (such as the confounding of the experimental variables) that jeopardize the comparability of the experimental groups, problems (such as non-response) that cast doubts on the generality of the results, and problems in determining the reliability of the results. The course will also consider some of the design decisions that almost always arise in planning experiments issues such as identifying the appropriate error term for significance tests and including necessary comparison groups.

Focus Groups as a Tool in the Development, Implementation, and Evaluation of Quantitative Social Science Research
This course will present an overview of how to conduct groups and address how they can assist in the design of social science research projects. Topics will include how focus groups can be used to aid in development of appropriate methodologies, samples, and questionnaire design for survey projects. We will also discuss how focus groups can help us to better understand or elaborate on the quantitative findings or to investigate why people answered as they did.

Inferential Issues in Web Surveys
There are many different ways that samples can be obtained for online surveys. These include open invitation surveys of volunteers, intercept surveys, opt-in or access panels, Amazon’s Mechanical Turk, Google Consumer Surveys, list-based samples, and the like. In most cases, the goal is to make inference to some large population. The different approaches to selecting samples and inviting respondents to complete a survey vary in their inferential properties. Threats to inference include sampling error, coverage error, and non-response error. In addition to selection methods, a variety of adjustment methods, such as weighting, propensity score adjustment and matching, are being used to mitigate the risk of inferential errors. The course will focus on the assumptions behind the different approaches to inference in Web surveys, the benefits and risks inherent in the different approaches, and the appropriate use of a particular approach to sample selection in Web surveys. The course has a conceptual rather than statistical focus, but a basic understanding of statistics will be helpful. This course is suitable for people who are considering conducting a Web survey for data collection or analyzing data from an existing Web survey.

Introduction to Cognitive Interviewing
This course will provide an overview of cognitive interviewing as a technique for developing and/or testing survey questions. We will briefly discuss participant recruitment and other planning details before turning to development of an interview guide, discussion of think-aloud and probing techniques, selection of probes, trade-offs of concurrent vs. retrospective probing, and how to choose the best techniques for particular situations. We will use demonstrations and exercises to give participants experience using the technique.

Introduction to Focus Groups
Focus group interviews are commonly used for survey development, content development, and qualitative data collection to capture rich information about attitudes and beliefs that affect behavior. An overview of the basics of focus groups supplemented with real examples and hands-on practice will highlight the most appropriate uses of focus groups, moderating focus groups, developing interview questions, analyzing and using results, as well as reporting findings.

Introduction to Longitudinal Analysis
This course will review various models often used to analyze longitudinal data including generalized linear (for example, logit) models. Subject-specific versus population- averaged interpretations will be discussed. Detailed attention will be given to marginal models for the population-averaged response. Implications of survey clustering and weighting on generalized estimating equation (GEE) methods will be described. Robust variance estimation using Taylor series linearization, jackknife, and balance repeated replication (BRR) will also be presented.

Introduction to Qualtrics
This is an introductory course in using the Qualtrics web-based survey system to develop and conduct an online survey. Topics will include basic survey creation, customization and distribution. We will demonstrate how to create a mailing list for inviting respondents to the survey, and how to avoid the most common pitfalls that new users make. We will demonstrate how to download collected data into an Excel spreadsheet or SPSS dataset. The course will not cover use of the online analysis tools within Qualtrics.

Introduction to Survey Management
This course will focus on the application of project management principles and techniques to the management of survey research projects. At the conclusion of the course participants will have a basic understanding of: The principles of project management as applied to survey research; How to plan a survey project; How to implement the plan and manage the work; How to manage the project budget; How to manage the project contract. The course will cover a broad range of survey management topics, including: proposal preparation, Work Breakdown Structures, Gantt charts, organization charts, staffing, budgeting, management tools to monitor the work, and types of survey contracts. Course participants will receive a workbook containing all material presented in class.

Introduction to Survey Quality
The course spans a range of topics dealing with the quality of data collected through the survey process. The course begins with a discussion of total survey error and its relationship to survey costs and provides a number of measures of quality that will be used throughout the course. Then the major sources of survey error are discussed in some detail. In particular, we examine a) the origins of each error source (i.e., its root causes), b) the most successful methods that have been proposed for reducing the error emanating from these error sources, and c) methods that are most often used in practice for evaluating the effects of the source on total survey error. The course is not designed to provide an in-depth study of any topic but rather as an introduction to the field of survey data quality. The purposes of the course are to provide an overview of the basic principles and concepts of survey quality with particular emphasis on the components of sampling and nonsampling error, to develop the background for the continued study of survey measurement quality through readings in the literature on survey methodology and to identify issues related to the improvement of the survey quality that are encountered in survey work and provide a basic foundation for resolving them.

Introduction to Survey Weighting
This course is an introduction to the basic concepts for weighting. It begins by defining the goals of weighting including weighting as a correction for differential selection probabilities, non-response and non-coverage. The course covers the process of developing weights for stratified two-stage sampling including computing design weights and methods for nonresponse adjustments and frame coverage error adjustments. Additional topics may include the effect of weighting on variance of the estimates and extreme weights.

Issues in the Analysis of Complex Sample Survey Data
This one-day short course will provide participants with an introductory overview of issues frequently encountered when conducting secondary computer analyses of data collected from sample surveys with complex multi-stage designs (e.g., PSID, NHANES, NCS), including design-based weight determination, software choice, and proper analysis methods. The workshop is not intended for participants looking to design a survey, but rather for participants who have a desire to analyze complex sample survey data.

