Posters - IASSIST 2015 Conference

Presenter: Florio Arguillas, Jr; Cornell Institute for Social and Economic Research (CISER)
Abstract: Less than a year after its implementation, CISER's data curation and reproduction of results service, a carrot that encourages Cornell social science researchers to share code and data associated with their publication, has evolved resulting from experiences learned along the way. This poster discusses how the service works, the issues encountered and how they were resolved, the costs involved and the factors influencing the costs; how the cost of the service has given rise to other services and trainings (data management, curation, metadata creation, code documentation and version control) that shifted labor to researchers and/or their assistants; and the supplementary post-reproduction services to promote and increase discoverability of the publications and their associated files.

Using Undergraduate Students to Provide Data Services

Presenter: Julia Bauder, Grinnell College
Abstract: Since 2010, the Data Analysis and Social Inquiry Lab (DASIL, pronounced “dazzle”) at Grinnell College has employed peer mentors to provide certain data services to students and faculty. These peer mentors--undergraduate students with backgrounds in statistics, economics, geographic analysis, computer science, and/or qualitative analysis—provide services including drop-in assistance to students working on data-driven projects in economics, GIS, psychology, and other social science disciplines; preparing data visualizations and data-driven exercises for faculty to use in their courses and datasets for them to use in their research; and providing hands-on assistance during in-class labs that involve data. The poster will cover the who, what, where, when, and why of using this model to provide data services at Grinnell, and also discuss how this experiment has worked out for us.

High Value, High Risk - Options for Restricted Data Dissemination at ICPSR

Presenter: Johanna Bleckman, ICPSR, University of Michigan
Abstract: As the body of social science research data grows, so does interest in obtaining data for replication and secondary analysis purposes. Research data involving individuals tend to be highly disclosive and often sensitive. Traditional methods of coarsening, truncating, or otherwise altering the data for public consumption limits both the utility of the study and the impact of the initial investment. There is growing interest in establishing innovative ways of sharing research data in its fullest form while minimizing disclosure risk and honoring confidentiality assurances given to research subjects. ICPSR employs three main methods of restricted data dissemination, and this poster will highlight these methods and demonstrate the features of the newest method, the ICPSR Virtual Data Enclave (VDE)-- a virtual computing environment offering full access to disclosed and sensitive restricted data that never leave ICPSR servers.

Bridging the administrative data gap: supporting researchers and data custodians at the Administrative Data Research Network

Presenter: Kakia Chatsiou, Administrative Data Research Network, UK Data Archive, University of Essex
Abstract: Researchers go an extra mile to access administrative data in the UK: administrative databases are operational records and need to be processed and documented prior to using them for research. There is no support mechanism for users of administrative data -- researchers quite often don't know how to access and analyze administrative data and have to negotiate access with depositors themselves. Support staff working in the administrative data setting supporting researchers need a variety of skills, too. Having a good knowledge of the data is not sufficient - staff need to be approachable and proficient in translating requirements, mediating and negotiating access and thinking laterally to solve any problems arising. Finally, legal pathways to access for research purposes are still in development and quite often vary between different data custodians. The Administrative Data Research Network is a UK-wide partnership between universities, government departments and agencies, national statistics authorities, funders and the wider research community aiming to bridge these barriers. This paper will describe how the ADRN User Services team have been bringing together academic, government and third sector researchers, data custodians and data professionals to enable access of linked, de-identified administrative data for research in the UK.

ICPSR: A Common Data Platform Serving Unique Data Needs

Presenter: Linda Detterman, ICPSR - University of Michigan
Abstract: ICPSR is a collection of data collections related to one another by a common data platform, common guidelines for data ingest and processing, and a common presentation of data for discovery (data catalog) and use by the data community. ICPSR's collections number almost 20 including a member (subscriber) collection, several agency and foundation collections, and most recently, a self-deposit, public-access collection. This poster will diagram and present ICPSR's shared infrastructure approach that enables research data collections to be unique from a common platform.

