Power BI: Data Wrangling and Fish

Danielle Degiorgio, Digital and Information Literacy Project Adviser, Edith Cowan University Library
Sue Khoo, Librarian (Digital and Information Literacy), Edith Cowan University Library

What is Power BI?

Power BI is a Microsoft data visualisation tool that displays data in an easy-to-read format and allows users to interact and show relationships between different data sets.

Why Power BI?

Our goal was simple, we wanted to connect the mapping of digital and information literacy skills across the course curriculum to the teaching and learning activities we were doing each semester. We just had one problem, we were recording our data and statistics in multiple spreadsheets.

As luck would have it, the 2021 VALA Tech Camp was hosted at Edith Cowan University Library that year and we were introduced to Power BI through a series of workshops. Soon after we decided to use Power BI to help us keep track of student statistics in a more visually appealing way and it let us connect multiple sources of data. This meant we could compare, filter, and visualise relationships between multiple spreadsheets which allowed us, and more importantly our manager, to see our progress across courses.

Power BI: Visualisation of mapped digital and information literacy skills in courses.

Things we got Power BI to do:

  • Connect information from multiple spreadsheets to show how much digital and information literacy skills coverage we have in each course. 
  • Filter and display subsets of data. 
  • Be hosted in Microsoft Teams for ease of access where the report can be shared and displayed as a tab in Teams. 
  • Automatically update data from SharePoint, so having all our sheets hosted on SharePoint / Microsoft Teams mean we can easily add data into the model. Our Power BI reads directly from SharePoint files and updates at 9am every day.

Skills: What magic do you need?

  • Spreadsheet and table management – Power BI relies on external data. You must have the data cleaned and stored in a data source such as Excel (or databases such as Salesforce or Access).
  • Logic and relationship management – Connections can be 1-1 and 1-many but only one model may exist at a time. If there are conflicts Power BI will complain.
  • Ability to play with formulas and data types – If you need a relationship that isn’t expressed in the Power BI map you will need to learn to write the formula for it.
  • How to put together a graph – Knowing what graph suits your needs be it a scatter plot, ribbon chart, pie chart, or fish.
  • Professional Google skills – If something goes wrong, be ready to Google it!

Pitfalls: What to watch out for

  • A lot of trial and error and Googling – No training will prepare you for what you want to do. There may be things you want to do but Power BI only gives you the basic tools. You will have to build what you want from there.
  • Broken or dirty data – Power BI relies on relationships between different tables and inputs to build the model. If a piece of information is missing and if that is the connection in the model, it will skip that line. This has resulted in expectations not meeting what was displayed.
  • Know your data story – Power BI does not do data interpretation. You need to know what you want to tell. This is one of the main issues on the final display of information.
  • Permissions – Our shared spreadsheets and the dashboard were stored in places where we didn’t have full access to use. Arrange the files so each input has the right permissions to do SharePoint integration.

How do you get started?

But what about the fish?

The most important thing to remember is to be creative and have fun with your data!  

Power BI: Number of students seen per School using the Enlighten Aquarium visual. Enlighten Aquarium won a people’s choice award for the ‘Power BI Best Visual’ contest in 2016.

A new direction: Our journey creating a chatbot

By Bryony Hawthorn, Information Services Manager, University of Waikato Library, bryony.hawthorn@waikato.ac.nz

Background
The University of Waikato Library has been using a live chat service successfully for more than 14 years. This is a very popular service with students – and that was even before the pandemic flipped our lives upside down!

In 2019 library staff numbers were reduced, and we realised we may not always be able to staff the live chat as we have done in the past. This led to the idea of a chatbot.

Chatbot box. University of Waikato Library.

Meet our chatbot, Libby
We chose to build our chatbot using the LibraryH3lp platform as we already use this for our live chat service. So bonus = no extra costs! We named our chatbot Libby.

Libby’s interface is similar to live chat so it creates a consistent experience for users. The only difference is the colour: green for live chat and orange for the chatbot.