Latent Class Analysis of Survey ErrorThis course presents a statistical framework for modeling and estimating classification error in surveys. It begins by examining some of the early models for survey measurement error (Census Bureau models; Kish model; etc.) and demonstrating their similarities, strengths and weaknesses. Then these models are cast in a general latent class modeling (LCM) framework where the true values of a variable are assumed to be unobserved (latent) and the survey response constitutes a single indicator of the latent variable. The model parameters include the target population proportions for a categorical variable to be estimated in the survey and the probabilities of misclassification probabilities (for e.g., false positive and false negative, for dichotomous response variables) for measuring the variable. Survey item reliability and construct validity as well as estimator bias are defined and interpreted in this context. Methods for estimating the model parameters and issues of model identifiability will be discussed. One advantage of viewing survey classification error model as a LCM is the availability of general software for estimating the error components. However, the assumptions of the traditional LCM can be somewhat restrictive. An even more general model can be obtained by viewing the LCM as a type of log linear model with latent variables. In doing so, a wide range of error structures and error evaluation designs can easily be discussed and analyzed using log-linear modeling notation and methods. A number of examples and illustrations will be presented to demonstrate the estimation methods and the interpretation of the latent class analysis results. The utility of the models for evaluating and improving survey data quality will also be discuss and demonstrated.

Mixed Method ResearchThis course equips students to design, conduct, and critique mixed method research. From a pragmatic perspective, we will explore the strengths and weaknesses of a variety of data collection methods, and evaluate strategies for combining them. We will focus on mixed method research designs incorporating in-depth interviews, focus groups, participant observation, archival research, survey interviews, experiments and/or hybrid methods. The majority of the workshop will center on research design and data collection issues with some time devoted to strategies for analyzing and presenting data coming from multiple methods. This course is designed for those who are relatively new to mixed method research and interested in the principles that should guide it. Participants are encouraged to come with a specific set of research questions in mind, or projects in development, to use as examples during workshop discussions and activities.

Mixed Mode
Given the declining response rates and increasing cost of surveys over the past several years, researchers are increasingly turning to mixed-mode surveys. Mixed-mode surveys also offer the promise of reducing measurement error or improving data quality for certain types of items. Specifically, mixed-mode surveys attempt to combine the benefits of reduced cost and improved data quality associated with self-administered methods (e.g., Web and mail) with the improved representational qualities (especially coverage and nonresponse error reduction) associated with interviewer-administered methods (face-to-face or telephone). A wide variety of different mixed-mode methods have been evaluated and deployed. This course offers an overview of the different approaches to mixed-mode survey design, presents a summary of what is known and not known about mixing survey modes, and offers a set of guidelines to help the researcher make decisions about mixing modes of data collection.

Multi-Item Scales
This course is designed to inspire participants from all disciplines that it is possible to write your own high-quality multi-item scale. This course offers an introduction on how to do this. It covers: The advantages and disadvantages of the use of multi-item scales An introduction to psychometric principles How to design a multi-item scale within these principles An introduction to some basic statistical tools for assessing the quality of a multi-item scale (i.e., use of Cronbach’s alpha to measure reliability and a basic exploratory factor analysis to measure dimensionality) The day will be a combination of lecture and workshops in designing multi-item scales and critiquing existing ones. We will also spend 1.5 hours towards the end of the day in the computer lab practicing applying the basic statistical tools to the evaluation of some pre-existing scales.

Multilevel Analysis for Grouped and Longitudinal Data
Social research often concerns relationships between individuals and the social contexts to which they belong. Individuals and their social contexts can be conceptualized as a hierarchical structure, with individuals nested within groups. Classical examples are organizational research, with individuals nested within organizational units, and cross-national research, with individuals nested within their national units. Such systems can be observed at two levels, and as a result we have data with group level variables and individual level variables. To analyze such hierarchical structures, we need multilevel modeling, which allows us to study the relationships between variables observed at different levels in the hierarchical structure. Multilevel modeling can also be used to analyze data from longitudinal research, by viewing measurement occasions as being nested within respondents. This has several advantages compared to more classical approaches to longitudinal data. This short course is intended as a basic and nontechnical introduction to multilevel analysis. It starts with a description of some examples, and shows why multilevel models are necessary if the data have a hierarchical structure. It then covers the basic theory of two- and three-level models. Next it explains how multilevel models can be applied to analyzing longitudinal data, and why and when this may be an attractive analysis approach, as compared to more classical analysis methods such as multivariate analysis of variance (Manova). Examples shall be given on how to conduct these analyses using the SPSS Mixed procedure, which is available in SPSS starting with version 11.5. The course assumes reasonable familiarity with analysis of variance and multiple regression analysis, but prior knowledge of multilevel modeling is not assumed.

Multiple Imputation: Methods and Applications
Multiple imputation offers a general purpose framework for handling missing data, protecting confidential public use data, and adjusting for measurement errors. These issues are frequently encountered by organizations that disseminate data to others, as well as by individual researchers. Participants in this workshop will learn how multiple imputation can solve problems in these areas, and they will gain a conceptual and practical basis for applying multiple imputation in their statistical work. Topics include the pros and cons of various solutions for handling missing data; the motivation for and general idea behind multiple imputation; methods for implementing multiple imputation including multivariate modeling, conditional modeling, and machine learning based approaches; methods for checking the adequacy of imputations via graphical display and posterior predictive checks; and applications of multiple imputation for scenarios other than missing data.

New Technologies in Surveys
Rapid advancements in communications and database technologies are changing the societal landscape across which public opinion and survey researchers operate. In particular, the ways in which people both access and share information about attitudes, opinions, and behaviors have gone through perhaps a greater transformation in the last decade than in any previous point in history and this trend appears likely to continue. This course examines some of the research findings to date with respect to the use of mobile and social media platforms as vehicles for collecting information on attitudes, opinions and behaviors. For each area, we will explore current applications, known best practices, and cautions, including smartphones (for surveys, GPS, and visual data collection) and social network platforms (surveys and other forms of information). Examples will be provided from several topic areas, including assessment of political attitudes, health-related studies, and consumer research. The final section of the course delineates some of the more fruitful areas for on-going research to improve our understanding of these technologies and the role they can play in assessing public opinion.

Nonresponse Analysis
This course has the objective of introducing participants to methods for measurement of nonresponse bias. The conceptualization of nonresponse bias and causal models are briefly reviewed. The focus of this course is the presentation of methods and findings from the research literature on the measurement of nonresponse bias, classified by the source and type of auxiliary information used to estimate bias. A very brief preamble to approaches for the reduction and the adjustment of bias due to unit nonresponse is provided. Prior knowledge about data collection methods in surveys is beneficial. This course does not intend to provide technical instruction on how to estimate nonresponse bias.