Bridging the Data Divide: Economical Repository Management in the openICPSR Cloud

Presenter: Linda Detterman, ICPSR - University of Michigan
Abstract: Organizations desire to meet changing federal research requirements requiring public access data sharing as well as satisfy the growing global call for replication and transparency of data analyses. However, organizations have limited resources to expend on user support, technical staffing, and the infrastructure development and maintenance needed to support public-access, data-sharing services. This environment has made hosted data sharing in the cloud a solid alternative to costly do-it-yourself (build-it-yourself) repository management. With a growing need at the organizational level for effective and experienced, but economically feasible data sharing, ICPSR researched and developed a public-access, data-sharing service for use by institutions and journals. This poster will highlight the research findings and demonstrate the features and benefits of openICPSR for Institutions and Journals, a fully-hosted, data-sharing service for use by organizations, departments, and journals of all sizes.

DANS Strategic Plan 2015-2020: Sharing data together

Presenter: Ingrid Dillo, DANS (Data Archiving and Networked Services)
Abstract: DANS promotes sustained access to digital research data. For this, DANS encourages scientific researchers to archive and reuse data in a sustained form, for instance via the online archiving system EASY (easy.dans.knaw.nl) and the Dutch Dataverse Network (dataverse.nl). With NARCIS (narcis.nl), DANS also provides access to thousands of scientific datasets, e-publications and other research information in the Netherlands. The institute furthermore provides training and consultancy and carries out research on sustained access to digital information.

Elements in our new strategic plan 2015-2020 are:

1. the federated data infrastructure in the Netherlands
2. the care for living data and
3. the need to change the business model by charging institutions that deposit data in the DANS-repository for the basic storage costs.

Driven by data, DANS with its services and participation in (inter)national projects and networks ensures the further improvement of access to digital research data. Please visit dans.knaw.nl for more information and contact details.

TERESAH - Authoritative Knowledge Registry for Researchers

Presenter: Johan Fihn, Swedish National Data Service
Abstract: TERESAH (Tools E-Registry for E-Social science, Arts and Humanities) is a cross-community tools knowledge registry aimed at researchers in the Social Sciences and Humanities. It aims to provide an authoritative listing of the software tools currently in use in those domains, and to allow their users to make transparent the methods and applications behind them. TERESAH has been developed as part of the Data Service Infrastructure for the Social Sciences and Humanities (DASISH), a Seventh Framework Programme funded project. DASISH collaborates with the five ESFRI Infrastructures in the field of Social Science and Humanities (CESSDA, CLARIN, DARIAH, ESS, and SHARE). The tools and knowledge registry is aimed at researchers from all disciplines and sectors, research infrastructure builders and users, as well as IT personnel. It aims to include information about tools, services, methodologies, and current standards and makes use of existing social media for dissemination and discussions. TERESAH is open source software and has been developed with a reusability plan in mind, meaning that anyone can install and run a TERESAH instance of their own with minimal effort required. This poster will give an overview of TERESAH's structure and features including live demos.

Overcoming Issues and Challenges Facing Social Sciences Data Services in the Near Future

Presenter: Maria A. Jankowska, UCLA Charles E. Young Library
Abstract: Social sciences and humanities data services and collections are changing rapidly due to data intensive research, new research data drivers, volume, variety, and velocity of data. This poster focuses on libraries transition from the traditional model of data services focused mostly on secondary data to a new model supporting research communities in discovering primary, secondary data, and the stewardship of research data. Additionally, the poster present the challenges facing data services and collections librarians in the near future and proposes strategies in managing these challenges.

