Wrangling Library Data: Analytics, Dashboards, and Spreadsheets

Data is everywhere in the library world. No matter what team or department you are in, you have access to numerous data points on a regular basis. Analyzing data for the decision making process is a step in the right direction. To truly leverage the greater potential of that data, be an analytics advocate. Anyone can be an analytics advocate--it all starts by being intentional in the library data tracked and analyzed. These small steps can revolutionize how your library uses data.

You don’t have to be a graphic designer to present your library statistics in a way that effectively communicates value. In this session, you’ll learn quick and easy tips for displaying your statistics so that they tell a powerful story about your library, whether your data visualization aspirations consist of adding a few Excel charts to a board report or designing a complex infographic for your website.

Through the use of Google products, you can collect raw data and turn it into a wealth of useful information about how your space is being used and when you are busiest. In this session, you will learn about the process, tools, features, and formulas you need to create an up-to-date user-friendly website with information about how your library is being used and what your staff is doing. You will learn about the challenges faced, the solutions implemented, and how you can collect and report information about your library.

Although metadata professionals continue to perform traditional activities such as original cataloging and subject analysis, responsibilities have evolved to include metadata analysis, enhancement, normalization, and quality control. At Northeastern University, the Digital Metadata Unit works collaboratively with Library and external partners to provide access to variety of digital resources such as Special Collections and Archives materials, faculty research and university publications, media, and electronic theses and dissertations. The Unit utilizes various tools and resources such as Google Forms and Sheets, scripts and regular expressions, Open Refine, and Marc Edit, to improve and enhance the discovery experience. In terms of quality control, we have implemented rigorous, yet flexible benchmarks based on RDA standards, Digital Commonwealth Metadata Requirements, and DLF/Aquifer Guidelines. Furthermore, our Digital Repository provides robust versioning, validation, and export/import features which enable us to access multiple versions of a particular metadata record, automatically screen metadata for structural problems and issues, and bulk update thousands of records within a short period of time.

Most large libraries rely on approval plans for book purchasing, but the very nature of approval plans – the large number of books that arrive and the breadth of the criteria used to select them – can make it difficult to know if the plan is effectively meeting the library’s goals. In this case the goals were to purchase items that were expected to be used while ensuring that key publishers were represented in the collection. This presentation will show how the presenter used vendor reports, the ILS, and data visualization software to collect and combine data, with the intent of evaluating approval plans based on pricing and number of loans. Reports from the vendor showing what books had been shipped were combined with circulation data from the ILS (Ex Libris’ Alma). The analysis was broken down by subject area and publisher, with the measures under consideration being price, circulation, and number of books received. Visualizations created in Microsoft Power BI made a fine-tuned analysis possible. The presenter will explain the calculations that were made based on the data and how these were used to recommend adjustments to her library’s approval plans.

Tulsa City County Library has made many system changes in order to gather better data. This presentation will discuss how these changes have benefitted our library system, what kind of reporting we use to make decisions, and the outcomes of these decisions.

While we may know who the key players are that we connect with in departments, a full network analysis of e-mail contacts can locate the intersections between projects and researchers. Using Gelphi and Immersion, I conducted a network analysis of my e-mail in order to determine areas of opportunity and connection with my liaison department of cooperative extension, in order to answer the following questions: What patterns did I see? What surprised me? What might I do differently with my network to reach underserved groups? Networking tools are just one technique you can use to possibly uncover additional outreach and engagement opportunities.

Microsoft Excel was first released on the Windows platform 30 years ago and has since become widely used. Although new tools for manipulating, analyzing, and visualizing data are constantly emerging, Excel remains a potent tool—and not just because of newer features such as PivotTables, Data Tools, Analysis ToolPak, Sparklines, and the like. Simple functions such as TRIM, MID, SUBSTITUTE, FIND, ROUNDDOWN, and VLOOKUP can be used to manipulate data sets in powerful ways.

This session will apply selected functions to realistic library data sets. Demonstrations will include:

• deriving time-series categories from date and time stamps
• dealing with messy data points such as call numbers and publisher names
• manipulating textual data (removing extraneous punctuation and symbols, eliminating stopwords, identifying misspellings and themes, etc.)

Viewers will gain access to takeaway resources—both handouts and spreadsheets with utilitarian functions.