Nonresponse from the Total Survey Error Perspective: An Overview
The Total Survey Error (TSE) paradigm embodies the best principles, strategies, and approaches for minimizing the survey error from all sources within time, costs, and other constraints that can be imposed on the survey. This approach can be viewed as resting on the four pillars of survey methodology: survey design, implementation, evaluation, and data analysis. This course provides an overview of the TSE paradigm as it applies to one critical source of error: nonresponse. Structured around these four pillars, the course presents the best methods and lessons learned for dealing with nonresponse in survey, data collection, data analysis and evaluation. The survey focuses particularly on the interactions of response mechanism with other error sources and how nonresponse interventions can lead to unintended consequences for TSE.

Practical Approaches to Web Survey Implementation
Practical Approaches to Web Survey Implementation Are you interested in learning how the Web can be used effectively in survey research? Have you wanted to see how Web surveys are actually implemented? In this course, we will trace the process that begins with a questionnaire and follow it through to survey implementation and management. Topics will include: Design implications on survey quality; Interactive, navigation, and screen design features in Web surveys; Minimizing error through standards and testing; Respondent contact strategies and sample management; Security and confidentiality issues; New innovations in Web survey design and implementation; Web survey software/service options and selection criteria. This course will be taught with the survey practitioner/manager in mind and will be very useful for graduate students or researchers desiring to learn more about Web surveys. Knowledge of complex survey methods or computer programming is not required. Participants in this course should have a basic understanding for the practice of survey research, and some experience in using the Web. The course will not be tied to a specific web survey product or service, however, the presenters will share their experiences with several products available.

Psychology of Survey Response
This course examines survey questions from a psychological perspective. It describes the major psychological components of the response process, including comprehension of the questions, retrieval of information from memory, combining and supplementing information from memory through judgment and inference, and the reporting of an answer. It discusses several models of how respondents answer questions in surveys, reviews the relevant psychological and survey literatures, and traces out the implications of these theories and findings for survey practice, especially for the design of questionnaires.

Questionnaire Design
This course focuses on the design of questionnaires used in survey research, exploring the theoretical issues that arise in their development, application and interpretation as well as the practical aspects of questionnaire design that are often not taught in formal courses. It involves lectures as well as a variety of hands on exercises. The emphasis is on the selection of appropriate measurement techniques for assessing both factual and non-factual material using survey questions. Topics include: cognitive guidelines for question construction to ensure respondent understanding; techniques for measuring the occurrence of past behaviours and events; the effects of question wording, response formats, and question sequence on responses; combining individual questions into a meaningful questionnaire; special guidelines for self-completion surveys versus interview surveys; strategies for obtaining sensitive or personal information; an introduction to the various methods to test questionnaires.

Response Rate Issues in Telephone Surveys
This course addresses survey design and implementation decisions that affect response rates in telephone surveys. Topics to be covered include: advance letters, incentives, survey introductions and refusal conversions, dealing with cell phones and answering machines, and ramifications of the federal Do Not Call list. The presentation will summarize methodological research on telephone survey response rates and provide broad guidance on when response rates matter most to survey quality.

Social Media’s Role in Survey Research
The ubiquity of social media in the world today presents new opportunities and challenges when it comes to social research. This course considers the use of social media in survey research. Throughout the survey lifecycle (questionnaire design and testing, subject recruitment, respondent tracking and longitudinal panel retention), social media platforms offer some new ways to reach respondents at a time when traditional methods have seen declining participation. Social media data can also be considered as supplementary or proxy data for surveys. This course will present specific examples of the use social media in survey research, highlighting the topics, methods, and ethical considerations that accompany this growing sub-discipline. We end with considerations for the role of social media in public opinion research in the future as this area of research evolves.

Subjective Measurement in Surveys
For 100 years, a huge amount of social science research has been done using structured questionnaires. During most of this time, the wording and ordering of questions has traditionally been viewed as “an art, not a science.” This course will challenge that perspective and make the case that the accumulation of a great deal of knowledge throughout the social sciences about effective question-asking does indeed offer a basis for recommendations about how best to measure subjective phenomena. The course will expose participants to cutting-edge techniques for questionnaire design, for guarding against measurement artifacts, and for analyzing data in order to overcome the biasing impact of such artifacts. Second, the course will raise awareness among participants about many design issues regarding which there are not simple solutions. Thus, the principal goals of the enterprise are three: (1) to help researchers to design better questionnaires, (2) to allow consumers of questionnaire data to look from a new perspective as they analyze and evaluate the meaning of their findings, and (3) to energize future methodological research to identify and refine principles of optimal questionnaire design. Topics covered will include: Introduction and Theory, Open vs. Closed Questions, Rating vs. Ranking, Response Order Effects, No Opinion Response Options

Survey Interviewing Techniques
This course examines the thinking and empirical research behind current interviewing practice, alternative approaches to interviewing and the embodiment of interviewer-like behaviors in computer-based, self-administered questionnaires. The course will consider the rationale for and against the “standardized” approach in which interviewers strive to use identical wording for all respondents and which is the industry standard. We will then consider the arguments for and against “conversational” interviewing, an alternative approach in which interviewers use the words they deem necessary to ensure uniform question understanding across respondents. The primary measures used to evaluate the techniques are response accuracy and interview duration. We will discuss differences between the approaches in both telephone and face-to-face interviewing and will distinguish between ideal implementation of these interviewing techniques and what interviewers actually do. The course will then explore the concept of “rapport” between interviewer and respondent as a way to understand the impact of interviewers on answers to sensitive questions. We will discuss some of the research on formal versus interviewing style, an experimental attempt to vary rapport. The final part of the course will cover recent attempts to embody some of these concepts in computerized self-administered questionnaires. This includes web questionnaires that allow the respondent to obtain clarification from the questionnaire (sometimes by clicking, sometimes by speaking) and that can offer clarification when the respondent seems to need it. Finally, we will discuss very recent research on virtual interviewers embedded in the interface of web-based questionnaires. Among the issues we will consider are the importance of the virtual interviewer’s visual realism and its dialog capability. Do respondents provide more socially desirable answers when the interviewing agent looks more like a human? Do respondents communicate more socially, e.g. more nods, uh huhs, and thank you’s under these circumstances? Does an interviewing agent that can recognize respondents’ confusion help enough to warrant the development costs?