Understanding Researcher Needs in Data Management: a comparison of four colleges in a large academic American university

Presenter: Lisa Johnston, University of Minnesota; Carolyn Bishoff, Steven Braun, Alicia Hofelich Mohr, Josh Bishoff
Abstract: The diverse nature of research makes identifying needs and providing support for data management a complex task in an academic setting. To better understand this diversity, we compare the findings from three surveys on research data management delivered to faculty across 104 departments in the University of Minnesota - Twin Cities campus. Each survey was separately run in the Medical School, the College of Liberal Arts, the College of Food, Agricultural, and Natural Resource Sciences and the College of Science & Engineering and modified to use language that paralleled the different cultural understandings of research and data across these disciplines. Our findings reveal common points of need, such as a desire for more data management support across the research life cycle, with the strongest needs related to preparing data for sharing, data preservation, and data dissemination. However, the results also reveal striking differences across the disciplines in attitudes and perceptions toward data management, awareness of existing requirements, and community expectations. These survey results can be used by others to demonstrate that a one-size-fits-all approach to supporting data management is not appropriate for a large research university and that the services developed should be sensitive to discipline-specific research practices and perceived needs.

DRUM-roll Please: Introducing an interdisciplinary data repository with a focus on curation for reuse

Presenter: Lisa Johnston, University of Minnesota
Abstract: The Data Repository for the University of Minnesota (DRUM) (http://z.umn.edu/drumposter) is a service that launched in November 2014 enabling campus researchers to provide long-term, open access to their research data. DRUM reflects the Libraries' commitment to providing broad and enduring access to the intellectual output of the University. Making research data openly available in DRUM has numerous benefits, including: the ability to provide a persistent identifier (DOI) to data for citation purposes, compliance with data sharing and preservation requirements of funding agencies, and tracking the downloads of data in order to demonstrate impact. DRUM is one of a number of library data services that include support for writing data management plans (DMPs), training faculty, staff and students in data management best practices, and digital preservation and curation of the digital objects. This poster will describe the open source architecture behind the data archive, the policies for deposit, and the importance of acceptance criteria and curation actions that are taken to ensure that data are discoverable in a way that maximizes potential for re-use.

Telling tales: The power of data stories to illustrate and reach out

Presenter: Inna Kouper, Research Data Alliance, Indiana University
Abstract: In September 2014 a librarian asked for examples of good and bad data management practices on the JISC Research Data Management mailing list. The examples were to be used in training courses and engagement efforts. Following this exchange and several suggested links to examples, a session at the Research Data Alliance (RDA) plenary (a global organization to facilitate data sharing) further highlighted the need for a repository of such examples. The repository would serve as a community resource to promote best data management practices, but also serve larger goals of effecting change in cultures around research data. Responding to this need, the UK Digital Curation Centre and the RDA Engagement Interest Group are launching a service to collect and organize stories about failures and successes in research data management, sharing, and re-use. This poster will describe our effort to date with launching this service. We will outline the framework for organizing the stories and our initiatives to collect them. We will also share preliminary results from the first round of story solicitations and highlight the challenges of making the stories useful. We hope that the poster will stimulate a discussion about education, engagement and outreach in social science data exchanges.

Implementing a Data Citation Workflow within the State Politics and Policy Quarterly Journal

Presenter: Sophia Lafferty-Hess and Thu-Mai Christian, Odum Institute for Research in Social Science, University of North Carolina
Abstract: Journals are increasingly instituting data sharing policies to encourage replication and verification of research results. Workflows that support citing and archiving data alongside the publication of peer-reviewed articles can assist researchers in receiving scholarly acknowledgement for data products and ensure data are properly preserved. In this poster, we will summarize a project sponsored by the Alfred P. Sloan Foundation and the Inter-university Consortium for Political and Social Research (ICPSR) to implement a prototype data citation workflow within the State Politics and Policy Quarterly (SPPQ) journal publication workflow. The project developed a human-driven workflow to archive, share, and link underlying replication data to their associated scholarly publications. Through the development of the workflow, the project team examined some of the challenges and opportunities of integrating data archiving and sharing into existing publishing systems. This poster will present the prototype workflow and key lessons learned such as the importance of relationships and the challenge of working with multiple system.

