Social Science Research Commons

Permanent link for this communityhttps://hdl.handle.net/2022/18403

Browse

Now showing 1 - 20 of 154

A Brief Introduction to Multilevel Modeling: Concepts & Applications
(Indiana University Workshop in Methods, 2015-02-27) Rutkowski, Leslie
In this two-hour workshop, participants will be provided with a brief overview of multilevel modeling concepts and several applications, including random intercepts and random slopes models. Several examples will be provided along with SAS syntax and a data set will be made available. Participants will have an opportunity to fit several models in SAS and interpret the results.
A Comparison of Online Panels with GSS and ANES Data
(2015-06-24) Zack, Elizabeth; Kennedy, John
In the past five years, researchers have increasingly used low cost data collection methods to conduct surveys. Survey data collection software platforms such as Qualtrics and Survey Monkey allow researchers to easily and cheaply create questionnaires for distribution. Similarly, low cost methods are available to recruit survey participants. Some of these include online panels, Amazon’s Mechanical Turk (MTurk), and Google Consumer Surveys. With these tools, researchers are much less dependent on professional survey researchers to conduct surveys. More importantly however, survey researchers are now using non-probability online panels as a substitute for probability samples. In 2010 and 2013, AAPOR released task force reports that analyzed the challenges encountered when using online panels and nonprobability samples for high quality survey research. In 2014, a book on the use of online panels in survey research included chapters written by respected survey researchers (Callegaro et al, 2014). In the past five years, at least 20 peer-reviewed methods articles were published on the use of Mechanical Turk for social and behavioral science research. Despite the cautions raised about the appropriate use of online panels, they are being used more often, e.g., the YouGov panel and CBS News. In this presentation, we will discuss the results of a number of experiments we conducted that compared distributions in questions asked recently in the ANES and GSS to similar questions asked with MTurk samples and a Qualtrics online panel. In addition, we will show how simple multivariate models are similar and different using data from both the probability and non-probability samples. This presentation will contribute to the continuing research into the appropriate uses of online panels for survey research.
A Realist's Guide to Gerrymandering
(Indiana University Workshop in Methods, 2019-11-08) Duchin, Moon
Moon Duchin is an Associate Professor of Mathematics at Tufts University.
Abortion Attitudes: The difficulties of simple questions about complex issues
(Indiana University Workshop in Methods, 2022-08-26) Jozkowski, Kristen; Crawford, Brandon
The recent Supreme Court decision in the Dobbs v. Jackson resulted in the overturning of Roe v. Wade, ending the constitutional right of a pregnant person to choose to have an abortion until fetal viability (i.e., until about 24 weeks). Although this ruling marks a significant shift in abortion policy and climate in the U.S., the vast majority of polls measuring abortion attitudes show that there has been little change in attitudes since the 1970s. What does that mean about the relationship between policy and public opinion and how we measure abortion attitudes? In this workshop, we will provide an overview of abortion as a contentious social issue, historical trends of abortion attitudes, and the implications of the Dobbs v. Jackson decision. Next, we will share an overview of our project: Developing and Assessing Measures for Social Surveys (DAMSS). Then we will present some of our preliminary findings and discuss what makes a “good” question actually a “bad” question? We will conclude with what these preliminary findings mean for how we assess abortion attitudes and what the DAMSS project will be doing next. No previous experience with abortion attitudes is needed. Although the focus of this presentation will be abortion attitudes, the findings and methodology have relevance for other social issues.
Accessing Complex Systems: An Introduction to Agent-Based Models
(Indiana University Workshop in Methods, 2020-03-06) Uribe-McGuire, Alicia
The world we live in is complex. Individuals change, evolve, and grow as they experience the world and others around them. Agent-based and computational models can help us account for the complexity and dynamics of interactions among individuals. This seminar will include an introduction to agent-based models, explore some prominent examples, and provide a brief primer on how to approach these models.