Survey Process Improvement Using Paradata Analysis
During the last twenty years survey data have been increasingly collected through computer assisted modes. As a result, a new class of data * called paradata * is now available to survey methodologists. Typical examples are key-stroke files, capturing the navigation through the questionnaire, and time stamps, providing information such as date and time of each call attempt or the length of a question-answer sequence. Other examples are interviewer observations about a sampled household or neighborhood, recordings of vocal properties of the interviewer and respondent, information about interviewers and interviewing strategies. While the type of available paradata varies by mode, all share one feature–they are a by-product of the data collection process capturing information about that process. This course covers the potential of paradata for social survey research. The course will give an introduction and overview of methodological issues involved in the collection and analysis of paradata. We will discuss several research examples including but not limited to the use of paradata to monitor fieldwork activity, guide intervention decisions during data collection (e.g. through responsive design), and to address various total survey error components (in particular measurement error and nonresponse bias). Cases-studies will draw attention to the challenges in automated data capturing and modeling of the complex structure of paradata. The objective is to provide the participants with an overview of best practices as well as cutting edge research on the use of paradata and to help participants gain a thorough understanding of the role of paradata in increasing survey quality and reducing total survey error.

Telephone Interviewing Techniques
Interviewers are the heart and soul of telephone survey research. A researcher’s theory, instrument, sampling, and programming may be perfect, but it will all be for naught without highly-trained interviewers to gather the data. This session will examine telephone survey interviewers’ work at multiple levels, including respondent-interviewer interaction at the micro level, the survey research organization context at the meso level, and the concept of total error in surveys at the macro level. (Note: Much of the material applies to in-person interviewing as well.)

Focusing on best practices for collecting high-quality telephone survey data and minimizing interviewer-related error, this one-day workshop will cover the nuts and bolts of telephone interviewers’ work, from telephone etiquette, to call disposition codes, to why interviewers call back people who have “refused” to participate. It will also cover the finer points, such as how interviewers control their voices, know when to probe respondents’ answers, manage an interview’s pace, distinguish between “hard” and “soft” refusals, and employ the cognitive psychology of survey response to handle certain types of difficult respondents: the Lucies, the bullies, the victims, and the dissenters. Students will learn how interviewers’ work varies distinctly over the life cycle of a typical telephone survey ? from the first few days, to the middle of data collection, to the final few interviews. In the process we will talk about how scientific telephone surveying survived the 1990s, when telemarketing firms hijacked telephone survey methods to engage in deceptive practices, and how that episode affects interviewers’ ethical practices today.

Telephone Surveys in the Era of Cell Phones
This course will address how, when and why to incorporate cell phones in telephone surveys. For decades, researchers have relied on landline random digit dial (RDD) telephone samples to survey the general household population. As a growing number of households rely solely or mostly on a cell phone, telephone surveys conducted only by landline are at risk of producing biased estimates. This course identifies methods of designing and implementing dual frame telephone surveys and describes some of the key issues in these surveys. The course will examine best practices, costs and challenges in cell phone surveys, and will discuss the statistical issues involved in combining landline and cell phone samples. It will also briefly address alternative ways to cover households that are cell phone-only.

The Analysis of Clustered Data
This short course is aimed at the audience of statisticians and analysts who have to analyze data with cluster-correlated outcomes. The course will describe complex concepts such as GEE and HLM in plain terms with simple examples (e.g., based on three observations), and provide intuitive understanding of the basics principles behind the methods for analyzing correlated continuous and categorical (mostly binary) data. The course will explain the difference between population averaged and cluster-specific (hierarchical) models and the basics of generalized estimating equation (GEE) methods. The definition of clusters is very broad and covers primary sampling units such as neighborhoods, schools, hospitals, as well as individuals when multiple measures are taken for the same individual or entity.

Total Survey Error
The Total Survey Error (TSE) paradigm embodies the best principles, strategies, and approaches for minimizing the survey error from all sources within time, costs, and other constraints that may be imposed on the survey. The TSE paradigm can be viewed as resting on four pillars of survey methodology corresponding to survey (a) design, (b) implementation, (c) evaluation, and (d) data analysis. Regarding (a), the TSE paradigm specifies that surveys should be designed to maximize data accuracy subject to budgetary and other constraints by minimizing the cumulative effects of error from all known sources. For (b), the paradigm specifies that strategies should be in place to monitor the major error sources, adapting the survey design as necessary to minimize the TSE through real-time interventions and design modifications. Regarding (c), the TSE paradigm emphasizes the importance of regularly assessing the joint effects of survey error on estimation and analysis so that continuous improvement and future design optimizations are possible. Finally, for (d), the TSE paradigm specifies that data analysis should appropriately consider the complex sampling design and the effects of nonsampling errors on the analytical results. Structured around these four pillars, the course presents the best methods and lessons learned by the instructor over four decades of survey practice in government, academia, and private industry.

Usability Testing for Survey Research
Usability testing in survey research allows in-depth evaluation of how respondents and interviewers interact with computerized questionnaires. The purpose of this course is to apply the principles of usability testing that is currently being used for website and product testing, and adapt these methods to show how they can apply to survey research. Usability testing is similar to cognitive testing except rather than testing how well respondents cognitively understand survey questions, it shows how well respondents use and navigate web surveys and web survey features. The purpose of this course is to provide researchers with a foundation in usability testing as a method for pretesting web surveys.