We create the responses that Libby sends. The chatbot administration back end has been set up to be simple to use and this means library staff creating responses don’t need to be tech experts. We’ve chosen to focus primarily on library-related topics.

Bumpy beginnings
Libby was very basic when we started. We struggled to get her to reply to keywords (the user had to type the EXACT word or phrase we had in our response bank) and she couldn’t return multiple responses to a single question. Because of this, Libby’s most common response was, “Sorry, I could not process your request. Please try a different word or phrase”. Let’s just say it was a bumpy beginning and a frustrating experience for our early users.

Stepping up
The road became a lot smoother when we introduced a natural language toolkit. This included:
● Text filtering – keywords can appear anywhere in a user’s question so no need to type an exact phrase anymore.
● Removing stop words (e.g. a, at, the, not, and, etc).
● Tokenizing – isolates words so they are compared separately.
● Stemming – allows for different endings for keywords.
● Synonyms – increase the range of words that trigger a response.

We also improved the way Libby greets users and made it clear how to receive help from a person. Most recently we added a module to assist with spelling errors.

One of our biggest successes has been introducing a prompt to encourage users to type their email address if they want a follow up from a librarian. Prior to Libby’s introduction, if the chat service was offline, users were told to email the library for assistance. This didn’t happen very often. But now users find it easy to add their email address and thus allow us to contact them. This has markedly increased the number of users receiving further help.

Example chat with chatbot. University of Waikato Library.

What we learned along the way
● Don’t do it alone. Use those around you with the right technical experience.
● Simple fixes can make a big difference.
● Make it clear to your users they are chatting with a bot who won’t be able to answer everything.
● Make it easy for users to request a follow up from a librarian.

Libby is still a work in progress and our journey is ongoing. Who knows where the road will lead. There are other ways to build a chatbot and some are simpler than what we have done. If you are interested in creating something similar, do look around for options to find something that will suit your needs.

If you’d like to learn more about our journey so far, you can watch our presentation from the LearnFest2021 conference.

Building a digital skill set with Aus GLAM Blogs

By Hugh Rundle, Manager, Digital Innovation, La Trobe University Library

Several years ago I created an aggregator service for blogs by Australasian GLAM (Galleries, Libraries, Archives, Museums) workers. Initially this was a simple Twitter bot, but later I built a web application that eventually allowed blog authors to register their blog, and readers to search by keyword, browse by tag, or subscribe to new posts via RSS or directly into their Pocket list, including the ability to filter out content they may be less interested in. If you’re interested in what Australian librarians are thinking and talking about, this is now a useful place to look.

One of the obvious questions for people wanting to start independently learning computer programming skills is where to start. The best advice I was ever given was to start by working on a real project you want to see exist, or contributing to an existing project you like. Aus GLAM Blogs was one my first “real” coding projects. I had unsuccessfully tried to teach myself some kind of computer programming for a couple of years, but it didn’t “stick” until I had something tangible to work on. Having a real life project to work on – especially one that was operating in public – really helped provide an incentive and focus to develop and practice the skills I needed to complete the job. The first version was quite rudimentary – a simple text file of manually-entered RSS feed URLs, some JavaScript loops, and a Twitter account back in the days when Twitter API keys were very easy to obtain within a couple of minutes. It barely worked at all, but it was something I thought would be helpful to bridge the gap between library bloggers looking for an audience and Library Twitter looking for good local content.

Last year I wrote about re-building Aus GLAM Blogs from scratch when I had developed more knowledge and skills. Developing the app in incremental stages meant that it wasn’t completely overwhelming. This sort of project-based learning approach can be used in many contexts, but is particularly useful when building your digital skills. Creating a web application meant I needed to host it somewhere, which led to learning about Linux server management. Gradually increasing the scope of the application led to developing an understanding of how databases and software applications interact. Re-writing the entire thing led me to consider problems of data normalisation and to what extent it is useful and acceptable.

