2017-2018 Workshop in Methods

Fall 2017

Friday, September 1, 2017

Dr. J. Scott Long, "Reproducible Results and the Workflow of Data Analysis"

2-3:30pm, Social Science Research Commons Grand Hall (Woodburn Hall 200)

Many disciplines are paying increasing attention to reproducible results. The fundamental idea is that other scientists should have access to your data and be able to obtain the same results—this is reproducibility. More generally, your results should be robust so that other scientists can confirm your findings using other data. Increasingly journals require authors to provide their data and analysis file before a paper is accepted to verify that that results. Producing reproducible results is highly dependent on your workflow for data analysis. This workflow encompasses the entire process of scientific research: Planning, documenting, and organizing your work; creating, labeling, naming, and verifying variables; performing and presenting statistical analyses; preserving your work; and ending with reproducible results. Most of the work in statistics classes focuses on estimating and interpreting models. In “real world” research projects, these activities may involve less than 10% of the total work. Professor Long’s talk is about the other 90% of the work. An efficient workflow saves time, introduces greater reliability into the steps of the analysis, and generates reproducible results.

Dr. Long is Distinguished Professor and Chancellor’s Professor of Sociology and Statistics at Indiana University.

Materials on IU ScholarWorks (presentation slides, event flyer) 

 

Friday, September 15, 2017

Dr. Emily Meanwell, "Qualitative Research Methods: An Introduction"

Social Science Research Commons Grand Hall (Woodburn Hall 200), 2pm

What is qualitative research? When and how might qualitative research methods help you answer your research questions? This workshop will provide an overview of qualitative research methods, including interviews, focus groups, ethnography, and qualitative content analysis. For each, we will discuss the basics of data collection and analysis, best practices, and illustrative examples. Lastly, we will consider how these methods contrast with, and/or may be combined with, other empirical research methods, including quantitative and survey research methods.

Emily Meanwell is the Director of the Social Science Research Commons and the Study Director for the Sociological Research Practicum. She received her PhD in sociology in 2014 and has used a range of qualitative methods, including interviews, ethnography, and content analysis, in her research.

Materials on IU ScholarWorks (presentation slides, event flyer) 

 

Friday, September 22, 2017

Dr. James Ziliak and Dr. Charles Hokayem, "Kentucky Research Data Center Infosession"

Social Science Research Commons (Woodburn Hall 200), 2pm

The Kentucky Research Data Center (KRDC) is a collaboration between the University of Kentucky and the U.S. Census Bureau established by a grant from the National Science Foundation in 2016. KRDC is part of the nationwide system of Federal Statistical Research Data Centers whose mission is to expand the data infrastructure available to qualified scholars and students with approved projects by providing access to restricted individual- and firm-level data from participating federal statistical agencies. KRDC is maintained by a regional consortium of leading research institutions, including Indiana University. This infosession is designed for IU researchers interested in developing research projects using the KRDC.

James Ziliak is the KRDC Executive Director, Carol Martin Gatton Endowed Chair in Microeconomics, and the Founding Director of the Center for Poverty Research at the University of Kentucky. Charles Hokayem is the KRDC Administrator with the United States Bureau of the Census.

Materials on IU Box (presentation slides, event flyer)  Video  

 

Friday, September 29, 2017

Dr. Robert Calin-Jageman, "The New Statistics and Open Science: How to Get Started"

Social Science Research Commons (Woodburn Hall 200), 1-3pm 

Recently the Association for Psychological Science revised its publication guidelines to reward Open Science practices and to encourage the use of the “New Statistics” as a better alternative to null hypothesis significance testing (NHST).  Other journals and professional societies seem to be moving in the same direction, often in collaboration with funding agencies.

This workshop will provide a practical introduction to the New Statistics and some emerging Open Science practices.  We will worth through examples from several common research designs.  We will also explore resources that can help you adopt these approaches in your own research.  

Robert Calin-Jageman is a professor of psychology and the neuroscience program director at Dominican University. He has taught statistics and mentored students in psychological science for 10 years, publishing with 16 undergraduate co-authors (so far). His research focuses on how memories are formed and forgotten. He has also been active in exploring the replicability of psychological science and promoting Open Science. He received his PhD in Biological Psychology from Wayne State University.

Materials on IU ScholarWorks (presentation slides, demo, event flyer)   Video 

Introduction to the New Statistics website 

Friday, October 13, 2017

Jefferson Davis, "The Grammar of Graphics: An Introduction to ggplot2"

Social Science Research Commons (Woodburn Hall 200), 2-3:30pm

In The Grammar of Graphics, Leland Wilkinson created a systematic way to think about statistical graphics and the presentation of quantitative data. The package ggplot2 by Hadley Wickham implements of Wilkinson's system for the language R.

The talk will discuss the following:

- What are statistical graphs and what are some ways to talk about them?

- Examples of common plot types done in ggplot2.

- Using ggplot2 to display common statistical transformations.

- Grouping and faceting data in ggplot2.

- Using themes to add polish to graphs.

The talk requires no familiarity with ggplot2 or other libraries. Some familiarity with base R, however, will be useful.

Jefferson Davis is a software analyst with Research Analytics. He has worked on several projects that have used R and R libraries. Current projects include work in data visualization, machine learning methods, and large data sets.

Materials on IUScholarWorks (presentation slides, exercise files, event flyer)    Video

Friday, December 1, 2017

Justin Wild, "Structural Equation Modeling in Open-Source Software"

Social Science Research Commons (Woodburn Hall 200), 2-4pm

Structural Equation Modeling (SEM) offers flexible statistical models for the social science researcher. A variety of software packages are available for implementing SEM with researchers’ datasets and are becoming increasingly sophisticated. This talk will briefly outline SEM in comparison with more familiar statistical models (such as linear regression) and review several R packages tailored for the SEM community. In addition, these packages are compared to perhaps the most well-known commercial package available, MPlus.