Visual Design: A Hands-On Approach
This course focuses on how and why words, numbers, symbols and graphics independently and jointly influence answers to questions in Internet and paper surveys. It begins with theoretical background on why and how the visual aspects of questions are interpreted by respondents and guide their reading and comprehension of meaning. Applications of the theory and research to designing individual person and establishment surveys in ways that improve their usability for respondents will be provided. The course includes a discussion of the substantial implications these ideas have for the design of mixed-mode surveys in which some respondents are asked to report aurally (e.g. telephone) and others are asked to complete visually communicated (web or mail) survey questions. The substantial visual design challenges researchers are now facing with designing questions for smartphones will be discussed as part of the mixed-mode design issues that must be addressed in many surveys.

Visual Design for Self-Administered Questionnaires
This course focuses on how and why words, numbers, symbols and graphics independently and jointly influence answers to questions in Internet and paper surveys. It begins with theoretical background on why and how the visual aspects of questions are interpreted by respondents and guide their reading and comprehension of meaning. Applications of the theory and research to designing individual person and establishment surveys in ways that improve their usability for respondents will be provided. The course ends with a discussion of the substantial implications these ideas have for the design of mixed-mode surveys in which some respondents are asked to report aurally (e.g. telephone) and others are asked to complete visually communicated (web or mail) survey questions.

Web Survey Implementation
Are you interested in learning how the Web can be used effectively in survey research? Have you wanted to see how Web surveys are actually implemented? In this course, we will trace the process that begins with a questionnaire and follow it through to survey implementation and management. Topics will include: Design implications on survey quality; Interactive, navigation, and screen design features in Web surveys; Minimizing error through standards and testing; Respondent contact strategies and sample management; Security and confidentiality issues; New innovations in Web survey design and implementation; Web survey software/service options and selection criteria. This course will be taught with the survey practitioner/manager in mind and will be very useful for graduate students or researchers desiring to learn more about Web surveys. Knowledge of complex survey methods or computer programming is not required. Participants in this course should have a basic understanding for the practice of survey research, and some experience in using the Web. The course will not be tied to a specific web survey product or service, however, the presenters will share their experiences with several products available.

Weighting and Imputation Methods for Analyzing Incomplete Data
Every analyst encounters the problem of dealing with missing data. Numerous methods have been proposed for analyzing incomplete data. The most common and useful approaches are weighting the complete-cases to compensate for incomplete cases; and imputation of the missing values in the data set to be analyzed. This one-day short course will cover the broad principles of weighting and imputation methodologies and, especially focus on the multiple imputation framework for analyzing the incomplete data. Several examples will be provided and discuss some software options. This course will assume that participants have taken courses covering basic statistics and regression analysis using linear and logistic regression models. This course will emphasize applications but with the heuristic explanations of the underlying theoretical principles.

Weighting Survey Data
This course is an introduction to the basic concepts for weighting survey data. It begins by defining the goals of weighting including weighting as a correction for differential selection probabilities, non-response and non-coverage. The course covers the process of developing weights for stratified two-stage sampling including computing design weights and methods for nonresponse adjustments and frame coverage error adjustments. Additional topics may include the effect of weighting on variance of the estimates and extreme weights. When and How to Use Web SurveysThere are many different ways that samples can be obtained for online surveys and experiments. These include open invitation surveys of volunteers, intercept surveys, opt-in or access panels, Amazon’s Mechanical Turk, Google Consumer Surveys, list-based samples, and the like. In almost all cases, the goal is to make inference to some larger population or process. These different approaches have strengths and weaknesses. Understanding these is important to choosing the right method for the research question, given available resources. The course will review some of the inferential challenges, including sampling error, coverage error, and non-response error. A variety of adjustment methods, such as weighting, propensity score adjustment and matching, are being used to mitigate the risk of inferential errors, and these will be reviewed too. The course focuses on the practical choices a researcher needs to make in choosing a research method, in analyzing the data, and in describing the results of the research. This course is suitable for people who are considering conducting a Web survey to collect data for research or analyzing data from an existing Web survey.

When One Mode Is Not Enough: Methodology for Mixed Mode Surveys
When planning a survey, many decisions have to be made, and one of the most important decisions concerns the choice of data collection mode. In the first decade of the 21st century a large variety of data collection methods are available for social surveys and official statistics, which leads to methodological questions, such as, which method is best? Each method has advantages and disadvantages. Sometimes, the choice is easy and straightforward, but often the situation is more complex and one single method will not suffice. Therefore, multiple modes of data collection or mixed modes have become more and more popular in survey practice. The topic of this short course is the methodology for mixed mode surveys. I will discuss the advantages and disadvantages of mixed mode survey designs, give an overview of common forms of mixed mode design and reasons for using more than one mode in a survey. Special attention will be given to the impact of mixed-mode designs on survey quality, such as coverage, measurement error and mode effects, nonresponse, cost and timeliness. The course will provide students with a theoretical background on mixed mode methodology and empirical knowledge on the implications of mixed-mode for survey quality, logistics and resources.

Introduction to R for Social Scientists
This is a two-day course on R, an open-source programming language for statistical analysis and graphics. It provides the analyst with a wide variety of tools commonly used in statistical modeling with more flexible, objected-oriented facilities than other programs like Stata or SAS. This course is designed for those with little or no R experience. It will cover basic syntax and data loading, model estimation, loading and using written packages (including a sampling of popular packages), graphical presentation of model results, and Monte Carlo simulation. After completing the course you will know enough to be able to (1) conduct a typical statistical analysis for your own research and (2) search for the things you don’t know in an efficient manner.

Introduction to R for Statistical Analysis
The statistical programming language and computing environment S has become the de facto standard among statisticians and is making substantial inroads in the social sciences. The S language has two implementations: the commercial product S-PLUS, and the free, open-source R. Both are available for Windows and Unix/Linux systems; R, in addition, runs on Macintoshes. The purpose of this workshop is to provide a quick introduction to R and to show you how to accomplish a variety of tasks, including writing basic programs and constructing non-standard graphs. The statistical content is largely assumed known. Topics: R basics; data in R; statistical models in R; structural-equation modeling in R with the SEM package; R programming; and R graphics.