National Archive of Data on Arts and Culture: A Resource for Researchers, Policymakers, Practitioners, and the General Public

Presenter: Alison Stroud and Amy Pienta, ICPSR, University of Michigan
Abstract: The mission of the National Archive of Data on Arts and Culture is to share research data on arts and culture with researchers as well as those not experienced with statistical packages, such as policymakers, people working for arts and culture organizations, and the general public. Funded by the National Endowment for the Arts, the infrastructure of this data repository within ICPSR is also designed to be accessed through a computer or mobile device. This poster will help attendees to see how ICPSR has designed a repository to increase data access to the arts and culture field, a field not known to have or use research data. Methods to discover and learn about data available to download from NADAC and various techniques to explore the data online will be described. It will also highlight several user-friendly tools for analyzing and visualizing data for this wider spectrum of experienced and novice data users from the community.

Elevator: A tool for all media

Presenter: Colin McFadden and Rebecca Moss, University of Minnesota
Abstract: In today’s digital world, the creation of media is increasingly easy. In a University environment, managing that content is getting increasingly complex. For researchers, materials may at first be shared with only a select few, but eventually, they will be more broadly available. Some content will need to be preserved for the long term, other items will have a very short shelf life. The media that is managed as close to the period of creation, with the proper description by the creators, has a much better chance of being useful over its life. There are many solutions being made available through the Libraries, IT and commercial entities, but rarely do these options offer customizable metadata templates, granular access controls, and superior searching options. Elevator is one such option. It offers a flexible framework that allows content to move from a privately managed collection into more curated collections that can be more widely searched. Based on open source tools, and backed by Amazon cloud storage, Elevator is a powerful, easy to use, digital asset management tool that works for most types of research and teaching materials.

Easy DDI Organizer (EDO): Metadata management and survey planning tool based on DDI-Lifecycle

Presenter: Akira Motegi, Institute of Social Science, the University of Tokyo
Abstract: We will introduce a metadata managing software project launched at Social Science Japan Data Archive (SSJDA), which is affiliated with Institute of Social Science, the University of Tokyo. The aim of the project is to develop Easy DDI Organizer (EDO), a metadata managing software which helps researchers to edit and manage metadata based on Data Documentation Initiative (DDI) 3.1. The implementation of DDI 3.1 constitutes one of the greatest features of EDO, in which researchers can record metadata such as study purpose, sampling procedure, mode of data collection, questions, question sequence, variable descriptions, and bibliographic information along with the scheme of data lifecycle. File-import/export function is another salient feature of EDO. It supports importing variable level metadata from SPSS files and exporting codebook and questionnaire. Our poster session will introduce the contexts and features of EDO with some demonstrations provided. We will also discuss its future improvements such as public release and other function enhancements.

A Stata Course for Undergraduates

Presenter: James Ng, University of Notre Dame Libraries
Abstract: I teach a 1-credit hour undergraduate course on how to use Stata for empirical research in economics at the University of Notre Dame. I will share my experience teaching this course on these fronts: i) course logistics, ii) student outcomes, and iii) lessons learned: what worked well and what did not. My aim is twofold: to offer information that data librarians and instructors may find useful for their own data literacy initiatives, and to learn from the experiences of conference participants whose work involves promoting data literacy.

Co-circular RDM: A Pilot service for Graduate Students at the University of Toronto

Presenter: Andrew Nicholson, University of Toronto Libraries; Leslie Barnes, Dylanne Dearborn
Abstract: The University of Toronto Libraries (UTL) designed a co-curricular Research Data Management (RDM) workshop aimed at introducing graduate students to RDM principles and best practices. This poster will outline our method of developing the workshop and will detail the preliminary results gathered through student feedback. Findings presented will include the domains and divisions expressing interest in such a workshop and what RDM facets or areas of support have increased demand for RDM services at the University of Toronto. Envisioned as part of a larger initiative to supplement gaps in graduate professional skills training and resources, this workshop is an experiment in linking instruction and RDM service development in a large, distributed research university. Key areas covered include a research data overview and best practice pointers for collecting, describing, storing and sharing research data, with an emphasis on creating sound data management plans. Graduate students also learn about emerging research data policies in Canada, as well as RDM requirements implemented by other funding agencies and publishers. This workshop is being offered through the University of Toronto’s Graduate Professional Skills (GPS) program, which provides graduate students with training in areas such as teaching and advanced research for co-curricular credit on their transcript.