Advanced Topics in R
(Indiana University Workshop in Methods, 2014-12-05) Davis, Jefferson; Zhang, Hui; Michael, Scott
In this follow up workshop to “Introduction to R” researchers will have the opportunity for a deeper dive into R. Available on all IU's supercomputers, R is a flexible open source statistical programming language that can work with large and complex data sets. This workshop will address several advanced topics in R as well as giving participants the opportunity to use R on IU’s supercomputers. The topics to be covered include: R scripting on IU supercomputers; debugging, profiling, and performance analysis of R code; parallel programming in R, including the Rmpi and snowfall packages; and advanced plotting in R. Participants will have access to the supercomputer Big Red II during the session and will be able to see hands-on examples of running R code and submitting batch jobs in R.
An Introduction to IPUMS Health Surveys: Harmonized versions of the National Health Interview and Medical Expenditure Panel Surveys
(Indiana University Workshop in Methods, 2021-02-05) Drew, Julia A. Rivera
This talk builds on last week’s introduction to IPUMS, given by my colleague Kari Williams. Targeting social scientists interested in health research, I will provide an in-depth exploration of the harmonized versions of the National Health Interview Survey (NHIS) and Medical Expenditure Panel Survey (MEPS) data available through IPUMS Health Surveys. The NHIS data, collected by the National Center for Health Statistics, is the longest-running, annual survey of health in the world. It is the primary source of information on topics such as physical and mental health status, chronic condition prevalence, health care utilization, and health care access, making it a critical resource for the study of population health disparities. MEPS, collected by the Agency for Health Care Research and Quality, is the primary source of information on health care expenditures in the United States. The MEPS Household Component is a longitudinal, household-based survey, collecting information about the same people during five interviews over the course of a two-year period. Topics covered include demographic characteristics and transitions, employment, medical conditions, health care utilization and access, and health care costs. The public use files combine survey data with billing data collected directly from a subsample of health care providers and pharmacies serving MEPS respondents.
Analysis with MATLAB
(2018-02-27) Davis, Jefferson
This workshop will cover basic usage of MATLAB, different use cases and examples, and where to get and run MATLAB.
Analyzing longitudinal changes in mental health: Integrating variable-centered and person-centered approaches
(Indiana University Workshop in Methods, 2022-03-25) Chow, Angela
While variable-centered approaches to data continue to be widely employed in public health, psychology and many other disciplines, the use of person-centered approaches have become increasingly common. A third type of approaches, which integrates both variable- and person-centered approaches, has been rapidly developed and adopted in research which involves longitudinal data. In this workshop, we will first compare these three approaches. Then, drawing on a longitudinal study which examines the trajectory patterns in maternal depressive symptoms across pre- and postnatal periods, we will go through the steps of conducting general mixture modeling, an analytical procedure combining variable- and person-centered approaches. We will also discuss how this integrated approach can help to address research questions which are difficult to answer otherwise.
Analyzing ordinal data with metric models: What could possibly go wrong?
(Indiana University Workshop in Methods, 2018-03-23) Kruschke, John K.
We surveyed all articles in the Journal of Personality and Social Psychology, Psychological Science, and the Journal of Experimental Psychology: General that mentioned the term "Likert," and found that 100% of the articles that analyzed ordinal data did so using a metric model. We demonstrate that analyzing ordinal data as if they were metric can systematically lead to errors. We demonstrate false alarms (i.e., detecting an effect where none exists, Type~I errors) and failures to detect effects (i.e., loss of power, Type~II errors). We demonstrate systematic inversions of effects, for which treating ordinal data as metric indicates the opposite ordering of means than the true ordering of means. We show the same problems --- false alarms, misses, and inversions --- for interactions in factorial designs and for trend analyses in regression. We demonstrate that averaging across multiple ordinal measurements does not solve or even ameliorate these problems. We provide simple graphical explanations of why these mistakes occur. Moreover, we point out that there is no sure-fire way to detect these problems by treating the ordinal values as metric, and instead we advocate use of ordered-probit models (or similar) because they will better describe the data. Finally, although frequentist approaches to some ordered-probit models are available, we use Bayesian methods because of their flexibility in specifying models and their richness and accuracy in providing parameter estimates.