LISREL
This course is designed to provide a basic introduction to LISREL and PRELIS (a component of LISREL). The focus is on programming using the LISREL matrix syntax and on specifying structural equation models.
Mathematica: Computer Algebra/Calculus (2 days)
Mathematica is a software for symbolically manipulating mathematical expressions. The first day of this course will cover basic algebra, calculus and matrix algebra in Mathematica. The second day will cover common scientific extension of these operations including optimization, series approximation and function programming.

Matlab: Data Analysis (2 days)
Matlab is the frontrunner in software for scientific computation. This course will cover the basics of data analysis in Matlab. Material covered will include; introduction to the Matlab programming language, data management and basic description, ANOVA, Ordinary Least Squares regression modeling and Maximum Likelihood Estimation of Nonlinear Regression Models.

Matlab: Graphics
Matlab contains powerful tools for graphically exploring and describing data. This course will cover the basics of one, two and higher dimensional plotting in Matlab.

Matlab: Simulation Modeling (2 days)
Simulation Modeling (at times referred to as Computational Modeling) is a powerful tool for theoretically exploring social systems. Matlab has utilities that permit flexible model specification and extensive exploration of simulation results. This course will cover the basics of discrete event simulation in Matlab.

Missing Data: Lab Session
This course will first discuss the usual methods for dealing with missing values, such as listwise deletion and simple mean imputation and the consequences of each. We will move to regression model imputation without and with methods that compute variance estimates utilizing variance induced but the imputations. A lab session will be included to demonstrate software available for imputation and estimation using sythesized data.

MPlus
Mplus is a modeling program that integrates random effect, factor, SEM and latent class analysis in both cross-sectional and longitudinal settings and for both single-level and multi-level data. As such, this short course will only scratch the surface of Mplus’ capabilities. The basic structure of the program and how it can be modified will be taught in a hands-on way in the Odum Institute Computer Lab.

SAS
This is a four-part course. SAS part 1 of 4 will give an introduction to the SAS system and SAS windows. Topics to be covered include: creating and saving SAS programs; reading in data from simple and complex text data sets; typing variables; obtaining frequencies, contents, and univariate statistics. SAS part 2 of 4 will discuss formatting variable values; creating SAS libraries for storing and retrieving SAS data sets and format files; reading raw data from external files; creating new SAS data sets from existing SAS data sets, subsetting by observation and by variable. SAS part 3 of 4 will explain how to create new SAS data sets combining information from multiple existing SAS datasets; how to sort, concatenate, interleave, and merge data sets; how to perform the t-test, and test for no association in a contingency table. For SAS part 4 of 4, attendants will be allowed to suggest topics. Past topics include variable retyping, creating SAS datasets from SAS output; creating html and Microsoft Word tables, ANOVA, importing and exporting Excel files.
SEM in R Using Lavaan
Lavaan is an SEM package free to faculty and students. This talk will be 45-minutes long followed by a Q&A session. No registration is required.

SPSS
Part 1 of the course will offer an introduction to SPSS and teach how to work with data saved in SPSS format. Part 2 will demonstrate how to work with SPSS syntax, how to create your own SPSS data files, and how to convert data in other formats to SPSS. Part 3 will teach how to append and merge SPSS files, demonstrate basic analytical procedures, and show how to work with SPSS graphics. Please bring a flashdrive to class.
Stata
This is a 3-part short course (held over three afternoons). Stata part 1 will offer an introduction to Stata basics. Part 2 will teach entering data in Stata, working with Stata do files, and will show how to append, sort, and merge data sets. Part 3 will cover how to perform basic statistical procedures and regression models in Stata.
Stata II: Programming
This three-day course will serve as an introduction to the uses and how-to’s of STATA programming. The primary goal is to familiarize participants with the advantages and essential methods of STATA programming, as well as providing a foundation for more complex techniques and functions. The first two parts will be interactive-instruction covering the basics of STATA programming, including foreach loops, the use of macros, and the ins and outs of writing your own STATA programs. The final day will be in workshop format, where participants are encouraged to bring any specific programming questions or challenges for one-on-one assistance from the instructor. NOTE: Participants should have at least a baseline understanding of STATA before taking this course. Having taken or knowledge of the topics covered in STATA I is preferable.

101: From Idea to Article . . . and Beyond
Although graduate school teaches you many things, many Ph.D. students complete their dissertations having never been fully trained in academic publishing. How does academic publishing actually work? Where should you publish your research? How do you actually get papers accepted in strong academic journals? In this talk, we will discuss several major ideas in academic publishing (focusing on the social sciences). We will discuss: professional integrity and ethics; the role of academic conversations and communities as a guide for publication forums and journal selections; the mechanics of publishing in journals and other forums (including outlining, writing style, journal, legal, and newspaper submissions, the peer review process, revisions, and corresponding with editors); and acceptance and all that follows.

Advanced Access to Census Data
Hands-on workshop to help users understand the strengths of various Census (and other survey) data retrieval tools which allow the creation of custom cross-tabulations (that is, custom data tables). Tools to be covered include: DataFerrett; iPUMS/TerraPopulus (in beta); and the Triangle Census Research Data Center (TCRDC). The first two tools are freely available and focus on census data (U.S. for DataFerrett; international for iPUMS/TerraPopulus); researchers must apply to the Census Bureau (or other federal agency, e.g., the Centers for Disease Control) for access to the TCRDC in order to utilize survey microdata. TCRDC staff will present this portion of the class. Hands-on – 3 hours

Advanced Access to Census Data
Hands-on workshop to help users understand the strengths of various Census (and other survey) data retrieval tools which allow the creation of custom cross-tabulations (that is, custom data tables). Tools to be covered include: DataFerrett; iPUMS; TerraPopulus (in beta); and the Triangle Research Data Center (TRDC). The first three tools are freely available and we will focus on their census data content (U.S. for DataFerrett; U.S. and international for iPUMS and TerraPopulus); researchers must apply to the Census Bureau (or other federal agency, e.g., the Centers for Disease Control) for access to the TRDC in order to utilize survey microdata.
Advanced Options for Census Data Retrieval
The afternoon session assumes at least basic knowledge of Census data (i.e., taking the morning workshop should be adequate preparation) and will focus a spectrum of tools to access various U.S. and international data. Some specific tools covered will include DataFerret, iPUMS/Terra Populus, and the Triangle Census Research Data Center. Some hands-on exercises are included

Agent-based Modeling
This course offers an introduction to a new analytical method of agent-based modeling. It provides an easy way for the beginner to translate research goals into a dynamic model in simulation form. This course offers a step-by step, hands-on, interactive approach to conceptualizing, creating, implementing, and analyzing policy/political science simulation models. This analytical tool can be used in addition to traditional triangulation strategies to operationalize quantitative and qualitative variables (or a combination of both) into a simulation.

