Exploring physical vs digital titles in our Singapore Libraries (Part III — The End)
In the past two posts [Part I and Part II], I analysed the total copies and availability of non-fiction authors selected, and the analysis suggests that while there are more total titles of digital (electronic and audio) than physical copies (Part I), the availabilities of physical copies may not always lose their digital copies (Part II).
After analysing the data from a library user perspective in my first two posts, I want to use the same data to understand the available-to-total ratio, to get a rough sense of how “efficient” our libraries in the provision of non-fiction titles that I have sampled on. As mentioned in my previous posts, I am using a small data set that isn’t representative of the state of our libraries operations. Also, these ratios are calculated based on updated numbers extracted on 26th Dec, 2023, as I forgot to save the original data extraction that I did on 13th Dec 2023 :p
Available versus Total Copies
Doing a simple comparsion of available versus total titles by formats, we can see while our libraries have a larger supply of ebooks, they are also the least available among the three formats. This does suggest again that eBooks are the most popular format among us Singapore library users.
Book available-to-borrow ratio
Next, I created a ratio taking “available copies” over “total copies”. In simple terms, the closer this ratio is to 1, the more copies are available for borrowing. This ratio focuses on the probability of library users being able to borrow a particular book from our Singapore libraries.
The average available-to-total ratios for eBooks, Books and Audio are 0.165, 0.483 and 0.461 respectively. This means that about 85.5% (1 — 0.165) of electronic books are borrowed out at the time of my extraction. This percentage is much higher than the 51.7% and 54.9% borrowings of physical and audio books respectively. This seems to suggest the popularity and ease of borrowing electronic books over physical and audio books.
Viewing our ratios through other lens
To have a a better perspective of these ratio distributions, I created a box plot by title formats. Interestingly, we can see more electronic books with zero ratios (all electronic titles were borrowed out) than audio and physical books. There is also a wider separation across their more popular and less popular electronic books (between 0.36 and 0.6 books) than physical and audio books. Lastly, while electronic books are the most popular across all three mediums, there are still electronic titles that are not well borrowed, but such electronic books are fewer than physical and audio titles.
Box Plots by Authors
The box plot ratios by authors showed even more interesting patterns. Authors Adam Grant and Malcolm Gladwell had the most number of overly borrowed books (zero ratios).
Annie Duke and Ryan Holiday have a wider range of ratios across their books, suggesting that some of their books are not as in demand as their other books. Interesting, this could be because both authors had broadly written two types of genres, and of which one of their genres is more popular than the other. For Annie Duke, it was “Decision-Making” over “Gambling”, and for Ryan Holiday, it was “Stoicism” over “Growth Marketing and other topics”.
Nassim Taleb’s books have a wider spread of borrowing ratios, suggesting that not all his books are in demand. Nate Silver only had one book across the three different formats, so I will not over analyse his results (I hoped he had written more).
The top and least 10s
Lastly, looking specifically at the top 10 popular and unpopular books by their ratios, we can see that Ryan Holiday’s non-stocisim books are really not doing well, as with Nassim Taleb’s books, and Annie Duke’s audio title of “How to Decide”.
On the other hand, as I mentioned above, both Adam Grant and Malcolm Gladwell are doing extremely well. “Think Again” (200 audio books), “Give and Take” (47 eBooks), “Hidden Potential” (54 audio books and 161 eBooks) by Adam Grant are all completely borrowed! On the other hand, maybe our libraries should consider increasing the titles of “Blink” and “What The Dog Saw”, as both are also fully borrowed, but yet are not very well stocked by our libraries.
Lastly, I want to to look at the number of titles that had totally no borrows and titles that were totally borrowed out. In total, from the 26 Dec 2023 extraction, there are 4 titles that have no borrowings (as seen in Fig 5), while there are 18 titles that were totally borrowed out.
There were some “reservations”…
I could also have added books reservation data to my analysis as well to understand the actual demand signals of these titles. However, I understand that such data is only available for digital copies, so it may not be a fair comparison.
Reservations for digital titles are also free, while reserving a physical title costs $1.57 SGD, so the idea of reserving a digital and physical title would is quite different. On the other hand, one may argue that this difference is a inherent feature of borrowing digital copies, and should thus, still allow a fair comparison. Ultimately, I decided not to do so, but I wanted to share this thought exercise that I had regarding this issue.
Another point of consideration was how I identified demand for these books is that these proxies of demand are recorded at the point of borrowing. One can argue that it is so easy and costless to borrow a digital book compared to physical books that physical book borrowers have a stronger intent to read or complete the books they borrow (Disclaimer: I highly prefer physical books). Anecdotally, I have borrowed digital copies from our libraries, only to have not even read them after the eBooks were automatically returned to the library.
PS: Right now, I am trying to read an eBook called “Good Strategy, Bad strategy” on my laptop, as I am also taking notes while reading. Hopefully I can complete my first eBook, like ever.
Conclusion
While I mentioned that this analysis doesn’t show any definitive analysis on the book provision efficiencies of our Singapore libraries, at least among the selected author titles extracted, there are more titles that are borrowed out than books that are under-borrowed. It is also interesting to see the huge demand for some of these authors, with hundreds of their digital books being borrowed out. I cannot imagine the costs and logistics involved in lending hundreds of physical copies of any book title, and in a way, the large provision of such digital titles do hint towards some signs of logistics efficiencies made by our Singapore libraries.
If I want to sway the discussion back in another direction and question the validity of these digital copies being read, that will require a deep dive into the literature of digital and physical copy read or completion rates, as well as some studies regarding retention rates across consuming digital (audio and electronic) and physical titles. Overall, it was quite an interesting data exploration exercise for me, and I hope whoever that read it found it interesting too.
A better way to find (physical) SG library books
If you finished my post about library books, I assume you are a hardcore library user. And if you borrow a lot of physical books from our Singapore libraries (like me, for now at least), the web app I built may make it easier for you to know the availability of the physical library books, by their library locations.
I built this tool for myself as I found the experience finding books in our libraries frustrating, and I decided to release it to the public for anyone who wants a better experience finding physical books in our libraries too. I am also adding improvements to my web app, so if you visited my app in the past and not done so recently, do give us a try again! Feel free to provide any feedback to me as well!
To those who have not, feel free to connect with me on Linkedin, where I share more about my tech and data related stuff!