Assuming in Public: Use DAGs to improve transparency and causal estimation
(Indiana University Workshop in Methods, 2024-02-27) Brauer, Jonathan
Scientists routinely make causal inferences – whether implicit or explicit – about correlations generated from statistical analyses of experimental and observational data. However, while theorized causes are usually directionally specific, correlations are inherently symmetric or directionally ambiguous. Moreover, multiple causal structures can produce equivalent correlational results, posing significant threats to the validity of statistical inferences. Fortunately, advances from the “causal revolution” in science and statistics have provided us with powerful tools, such as potential outcomes and directed acyclic graphs (DAGs), to better understand causes and effects. This talk will focus on how DAGs can help us “assume in public” more effectively. By introducing DAGs early in the research workflow and adhering to simple rules for their use, we can formalize the causal assumptions underlying our theories and statistical models, thereby enhancing transparency and reducing avoidable biases in causal estimation. The presentation will cover the four foundational structures in causal systems, as represented in DAGs: complete independence, pipes, forks, and colliders. Real-world and simulated examples – drawn from the speaker’s blog posts – will illustrate key concepts, such as d-separation, “good and bad controls,” and adjustment sets. Finally, the talk will introduce tools and resources to help researchers more confidently and effectively navigate the assumptions and challenges of causal inference.
Beginning Text Analysis with R
(Indiana University Workshop in Methods, 2016-04-29) Gniady, Tassie
R is an open source language for statistical programming and graphics. With libraries oriented towards text mining, and one even called twitteR, using R to analyze social and humanities data has gotten easier than ever. This workshop will introduce some basics of R and guide you through a scaffolded approach to learning R that includes written tutorials, online web apps, dynamic notebooks, and downloadable code. In this session we will generate basic word clouds and cluster dendrograms.
Best Practices in SPSS
(Indiana University Workshop in Methods, 2025-03-07) Klein, Neela; Ray, Elizabeth
SPSS is a common data analysis program for work in Social Sciences. It offers a point of access for data cleaning, description, and analyses in a user-friendly manner. Different from programs like R that require coding, SPSS provides a “point and click” interface that allows you to use the program intuitively. Behind the scenes of this “point and click” interface, though, SPSS can provide, generate and execute code FOR YOU, making it an accessible option for researchers aiming to improve transparency and replicability of their analyses. SPSS is a powerful and approachable tool for anyone looking to view, describe, clean, edit, or analyze data with simple to complex statistical analyses. The goal of this workshop is to provide an accessible, applied, and practical understanding of how to use SPSS. The workshop will begin with a description of the software including a detailed map of how to interact with the software, how to view previously collected data, how to subsect data and create composite variables, and how to create both descriptive visuals of data. We will cover how to execute and interpret various statistical analyses (e.g. ANOVAs, correlations, and regressions). The workshop will include both the point and click method of interacting with SPSS as well as cover how to generate and work with syntax (i.e. SPSS code). Though not required, to make the most of your attendance, arrive with the SPSS software (provided for free for IU faculty, students, and staff) already downloaded.
Black Professionals at Work: Methodological Approaches for Studying an Underrepresented Population
(Indiana University Workshop in Methods, 2018-02-16) Wingfield, Adia Harvey
Adia Harvey Wingfield is Professor of Sociology at Washington University in St. Louis. She specializes in research that examines the ways intersections of race, gender, and class affect social processes at work, and is an expert on the workplace experiences of minority workers in predominantly white professional settings, and specifically on black male professionals in occupations where they are in the minority. Prior to her talk at the Workshop in Methods, Dr. Wingfield will speak at the Karl F. Schuessler Institute for Social Research (1022 E. Third St.) on "Professional Work in a ‘Postracial' Era: Black Health Care Workers in the New Economy," 12-1:30pm.