An Introduction to Social Network Analysis
This short course will serve as an introduction to the basic concepts and procedures of social network analysis (SNA), with an emphasis on descriptive measures. SNA has a long history in social psychology, sociology and anthropology, and has more recently been applied widely in public health & epidemiology, communication studies, political science, and other fields where relationships among individual people or other entities are the objects of study. The course will focus on the measurement of key properties at various levels: at the actor level in complete social networks (e.g. centrality) and ego-networks, the dyad and triad level, subgroups (e.g., cohesion & components) and at the global level (e.g. density & modularity). The course will also cover position and role analysis (e.g. structural equivalence and blockmodels) and a brief look at statistical issues. Demonstrations in UCINet and/or R will be provided, but the course is not intended to be a hands-on technical training in any software package.

An Introduction to Web Mapping and Data Visualization
This hands-on course will introduce three distinct online tools for mapping and visualizing data: ArcGIS Online, Tableau Public, and Google Fusion Tables. A companion to ArcGIS Desktop, ArcGIS Online is a tool that allows users to import data, create maps, and share content. Tableau Public is free software that enables users to easily create and share a variety of interactive visualizations and maps. Google Fusion Tables is a suite of tools enabling users to map, visualize, manage, discover, and share tabular data. Through a variety of exercises, attendees will gain familiarity with the interface and functionality of each platform, learning the strengths and weaknesses of each for various applications. Attendees are encouraged to create both a Google account and a Tableau Public account beforehand.

Basic Access to Census Data
Hands-on workshop to help users understand the strengths of various Census data retrieval tools, both freely available ones and those to which the library subscribes: American FactFinder, the Census Bureau’s freely available database; Social Explorer, a commercially licensed tool to which the library subscribes; and the grant-supported (so, free to you) National Historical Geographic Information System (NHGIS). These tools provide access to pre-constructed data tables published by the Census Bureau. Some are better for the most recent data and others are useful for historical data. Come learn how to choose the best tool for your research, and the ins and outs of each tool.

Causal Inference
This short course will provide a brief orientation to the counterfactual-based inference in observational studies where treatment assignment is non-random. The course will seek to answer causal questions such as “what is the impact of a single intervention A, such as a new climate change policy, on a single outcome Y, such as carbon emissions?”
Counterfactual analysis of causation does not require full specification of all causes and only require data to be balanced with respect to treatment (intervention) assignment. Randomization of the treatment assignment is expected to exclude all alternative causes and balance potential confounders to establish secure causal claims with certainty. However, in field observation studies (mostly in social or medical sciences), randomization is very difficult to achieve due to the non-experimental nature of the treatment where treatments are observed rather than assigned. In such studies, matching based methods are now widely used to invoke randomization and to make causal claims.
Morning session will focus on conceptual understanding of the counterfactual-based causal inference and afternoon computer practical will include step-by-step implementation of one of the matching methods – propensity score matching – in R. After completing the short course, you will know enough to (1) explain counterfactual conception of causal inference and (2) conduct a propensity score matching for your own research. Some basic knowledge of regression (linear and logistic) and R would be highly beneficial.

Census Data Introduction
This morning workshop is an introduction to Census methodology, contrasting the decennial census with the American Community Survey. Topics covered will include survey methodologies, Census geographies, data suppression and new topics added to the ACS. This session assumes no prior knowledge of Census data and includes hands-on exercises with the data.

Choosing a Programming Language
Interested in learning to code but unsure where to start? This workshop will explore the strengths and weaknesses of three languages frequently used with data and research – JavaScript, Python, and R. For each language, we will suggest packages and/or integrated development environments (IDEs) and help you set up your computer to start writing and running code. Finally, we’ll suggest some resources, on and off campus, to help you learn some of the basics.
All are welcome, but laptops are strongly recommended. This workshop focuses on coding for automation and data manipulation in a replicable way, not on developing formal software for others to use.

Collecting and Analyzing Textual Data
This two-day hands-on short course provides a brief introduction to quantitative text analysis and mining in the social sciences for those who have little to no experience with the topic. The first session will focus on the basics of the collecting and formatting the text, an overview of how to extract specific pieces of information, and how to process your documents. The second session will provide an introduction to supervised, semi-supervised, and unsupervised models used to classify or elicit information from the text. In this context, supervised means that the researcher provides some amount of information for an algorithm to be trained and then be used to make predictions or explanations. A basic working knowledge of R is necessary.
Create Your Own Dataverse Network (DVN)
This introductory course, intended for researchers and research teams creating/managing social science datasets, will focus on using Odum’s Dataverse Network as a tool to share data for collaboration and preservation purposes. It will cover creating a Dataverse, uploading data sets, and creating collections of data, as well as permissions and sharing features.

Data Curation: Managing Data throughout the Research Lifecycle
Part of the Data Matters: Data Science Short Course Series This course will provide an introduction to data management best practices as well as demonstrations of digital curation tools including the Dataverse Network™ open source virtual archive platform.

 

Do-It-Yourself Data Management
Researchers are typically faced with the task of maintaining numerous files related to analyzes from various sources such as data, programs and output. Absent some management plan, file storage systems often result in confusion, replication and wasted time. This course offers strategies for efficient file management to reduce those problems.

Excel for Social Scientists
This workshop provides an overview of ways Excel can be used to structure and even analyze data and to create presentations. It starts with examining basic tools and functions in Excel (including some shortcuts to simplify using common features of the program, ways that Excel can be used to manage textual data, to combine or divide data cells and sources, and to perform calculations such as sum, average, minimum, and maximum). It then proceeds to an overview of how to sort and transpose data and create common graphs and tables. Finally, it reviews how to perform basic statistical operations, including counts, frequencies, and even simple linear regressions, in Excel.