Scaffolding my own learning like this has enabled me to slowly build a technical skillset around computer programming and server management, and think more deeply about the sort of data management questions colleagues working with library metadata have to deal with every day. I will never consider myself an “expert” in coding or server administration, but through a personal project I’ve been able to build my knowledge over time. For a while this was simply to amuse myself, but I now find myself in a library job where these skills are really useful and help me to look at problems in a different way.

Your own interests may be different. Perhaps you want to be able to make animated videos, or build your own computer from parts, or fancy being the in-house Excel macro expert. If you’ve been telling yourself that you’re “just not a technical person” or you will “never be able to do it”, I don’t believe you. The trick is to find some small projects – ideally personal ones where there are few consequences of failure – and work on them because you’re interested in the actual thing you are making. It’s surprising how much you can learn “accidentally” just by focussing on what you want to create rather than on the skills themselves. Then, simply increase your ambition for the next project, and the next one, and the next one.

Oh, and don’t forget to share it in a blog post so the rest of the library world can share your learning journey!

Enabling high-quality, scalable and consistent design: supporting library staff through a library design system

By Sarah Fennelly, Digital Designer, Deakin University Library

Our problem

Deakin University Library’s Digital Experience (DX) team is charged with delivering high-quality, consistent and scalable digital solutions across a range of library platforms and channels. However, we found ourselves with multiple challenges in delivering good on this:

  • There wasn’t a library-specific design reference for our developers, designers and digital content managers to help them manage consistency of look, feel and functionality across our digital library ecosystem, and to maintain accessibility, brand and design standards.
  • We often ‘reinvented the wheel’ when it came to digital design; library staff presented with differing design requests and needs, and the DX team often ended up creating bespoke designs to accommodate this.
  • At other times, librarians created their own design elements with little or no input from the DX team, and with limited understanding of design best practice, accessibility and brand requirements.

Our Solution

Our solution was to create the Library Design System (LDS), a website that provides the source of truth for library-specific design elements used across our digital ecosystem. The LDS references and extends upon existing university-wide design guidelines and tool kits to provide targeted and more library-focussed design direction.

The LDS is divided into two main areas:

  • Referencing guide for the library’s professional designers, developers and content managers to use across web interfaces and systems.
  • Design assets and templates created to specifically help librarians and library staff without professional design expertise.

Referencing guide

The sections used predominantly by the professional design, developers and content managers are:

  • Components: These are a selection of interface elements that can be reused across the library’s digital ecosystem. These elements include buttons, colour, forms, grids, headers and footers, logos, navigation, tables and typography.
  • Interface inventory: A table that identifies library systems, platforms, products and applications with a student and/or researcher interface.

Design assets and templates

The sections used to help librarians and library staff without professional design expertise are:

Deakin Library page showing example of characters and graphics that can be used. character posing in different postions, with both professional and student attire and with different versions of character.
  • Characters: A suit of characters developed to represent our Deakin cohort.Graphics: Multiple sets of library and faculty specific graphics.
  • Graphics: Multiple sets of library and faculty specific graphics.
screenshot of various graphics available to be used from the learning design team.
  • Iconography: Currently 189 icons in branded accessible colours in both SVG and PNG formats.
  • Templates: D2L banners, photographs and the advanced search graphic.   
  • Print: Deakin printer details.
example of scenes with 4 steps.
Step 1 adding background
step 2 add second layer image (information desk)
step 3 add third layer image (bookshelf)
step 4 add fourth layer image (couch)
  • Scenes: A series of PNG graphics that staff can use to build their own images or scenes, by layering the graphics in provided H5P, PowerPoint or Adobe Illustrator templates.
  • Video and audio: Deakin branded top and tail videos, videos created by the DX team, in conjunction with the library Learning and Teaching team and the library communications manager.

The future

The LDS is not a ‘set and forget’ resource; it will continue to adapt as a design resource to support all library staff. We will continue to maintain, update and grow the LDS to reflect changing university and library brand needs and changes in the library’s complex digital ecosystem.