The goal of this session will be to develop a game plan data analysis.  We have all seen reports and presentations that are overloaded with Excel graph after Excel chart after Excel graph.  Your core message can be lost in forest of data.  The goal of this session is to help you develop a plan regarding what data is important, and how to report the library's value to stakeholders.  The presenter will use the example of her library's recent annual report that identified metrics meaningful to users, developed a plan to obtain and analyze data, and presented the findings in a single page report.  Data analysis tools used to develop this report include a time study, pivot tables, and date subtraction Excel formulas.

There is an old marketing heuristic that if you market to everyone, you are really marketing to no one. Broad messaging and “shotgun” approaches may seem like a good way to get out the message, but it tends to be very ineffective. Learn how Mid-Continent Public Library used Orangeboy to determine service markets, how to communicate with specific user types, and how to more effectively serve those users.

Expanding on their presentation at the 2017 LITA forum, Andrew See (Head, user Services) and Mike Taylor (Support Systems Analyst, Sr) will present on their efforts to collect hourly real-time occupancy counts throughout the library using the campus WiFi infrastructure and graphically display that data on a data dashboard. Presenters will speak to the advantages and practical uses of visualized data and also about the challenges in gathering data from various sources in terms of determining ownership, need, and negotiating access.

The Public Library of Albuquerque and Bernalillo County, serving some 300,000 residents of New Mexico, recently implemented an Internet-based software to track usage data for our 18 library branches. Springshare is a third party vendor behind the popular web-hosting service LibGuides, and for an extra cost provides access to LibInsight for recording and reporting statistics. The Digital Services department was tasked with: moving structure and historical data from Excel sheets to the online format; creating a process in which any staff member can enter records for daily usage of the branch; and enabling branch supervisors and one alternate to modify data and run reports periodically. Usage data includes paging slips, programming, outreach performed, reference and computer questions, volunteer hours worked, and others.

We'll show how we migrated our statistics online and answer questions about the process. We'd love to share our experience with the data, communication, and training required for the move, as well as hear what challenges other libraries anticipate as they gear up for similar projects.

The Central Michigan University (CMU) Libraries have fine-tuned their data collection capabilities and have leveraged SharePoint, Microsoft SQL Server, and Excel to store, combine, and visualize data sets. More important than the tools used, the CMU Libraries has fostered a culture of data-driven decision making that goes beyond charts and graphs. Over time, the Library has created a data portal and has equipped themselves with dynamic ways to obtain on-going feedback from users. This has created a multifaceted approach to data collection and the CMU Libraries are seeing real-time results. The Library now has the ability to help answer questions that arise in daily meetings and directly influence decision-making processes with context data and information. This knowledge and agility has allowed the Library to better understand the needs and expectations of users, while providing a support for the staff to make informed decisions and strategies. Four initiatives will be discussed that give an overview of how data collection can be leveraged to drive projects and effectively support decision-making processes. With each of these initiatives we will discuss the tools used, the process, and the outcomes:

1. The Library Computer Project: An exploration into if the Library has enough computers to effectively serve the needs and expectations of patrons.
2. The Library Website Project: An analysis of how to best update and modernize the Library website.
3. Printing in the Library: How data and usability testing helped to solve printer and wayfinding issues within the Library.
4. Google Analytics Enhancements: How Google Analytics has been combined with user-specific data dimensions.

Poor data management practices undermine libraries' commitment to patron privacy despite good ethical intentions. This talk will identify current data handling practices from published library learning analytics articles and suggest four areas of data management for improvement: anonymization procedures; security for data in transit; data retention; and scope of data collection. Examples of potential data mishandling will be presented as well as a description of the corresponding data management best practice and ideas for implementation. The goal is to improve data management practices in library analytics in order to protect patron privacy.

Librarians have unprecedented access to data about our collections, including usage and holdings data from database vendors and integrated library systems. Our institutions also collect data about our user populations. How can we harness the power of this data to make informed decisions about collection management and library services?

The Neumann Library database reviews, visualizes, and wrangles data into a dynamic, interactive tool for understanding database usage across time. It includes three years of usage data for all library databases, and a variety of usage metrics such as COUNTER, ebook pages viewed, and videos watched. It includes library data—such as source of funding, liaison, and renewal month. And it includes institutional data—such as college, program, semester credit hours, and enrollment.

This webinar will begin by discussing the rationale behind creating a visualization for the analysis of database usage: enabling the library to make better informed purchasing and renewal decisions.We will then describe the process of collecting and creating the multiple datasets that constitute the visualization. Finally, we will detail the process of creating the visualization, including the development of filters and other design decisions.