Causal Inference for Complex Observational Data
(Indiana University Workshop in Methods, 2019-09-20) Huber, Chuck
Observational data often have issues which present challenges for the data analyst. The treatment status or exposure of interest is often not assigned randomly. Data are sometimes missing not at random (MNAR) which can lead to sample selection bias. And many statistical models for these data must account for unobserved confounding. This talk will demonstrate how to use standard maximum likelihood estimation to fit extended regression models (ERMs) that deal with all of these common issues alone or simultaneously.
Challenges and Opportunities when Studying Hard-to-Reach Populations
(Indiana University Workshop in Methods, 2022-02-25) Chong, Chinbo
Are you interested in surveying hard-to-reach populations? In this talk, I will briefly discuss my book project and ongoing projects, focusing on my motivations for providing practical tips for collecting survey data among Asian Americans and Latinos. I will emphasize the challenges of studying hard-to-reach populations like Asian Americans and Latinos. In that spirit, I will introduce considerations and practical steps for pursuing research from sampling strategy, matching process, survey questionnaire ordering, the timeline for completing your project, to various negotiation stages to consider when working with a survey firm. Finally, I will offer limitations of this approach and how you might address them in the publication of your research.
Characters and Grammar: How Linguists Can Become More Fluent in R
(Indiana University Workshop in Methods, 2022-09-23) Myers, James
R is not just statistics software, but a full-fledged computer language, and with its thousands of extra packages linguists can program it to do things far beyond basic statistical analysis and graph-making (though of course R is great at those things too). In this talk I hope to give a painless introduction to the grammar of R for those who have not yet dared to try it, while still offering some new ideas to more experienced users. The empirical focus is on my own explorations of the “grammar” of Chinese characters, which will allow me to survey a variety of methods of particular use to Chinese linguists, including how to work with non-Roman writing systems, how to analyze text corpora, how to compile data from lab experiments, and of course how to run statistical analyses and make graphs, from the simple to the fancy, including new types of analyses that you can invent yourself with some basic concepts in probability and a bit of programming. If time permits, I will also demonstrate how diverse and powerful R’s extra packages really are by highlighting tools for sound and image processing (the latter useful for the study of writing systems and sign language). Above all, I will emphasize that there is no reason to feel intimidated: anybody can become ever more fluent in R through workshops like these, textbooks, internet searches, and most importantly, patient trial and error.
Correspondence Audits: Design Issues and Practical Examples
(Indiana University Workshop in Methods, 2018-09-07) Gaddis, S. Michael
During the past decade, field experiments in the social and behavioral sciences have gained in popularity as the internet has made implementing experiments easier, cheaper, and faster. However, although researchers may have a conceptual knowledge of how experiments work, the actual experience of implementing a field experiment for the first time is often frustrating and time consuming. Researchers without prior experience often struggle with a number of issues such as navigating IRB, obtaining true random sampling and assignment, understanding blocking, and interpreting different types of treatment effects. The initial learning curve may be steep but the rewards are plentiful as experiments produce highly valued original data, lend themselves to causal analysis in ways that traditional survey data cannot, and become easier to implement as a researcher’s experience level increases. This talk will introduce social scientists to the basics of a particular type of field experiment -- the correspondence audit -- and walk through a number of design issues that first time users often struggle with. Dr. Gaddis will provide practical examples from his own and others' work to illuminate some of the pitfalls of this method and help the audience gain confidence in embarking on their own field experiments.
Covering the 2020 Election and the Lessons of 2012 and 2016
(Indiana University Workshop in Methods, 2019-10-04) Bump, Philip
Philip Bump is a National Correspondent for the Washington Post.
Creating Computational Interfaces
(2018-03-20) Michael, Scott; Gniady, Tassie
This workshop will focus on Jupyter/Zeppelin notebooks and Shiny R interfaces, and how to develop and where to run them.

Browse

Browsing Social Science Research Commons by Title