DDI-XSLT

Presenter: Olof Olsson, Swedish National Data Service (SND); Jannik Jensen
Abstract: Transforming DDI-L into various formats via stylesheets. This technology is currently in use at the Danish Data Archive, the Swedish National Data Service and the DDI Alliance. The poster will bring updates from the DDI-XSLT platform including METS, DISCO and other formats. With the release of DDI Discovery Vocabulary (DISCO) the project will display its mapping from DDI-L to DISCO. In this context the project is reaching out towards a JSON mapping and invite participants to come and join an informal idea generation for scope and purpose of a future JSON mapping of DDI-L. For online content see: https://github.com/MetadataTransform/ddi-xslt/wiki/ The stylesheets are released as LGPL software and are available for public and commercial use - download at: https://github.com/MetadataTransform https://github.com/linked-statistics/DDI-RDF-tools

Someday We'll Find it, the Data Connection: Information Literacy Frameworks and Data

Presenter: Kristin Partlo, Carleton College; Lynda Kellam, UNCG; Hailey Mooney, Michigan State
Abstract: What implications does the Association for College & Research Libraries' (ACRL) new Framework for Information Literacy for Higher Education have on our approach to teaching about data? Building a strong data literacy instruction program involves bridging the language, standards, and goals of data science, statistical literacy, and information literacy. Practicing instruction librarians are likely to draw on concepts, skills, and competencies from across these areas. The holistic nature of the Framework situates students as both consumers and producers of information, which ties closely to data focused learning outcomes such as those based on competencies from the Data Information Literacy program. This poster will investigate this and other parallels and divergences between the Framework and learning outcomes in data and statistical literacy. Participants will be invited to share their views and reactions on how the Framework addresses the data information literacy needs they encounter in their work.

Linking DDI to the Semantic Web

Presenter: Benjamin Perry, Cornell University Institute for Social and Economic Research
Abstract: The Comprehensive Extensible Data Documentation and Access Repository (CED2AR), is an online repository for metadata on surveys, administrative microdata, and other statistical information. CED2AR runs directly from DDI 2.5 through a single, non-relational database. While the DDI schema is well developed for documentation purposes, it is not ideal for semantic web applications. Using the schema.org microdata markup, CED2AR allows search engines to parse semantic information from DDI. The solution further enhances the discoverability of DDI metadata, as the data are machine readable to several providers such as Google, Yahoo and Bing. The schema.org markup is not directly embedded within the DDI, so it doesn't directly export when a user downloads a codebook. However, CED2AR can also run as a zero install desktop application. Users can simply download their own copy of CED2AR, quickly import codebooks, and instantly see the schema.org enhancements the system offers. The only prerequisites for the software is Java version 7, and an internet browser. This presentation will demonstrate the advantages schema.org adds to DDI, and the ease of deployment CED2AR allows.

Visualizing Social Science Research in an Institutional Repository

Presenter: Ted Polley, Indiana University-Purdue University Indianapolis
Abstract: Using text mining and visualization techniques to identify the topical coverage of text corpora is increasingly common in a number of disciplines. When these approaches are applied to the titles and abstracts of articles published in an academic journal, it yields insight into the evolution of scholarly content in the journal. Similarly, text mining and visualization can reveal the topical coverage of items archived in an institutional repository. This poster will present initial results from mining the text and visualizing the abstracts of social science research in one university’s institutional repository. Generating a topic map visually demonstrates how research in a repository clusters around specific domains in the social sciences. These topic maps are potentially useful to librarians and researchers seeking to learn more about the topical coverage of items in their repository and determine if the research is reflective of the scholarly output from an institution more broadly.