Genetics and Social Science Research Faculty Workshop
Advances in genetics have heightened interest in what role genetic factors might play in the study of social science phenomena. Yet, training in genetics is not part of most social scientists’ background. This three-hour workshop provides social scientists an overview of using genetics in research. The workshop includes brief summaries of concepts in genetics and the growing relevance of genetics to social science research, as well as two case studies of how genetics is being used in current social science research at UNC Chapel Hill.

Harvesting Internet Data with Stata
This short course provides an introduction to using Stata to gather data from web pages and RSS feeds. We will discuss ways to automate the downloading of multiple pages. We will also review a set of commands useful for transforming the text on pages or feeds into variables that can be analyzed in Stata. Commands to be covered include: copy, file, filefilter, regexm, post. We will also review common problems when using Stata to parse text. Students with a familiarity with Stata will get the most out of the class, but no advanced skills are expected. This is a hands-on course. If you have questions, please contact Paul_Mihas@unc.edu.

Introduction to Census Concepts
Do you know that variables like income and educational attainment are no longer part of the decennial census? Do you understand the differences between the decennial long form methodology and that of the American Community Survey (ACS)? If your answer to these questions is no, this class is for you! We will compare and contrast content and methodology of the decennial census long form and the ACS, and review Census terminology and geographies. (This class or equivalent knowledge is required for the Basic and Advanced Access classes.)

Introduction to Effective Information Visualization
Participants will learn how to clean and structure data; see how freely and commonly available software can be used to create effective visualizations; and learn basic design principles, so you can go beyond the defaults and create eye-catching and impactful figures and infographics!

Introduction to LaTeX
This is a two-day course on LaTeX, an open-source markup language/document preparation system widely used in academia to produce high-quality typesetting. In addition to producing beautiful-looking documents, slideshows, and posters, LaTeX can make many features of the manuscript-writing process–the bibliography, insertion of figures and tables, and all those requirements that the Graduate School or journals require–quick and easy. This course is designed for those with little or no LaTeX experience. It will cover basic syntax, loading and using written packages (including a sampling of popular packages), graphics, style files, creating a bibliography, making slide shows and posters, and integrating LaTeX and output from statistical software like R or Stata.
After completing the course you will know enough to be able to (1) pronounce “LaTeX” correctly, (2) create a basic document, slideshow, or poster, and (3) search for the things you don’t know in an efficient manner.

Introduction to Web Mapping and Data Visualization
This hands-on course will introduce three distinct online tools for mapping and visualizing data: ArcGIS Online, Tableau Public, and Google Fusion Tables. A companion to ArcGIS Desktop, ArcGIS Online is a tool that allows users to import data, create maps, and share content. Tableau Public is free software that enables users to easily create and share a variety of interactive visualizations and maps. Google Fusion Tables is a suite of tools enabling users to map, visualize, manage, discover, and share tabular data. Through a variety of exercises, attendees will gain familiarity with the interface and functionality of each platform, learning the strengths and weaknesses of each for various applications. Attendees are encouraged to create both a Google account and a Tableau Public account beforehand.

Network Analysis: Statistical Inference with Exponential Random Graph Models
Exponential random graph models (ERGMs) are flexible statistical models for relational (i.e., network) data that are capable of representing and identifying an extensive range of interdependencies common in networks. Are gender, race or social class predictive of mixing patterns in a social network? Are ties in a directed network typically reciprocated? Does the network exhibit transitive triad closure? The simultaneous and stochastic manifestation of relational dependencies such as these can be identified and characterized with ERGMs. This workshop will introduce ERGM and demonstrate its application in the free and open source R statistical software. Participants will be provided with real-world network data as well as R code to apply ERGMs to that data.

Practical “Big Data”: Separating the Hope from the Hype (Two-day Course)
The phrase “Big Data” has come to designate a network of relatively new computationally intensive methods that merge machine learning and statistical methods for the analysis of very large data sets derived from secondary sources, usually the Web. These lectures will provide an overview of the most commonly used approaches, and how these do — and sometimes do not — differ from conventional social science statistical approaches. The lectures emphasize approaches and resources for gaining further knowledge and technical proficiency, rather than going into depth on any single method; with very few exceptions, all of the software illustrated will be open source.
Module 1: Big Data: sources and practical implementation. Web-scraping. Hadoop and other distributed databases, “cloud” computing, and the “map-reduce” approach. Resources in R and Python. Ethical considerations: privacy, intellectual property
Module 2: Working with unstructured text: regular expressions, natural language processing suites for pre-processing text; named entity and feature extraction
Module 3: Working with unstructured text: supervised text classification and unsupervised topic models.
Module 4: Working with large-scale semi-structured data: clustering, decision-trees, ensemble methods, and visualization

System Dynamics
This two-day course will introduce systems thinking and system dynamics computer simulation modeling, a computer-aided approach to policy analysis and design. Our goal will be to enhance skills in understanding and analyzing the complex feedback dynamics in social, economic, and environmental problems. The course will introduce system dynamics modeling through the STELLA and Vensim modeling platforms.

Back to top

Education Staff


If you have questions about our courses, or registration for any of our workshops, please contact Jill Stevens.


A headshot of Paul Mihas.

Paul Mihas

Assistant Director for Education and Qualitative Methods

919-962-0513
paul_mihas@unc.edu


A headshot of Jill Stevens.

Jill Stevens

Educational Coordinator

919-962-4157
jill_stevens@unc.edu

 

Back to top

Suggest a Short Course


We are always looking for better ways to serve the campus community. Is there a short course you would like to see offered? Use the form below to submit a short course suggestion.

Before submitting your suggestion, check out our list of current and past courses above. Feel free to submit a past offering or indicate that you would like to see more offerings of a current course in your submission.

Request a Short Course

  • (please be as specific as possible)

 

Back to top