Good design is imperative in our complex digital ecosystem. Design that reduces cognitive load through visual and functional consistency, allows our students to focus on learning and understanding the course content we deliver.

By Sarah Fennelly, Digital Designer, Deakin University Library, s.mccormick@deakin.edu.au

Not another 23 Things!

By Dr Karen Miller, Coordinator, Learning Success, Curtin University

There’s no doubt that 2020 was a rollercoaster year, as we zoomed up and down steep, pandemic-shaped learning curves, one after the other.  For myself and my colleagues at Curtin University Library, implementing a new student program added to the wild ride.

’23 Things’ is Curtin Library’s online digital dexterity program that was essentially created by students, for students. It is an open, shareable resource that can be easily re-used and adapted. In this post, I’ll tell you a bit about it, and how it came about. 

What is 23 Things?

 I’m sure many of you are familiar with the 23 Things model for online learning. The first 23 Things was created in 2006 as a professional development activity for library staff. Regular blog posts introduced participants to a different digital technology each week, and invited them to try it out and to share their thoughts. Since then, hundreds of 23 Things programs have been developed and adapted to suit a variety of audiences and contexts.

Curtin’s version of 23 Things 

After consultations and environmental scans to ascertain the best approach to help our students develop digital competency, Curtin Library decided to adopt the 23 Things model. We initially intended to simply re-use and adapt the excellent (and open) version from Edinburgh University

However, when some HEPPP funding became available, the project became incorporated into the Library’s ‘students as partners’ program. We employed a diverse group of students to create new content including interactive activities, videos and infographics. Bringing student voices to the forefront and fostering a peer to peer learning approach brought unique perspectives to the program and helped make the content relatable and accessible as our student partners shared their knowledge and experience.  

We then collaborated with the Library Makerspace to build engagement with participants. Our student partners developed and delivered face-to-face workshops with hands-on activities to complement the 23 Things topics, and during the pandemic lockdown worked hard to translate them into the virtual realm (not an easy task). We also encouraged participants to share in the Curtin Makers Facebook Group with our weekly Creative Challenges.   

As if that wasn’t enough, we decided to experiment with “transmedia storytelling”, a communication method which involves developing a story using multiple digital platforms. We felt this would be a good way to bolster conceptual learning and illustrate how digital skills could be applied in a workplace context. Using the fictional characters and narrative developed in Curtin Library’s online referencing game Certitude, we used weekly blogs, comics and tweets to weave the ‘things’ into a story, replete with office dramas such as copyright violations, accounts being hacked and computer meltdowns. 

Reflections and next steps   

While we had a lot of fun creating the content and engaging with participants, the 23 Things program (2020 version) wasn’t perfect. With the many different elements involved, we didn’t achieve the ideal of a seamless, integrated learning experience.  However, our approach was intentionally experimental, testing different ideas to see how they landed. We gathered a lot of data and feedback that we are now using to improve the program for its second iteration in 2021.

This year, 23 Things is part of Curtin Extra, the University’s extra-curricular credentials program, while also remaining open to anyone in the community who would like to participate.  We’re interested in finding out how best to keep participants actively engaged and how best to assess and demonstrate learning and impact. 

23 Things can be re-used and adapted

Our 23 Things program has been licensed under a Creative Commons license (CC-BY-SA).  Each module or ‘thing’ has been created as a single H5P file, and thus is very easy to download, re-use and modify.  The resource is available from the Digital Dexterity Educators Group on OER Commons, and I hope to add some supporting resources in the near future, including a document that maps each module to the CAUL Digital Dexterity framework.  If anyone would like to find out more or have a chat about the program, please feel free to contact me at karen.miller@curtin.edu.au or on Twitter @infoliterati .    