Justin Wild is a doctoral candidate at Indiana University studying Education Policy Studies, with a concentration in International and Comparative Education, and Inquiry Methodology. He has a Master’s of Arts in African Studies and a Master’s of Public Affairs from Indiana University. His research addresses K-12 language of instruction, language ecology, foreign language learning, evaluation in education, large-scale assessments, multi-method research, and cross-cultural issues. His geographic focus is Tanzania, East Africa.

Flyer   Presentation slides  Exercise files    Presenter Email  

Spring 2018

Friday, January 19, 2018

Dr. J. Scott Long, "Reproducible Results and the Workflow of Data Analysis"

2-3:30pm, Social Science Research Commons Grand Hall (Woodburn Hall 200)

Many disciplines are paying increasing attention to reproducible results. The fundamental idea is that other scientists should have access to your data and be able to obtain the same results—this is reproducibility. More generally, your results should be robust so that other scientists can confirm your findings using other data. Increasingly journals require authors to provide their data and analysis file before a paper is accepted to verify that that results. Producing reproducible results is highly dependent on your workflow for data analysis. This workflow encompasses the entire process of scientific research: Planning, documenting, and organizing your work; creating, labeling, naming, and verifying variables; performing and presenting statistical analyses; preserving your work; and ending with reproducible results. Most of the work in statistics classes focuses on estimating and interpreting models. In “real world” research projects, these activities may involve less than 10% of the total work. Professor Long’s talk is about the other 90% of the work. An efficient workflow saves time, introduces greater reliability into the steps of the analysis, and generates reproducible results.

Dr. Long is Distinguished Professor and Chancellor’s Professor of Sociology and Statistics at Indiana University.

Materials on IUScholarWorks (presentation slides, event flyer)

 

Friday, January 26, 2018

Dr. Richard Ball and Norm Medeiros, "Documenting Quantitative Research for Transparency and Reproducibility:  Principles and Standards"

2-3:30pm (talk) and 4-5pm (hands-on workshop), Social Science Research Commons Grand Hall (Woodburn Hall 200)

This talk will present a set of standards for the replication documentation (data, code, and supporting information) that authors should assemble and make public when they release studies reporting the results of research based on analysis of statistical data.  We will begin from first principles:  What purposes is replication documentation intended to serve?  And what must be true of the contents and organization of the documentation for a study if it is to fulfill those purposes?  We will then describe how these general principles are embodied in the particular documentation standards we propose.  Further discussion will include: (i) a comparison of our proposed standards with existing guidelines, such as TOPS, DA-RT, the BITSS Manual, and the "data policies" that have been adopted by a number of prominent journals, (ii) using the Open Science Framework (OSF), an on-line file management platform, for assembling and sharing replication documentation, and (iii) the curricular resources that are being produced by Project TIER for teaching and learning reproducible research methods.

Individualized hands-on workshop:  After their talk, Norm and Richard will be available to work with anyone interested in a more in-depth and hands-on exploration of any related topics.  Ex ante, our agenda is to go into more detail about using OSF--as a tool for managing a reproducible workflow for an individual project (like a doctoral dissertation or any empirical research paper) and/or as a platform to have students use for class exercises and projects.  But the session will be informal, flexible, and driven by the preferences of the participants.  We will start by having a quick discussion to learn what people are most interested in exploring, and update our ex ante agenda based on this new information.  We have a variety of demos and exercises we could present and work through, and we would also be happy to give one-on-one consultation and advice.

Participants planning to attend the hands-on workshop may bring their own laptops or use computers available at the SSRC.  Anyone interested in getting advice about managing a current or imminent research project should bring any data, code and related files that they have already assembled for the project.

This workshop will begin about 30 minutes after the conclusion of the talk,  and continue until (i) the participants have had enough, (ii) Norm and Richard lose steam, or (iii)  IU security comes around to turn out the lights and lock the door (whichever comes first).

Richard Ball is Professor of Economics at Haverford College.  His primary teaching areas are game theory and statistical methods, and he supervises several senior theses every year.  His research has included theoretical papers on political economy and empirical work on development and social issues. He earned his B.A. at Williams College (self-designed major in cultural anthropology and African studies); his M.S. at Michigan State University (agricultural economics); and his Ph.D. at the University of California, Berkeley (agricultural and resource economics). Richard has studied or worked in Sierra Leone, Chad, Egypt and Côte d'Ivoire.

Norm Medeiros is Associate Librarian at Haverford College (PA) where he oversees the collection management and metadata services division of the Libraries. Norm has been active in the Association for Library Collections & Technical Services (ALCTS) for many years, serving as its president for the 2015-2016 term. He has been co-Director of Project TIER (www.projecttier.org) with Richard Ball since they launched the initiative in 2013.

Flyer | Add to my calendar: Outlook (iCal/.ics); Google

 

 

Friday, February 2, 2018

Post and Riposte: A Panel Discussion on Open Science, Repositories and Data Sharing in the Social Sciences

2-3:30pm, Social Science Research Commons Grand Hall (Woodburn Hall 200)

Social science journals, funding agencies, and professional organizations increasingly encourage or require the archiving and sharing of data, code, publications, and other research materials for transparency and replicability. Panelists including Dr. Andrew Brown (Applied Health Science), Dr. Timothy Hellwig (Political Science), and Jamie Wittenberg (IU Libraries) will share their insights on the benefits and challenges of open social science scholarship, and their experiences with resources available to support the sharing of data, analysis, and publication files. Following the panel, join us for a question-and-answer period and a broader discussion with the panelists and other university stakeholders about supporting social science data sharing, transparency, and reproducibility efforts at Indiana University.