data literacy

Physical data visualization 2: The email data scarf returns

Emily’s email data scarf draped over the back of two chairs.

In December last year, I made a post about tracking how many emails I sent every day from July 27^th 2022-July 27^th 2023. This encompasses my entire first year as a professional librarian, and I’m really happy to say that the scarf I was making to embody this data is now finished! Well, I need to weave in the ends (crafters know the dread), but the actual crocheting part is finished.

Looking back on my initial post about the project, it’s funny that I mentioned it being a weekly routine to do my five rows… that responsible way of doing things did not stick. I crocheted from about March-July all in the last two weeks or so. I did keep up with entering my data into an excel sheet during that time, but between some traveling, moving, and life in general, I had to play some catch-up with actually crocheting the scarf.

I was also motivated to finish this project because I’m presenting it at the International Visual Literacy Association’s conference in just a few short weeks! The scarf and its color key will be part of a poster presentation. It’s also a chance to really dig into why I did this – and what changes, if any, it led to in my email behavior more generally.

At first glance on the scarf, the beginning of my year at Salisbury had a lot more emails. I used mail merge twice over the first few months (indicated by bobble stitches as opposed to single crochet), and there are two additional rows of red indicating that I sent over 12 emails that day. There’s also more pink tones in the first semester, which indicates 6+ emails being sent on any given day. After winter break, though, the color clearly shift more towards the purples, which stand for 5 emails or less in a certain day.

There are two long stretches of grey, which indicate when I was off: during winter break at Christmastime and at the end of May, early June. I sent at least one email while I was off both times, which you can see by the row of white in between the grey. White isn’t a common occurrence throughout the scarf – I’d say based off my own feelings that my work-life balance is generally quite good, and this visualized data proves that! I do have to send an email or message occasionally while out of the office, mainly due to the fact that I supervise student workers who are here in the evenings and on weekends.

On workdays, the average amount of emails I sent per day was 3.8. My counts were as follows, where the left is the number of emails sent in a given workday, and the right is the number of days that number occurred. The color it represents is in parentheses.

Number of emails	Number of days
0 (Brown)	15
1 (Brown)	22
2 (Dark purple)	40
3 (Dark purple)	37
4 (Medium purple)	32
5 (Medium purple)	34
6 (Light purple)	11
7 (Light purple)	7
8 (Light pink)	6
9 (Light pink)	7
10 (Bright pink)	2
11 Bright pink)	3
12+ (Red)	7

This is based on email threads. The data quickly got unwieldly when going by the strict number of emails (not to mention Outlook makes this sort of counting difficult), so I chose to do threads instead. If I replied twice in one day to a thread about finals week, for example, that would only be counted once.

I’m in the process of creating the poster now, and I’m really excited to talk to more folks about it at UIUC on October 6^th! I was too excited to have actually finished the data object to wait to post until I was totally finished with the poster. 🙂

New Year, New Weed

I think a lot of us have New Year’s resolutions or goal-setting on our minds as we start the spring semester, but this time of the year has me thinking more about our fiscal year goals. Heading into January means that we’re wrapping up the second quarter, and we can evaluate how the collection is measuring up to goals that were set before I started. The best way for me to determine progress is by looking at the data, and the most effective way to share that with my colleagues is through data storytelling. I’m still growing my data literacy, but narratives (the storytelling part) I can do.

One of the action items for our strategic plan is to incorporate new tools for assessment. I recently found out about Dossiers from BLUECloud Analytics, a SirsiDynix tool that is powered by Microstrategy to pull data and create visualizations. Using knowledge I gained from a Learning Analytics course at Mizzou during my MLIS, plus from consulting books like Storytelling with Data, Data Science for Librarians, and Data Visualization: A Guide to Visual Storytelling for Libraries, I crafted a brief presentation as an update to the annual collection report. Honestly, compared to other programs like Tableau, this Dossier was tough to make. Although, between creating it and writing this post, they have upgraded their system to include new features that I would have loved to use. I spent a lot of time figuring out the system, making the visualizations, and creating a visually appealing template. Besides finding out how extra I am, I think my colleagues had an easier time understanding the data, and gained a better understanding of where we stand. This is a small start towards incorporating data storytelling into our work culture.