How Canadians Access Statistics Canada Research Data

Presenter: David Price, Statistics Canada; Donna Dosman
Abstract: Statistics Canada has been providing access to Canadian research microdata for 18 years. This session will explore the principles through which Statistics Canada will allow access to information, governance of the access programs, the different types of research data that are available and technological solutions used to implement researcher access to microdata. Starting with the Data Liberation Initiative and how Public Use Microdata files are distributed and used in the research community. Then following up with a detailed look at how research data can be accessed through the Research Data Centers and the Real Time Remote Access system.

Archival GIS: Discovering Gay LA through Bob Damron's Address Guide

Presenter: Andy Rutkowski, University of Southern California
Abstract: This poster outlines the development of a mapping project focused on using archival material from the ONE Archives at the University of Southern California. The project began by taking one item - Bob Damron's address book with gay-friendly bars and other destinations - and mapping out those locations. These locations were used as a starting point to explore and map other archival holdings. As more locations were mapped from subsequent address books opportunities arose for spatial analysis. The poster also discusses the possibilities of using GIS as an approach for introducing students to archival collections. #mappingtheone bit.ly/mappingtheone

Solr Cloud-working for the UK Data Service: problems experienced and a different way forward

Presenter: Jonathan Sexton, UK Data Archive, University of Essex
Abstract: Solr Cloud working is supposed to provide a fault-tolerant, high availability system, providing a stable platform for blazing fast full-text and faceted searches. A Solr Cloud implementation would typically consist of (x) number of Solr servers (where indexed data is held), (x1) number of zoo-keepers servers (used to share data between Solr servers and maintain system state) and a load-balancer server (providing an evenly distributed load). The implementation of this type of system in the UK Data Service has proved to be anything but this, with indexes frequently getting out of synchronization and an extremely high maintenance overhead required just to keep the service stable and usable. A simpler, more maintainable system was required; simple replication. A single master server provides the source for the indexed data, with a number of slave servers automatically replicating this. Simple load-balancing across the slaves is achieved with the use of a light-weight in-house software component. Extensive testing has proved that there is no detrimental effect for the speed of retrieval for queries, the system is very stable and more reliable, the maintenance overhead very low and the flexibility for alternative configurations greatly improved.

Insights into Data Recovery Projects: A Case Study from the Roper Center

Presenter: Cindy Teixeira, Roper Center for Public Opinion Research
Abstract: Recently, the Roper Center staff embarked on one of the largest data and documentation recovery projects in its history. In 2014, our host institution announced it was decommissioning the local IBM mainframe system over the next year. The mainframe had been the main access point to our IBM 3480 and 3490E Tape Cartridge collection, our main storage solution from 1977-2001. With limited descriptive and technical metadata, the Roper Center recovered over 30,000 files including data, documentation, published materials and work process files. Based on this case study, this poster will identify the unique challenges and explore important factors to consider in planning a large scale data and documentation recovery project. We will discuss the various techniques and solutions used during file recovery process, and share our recommendations for any organization engaging in such a project.

DLF E-Research Network: Developing a Sustainable Community of Practice for Research Data Services

Presenter: Rita Van Duinen, Council on Library and Information Services / Digital Library Federation
Abstract: To address the need for libraries to be engaged in developing e-research services the Council on Library and Information Resources (CLIR) and the Digital Library Federation (DLF) have developed the DLF E-Research Network; a peer-driven mentoring group focused on sharing information on implementing research data management services as well as on participant-directed learning and shared skill development. Launched in 2014, institutional teams from academic libraries in the US and Canada participated in the DLF E-Research Network, a program aimed at building a mutually supportive community engaged in continuous learning about e-research support. Through a series of in-person meetings, webinars, practical activities, and virtual discussions participating institutions were able to evaluate, refine and further implement research data services. The E-Research Network provides members working in research libraries and data centers the opportunity to develop data management strategies, policies, tools, and services. This poster will describe the DLF E-Research Network, its membership benefits and contributions to the field, as well as highlight participant’s objectives and outcomes. The poster will also demonstrate how the E-Research Network is designed to help build the capacity needed for professional development in research data services.