Day 5 of Championing the Digital Dexterity Framework Virtual Festival – Gaming together to digitally connect

By Sarah Howard, Associate Director, Library, Queensland University of Technology, Kat Cain, Manager Digital Literacy Programs, Deakin University Library and Nica Tsakmakis, Senior Librarian, Library Academic and Research Services, Australian Catholic University

Friday often brings with it work fatigue and dimming of enthusiasm. We have been staring at our computer screens for countless hours all week. Sure we need to, but we also have to balance that with digital wellbeing. Luckily the final day of the CAUL Digital Dexterity virtual festival landed and boy did it revive us! 

via GIPHY

What’s more – it clearly responded to the Digital Identity and Wellness area within our CAUL Digital Dexterity Framework.  

The session kickstarted with a fun presentation by Deakin Library’s Jane Miller. Jane shared real practice examples of games or other activities that have facilitated Library team building and group development. Fantastically, Jane’s tips and tricks had broad applicability despite the diversity of participant contexts. It’s amazing how a children’s game when used with adults builds negotiation, cooperation, imagination and having fun.

The opportunity to explore together online games Deakin Library has used was invaluable. These included Mentimeter, online jigsaw puzzles, and 9truths. Jane even demonstrated how to host a trivia game without the use of a program. At the end of the session participants shared their own favourite games through mentimeter poll. The collated list of games will be published soon, but it was clear that board game arena was a favourite.

Throughout the day we discovered so many benefits of workplace gaming. What stood out was the growing need for virtual socialising in our post-Covid work world. Games and activities help strengthen all types of teams, including the high performing teams who have those online awkward silences.

If you missed the session watch the recording below. Do take some time out of your day and take a peek at what was shared. You won’t regret it!


DigiDex – Championing the CAUL Digital Dexterity Framework – Day 5, Friday 12 February 2021

open video in new tab


Can OERs address Teaching and Learning challenges in Higher Education?

By Frank Ponte AALIA (CP) Academic/Research, Manager, Library Services (Teaching), RMIT University Library, frank.ponte@rmit.edu.au 


Open Educational Resources (OERs) are teaching, learning and research materials that are published under creative common licenses. These licenses specify how materials can be used.  OERs can include textbooks, curricula, syllabi, lecture notes, assignments, tests, projects, audio, video and software.

Driven by a reduction in international student arrivals, the sharp pivot to emergency online learning and a prolonged period of reduced revenue, universities are now pondering a new normal and envisaging a hybrid learning future. OERs have intersected and underscored many of the teaching and learning challenges faced by higher education institutions and advocates are now considering the efficacy of OERs to address these challenges.

Some of the Teaching and Learning concerns that OERs can address are outlined below:   

  • Student Affordability - Every student begins the semester with the same learning materials and has free digital access. Check these platforms and begin your search.  
  • Ongoing access and accessibility – Free access to learning resources in perpetuity. Access to resources in multiple formats. (select download this book to uncover the formats available.) 
  • Equity - Studies have found that using OER resources in coursework can increase grades for all students, but more so for low socio-economic students and ethnically diverse students.  
  • Retention - Studies suggest that students using OER textbooks were more likely to complete their course than those using commercial texts. 
  • Deeper Learning using Open Educational Practices – Studies suggest that OERs provide academic staff with the flexibility to customize the curriculum altering the student learning experience to achieve deeper engagement.  
  • Cultural diversity: OER’s can be used to reflect the diverse student voice. They can include gender neutral language, reflect the diverse names of students who make-up the classroom and include first nations representation and recognition.

Higher education has been challenged by many social justice issues in 2020, accelerated and reinforced by COVID.  The concepts above are some of the emerging issues that are being examined in higher education. Studies are now suggesting that OERs have a larger part to play in stemming the social justice disparities to create an even playing field for all students. Libraries are at the forefront of this fight demonstrated by this open letter in the United Kingdom signed by over 3000 Librarians, academics, and students to investigate the academic Ebook market and its practices. 

In the next post I will explore the ways to promote and advocate for OER use at your home institution.