Page of BLUECloud Analytics Dossier from ERAU

The biggest takeaway from this project was that deselection of materials had a largely positive impact on the age of the collection, greater than just adding brand new materials could. It’s like trying to mix a grey paint; you’re going to need to dump a whole lot of white onto your black paint to get it to lighten up. It’s so much more effective if you take all the old, unused stuff away first. Committing to keeping up with how we are progressing towards our goals is the only way I would have found out that the time invested by liaison librarians into collection development has been paying off – and more importantly, just how much of an impact their actions made. I think it is so much more valuable to see that quantitative comparison in the data than to simply say “good job.”

There is an IMLS project coming out of the University of Illinois Urbana-Champaign for a “Data Storytelling Toolkit for Librarians” that I am really excited to learn more about. With a resource like that, we can all learn more on how to gain insights from our data, and especially how to share our impact with our stakeholders, whether they be internal or external. When people ask me what the most beneficial classes during my MLIS were, I always list Learning Analytics among them. We live in a data culture, and in my first year as an academic librarian, I am definitely seeing how it is starting to seep into my everyday work.

Physical data visualization & data literacy

It’s hard to believe that my first semester as an academic librarian is nearly over. Scheduling the reference desk and hiring a new student worker for spring semester have taken a lot of my attention these past few weeks. I’ve started a personal project based on my work in the meantime though, which was inspired by this book chapter by McDonough and Lemon. Essentially, they created physical artifacts of their work data – specifically, the number of meetings they had – and it helped them become more mindful of their work life balance. I decided to track my own email data and crochet a scarf based on the number of emails I send in a day. I’ll do this for my entire first year as a librarian, so I did go back and record how many emails I’ve sent every day since July.

There are a few notes about the project: I only counted different email threads in the emails I sent per day, not the number I sent within a thread (so if I had an email about finals week planning and I replied twice on Thursday, once on Friday, it would be recorded once per day). My color key is as follows:

Off: grey
Weekend: White
12+: Red
10-11: Bright pink
8-9: light pink
6-7: Light purple
4-5: Medium purple
2-3: Dark purple
0-1: Dark brown
Bobbles: Mail Merge was used

The next question you might have is: why bother recording this data in a physical format? I wonder about this too, so I wanted to try doing it myself. I am endlessly fascinated by this phenomenon of using crafting and other physical forms to track one’s own data. We see it in the trend of temperature blankets (knitting or crocheting a row a day based on the average temperature wherever you are), mood tracking in bullet journals, and embroidering an icon a day for a year. Why are people compelled to do this? How does it contribute to mindfulness? For me, it is a nice routine to crochet my five rows for the week on Friday or Saturday.

After that, you may ask: what does this have to do with your librarianship or higher ed? For one, it’s helping me track my own work-life balance. I have to crochet a new row of white every time I send an email on the weekends or on a day off, so that makes me very aware of how often I’m even checking my Outlook inbox. I am also the liaison to our new data science major, so I hope to someday bring in collaborations with my faculty and students that focus on these physical visualizations as a concept. Beyond that, it’s very much a personal interest. I took Data Storytelling for my masters’ degree, and my time at the Library of Congress as a junior fellow really focused on data as well, so it’s something I want to continue to explore and nurture.

Data literacy is also at the forefront of my mind when I’m creating and editing lesson plans. I like the definition that Carlson et al (2011) put forth: “data literacy involves understanding what data mean, including how to read charts appropriately, draw correct conclusions from data, and recognize when data are being used in misleading or inappropriate ways.” That’s a big ask, especially when we as academic librarians are often trying to fit as much information literacy as we can into one session. I think data often has this connotation of being accurate, factual, or inherently correct; and though this is slowly changing, it’s also a bit scary as a concept. Big Excel sheets with rows and rows of data would frighten anybody without the knowledge to dive into and interpret that data.

The idea of “big data” and machine learning is in our faces all the time, too. To me, a crucial part of data literacy as a concept is remembering the real human beings behind the data – or data humanism, a concept by Giorgia Lupi. This is part of understanding what data mean, as per Carlson et al’s definition. The number of emails I send is one thing, but it’s attached to me as a new librarian, a white woman, a new Maryland resident… my list of positionalities can go on. The same goes for any institutional data we collect on students, for charts from a database like Statista that supports a student’s final presentation topic, and the like. Even though it often doesn’t have personal identifiers, that data came from someone.

Perhaps that’s the power of this slow data visualization; taking the time to record how many emails I send in a week isn’t revolutionary like some data viz projects are, but it is forcing me to appreciate the work that goes into data collection and the humans behind it. Data literacy isn’t just knowing how to collect, find, or process data; it’s reflecting on where, and who, it came from.