Blog
Font choice; serifs, sans- serifs and legibility
By Emma Smith
July 12, 2021.
How should you choose a font? You may put a lot of thought into the message a font sends about your brand. Is it friendly and familiar or formal and business-like? You may consider whether it’s aesthetically pleasing. Underpinning all these decisions should be an evaluation of the font’s legibility in all the situations you plan to use it.
Because of the range of users of your product and the range of situations in which you’ll use the font, legibility can be difficult to assess. Let’s have a look at one of the factors that determine legibility- the presence of serifs.
The presence of serifs is probably something you already consider. Serifs are the small ‘ticks’ or extra strokes present at the end of the main strokes of a letter. I’m sure you’re aware of the general understanding that sans-serif fonts (or fonts without serifs) are best used for headings and signage, but serif fonts are best used for large bodies of text. The idea behind this is that serif fonts aid reading speed by providing a little bit more differentiation between similar looking but different letters and a better kind of ‘visual flow’ between letters within a word. This idea is such common knowledge that it’s difficult to pinpoint where it began. A quick look at the history of typography (see Lowry, and Abrams, 1989) shows that the very first typefaces were inspired by calligraphic writing and set the convention for the use serif fonts when printing large amounts of text. Sans-serif fonts were not invented until centuries later and they had a very industrial connotation (Melton, 2020). What this means is that those who read a European language have had centuries of experience reading large blocks of text in a serif font. Could it just be practice that makes it seem as though serif fonts are best for large blocks of text and sans serif is best for logos, headings, and signage?
A recent study by Arditi and Cho (2005) has attempted to tease this question apart. As they point out, prior to the creation of digital methods that help us easily create iterations of the one font, it has been very difficult to conduct the kind of study that allows us to investigate what it is specifically about the appearance that makes it easier to read. This is because the researchers could not finely control the appearance of the fonts they studied. Arditi and Cho (2005) have done just that. They created a single font base and manipulated the presence of serifs and the size of the serifs. Crucially, they also manipulated the spacing between the letters. This is something often confounded with the presence of serifs because serifs require there to be more space between letters.
Before looking at the results of their study, it’s a good idea to look at how they measured legibility. They’ve used three very common methods used in legibility studies. The first looks at the smallest size a string of letters (5 random letters in this case) need to be for their participants to accurately identify at least 4 the letters in 69% of the times they were shown the stimuli. The next test measured how fast participants could accurately read sentences shown on screen in a large font with conventional capitalisation. The final test was also a reading speed test, but participants read longer blocks of text with scrambled sentence order. When it comes to scientifically assessing legibility so you can infer a real-world situation, clearly, it’s a difficult task.
What did they find? In the two tests that measured sentence reading, it made no difference whether the font had serifs or what size serifs were. These results were the same for all their participants no matter what their level of visual acuity. These tests suggest that if you use a large enough font size the presence of serifs is immaterial to reading. Where it does seem to matter is in the fist test, where the font size is challenging your ability to make it out (or is close to your threshold for resolving it). Here, what mattered most no matter what the participants’ level of visual acuity, was really the spacing between the letters. A gap that is 10% of the height of the capital letters was significantly more legible than a smaller spacing, no matter the serif configuration.
This study suggests some very good news when choosing a font for your project. What it tells us is, if you use a good font size (and allow responsive font sizing for vision impairment) and a font that isn’t too cramped together, you can pay more attention to the feel and messaging of your font of choice.
References
Arditi, A., & Cho, J. (2005). Serifs and font legibility. Vision Research, 45(23), 2926-2933. https://doi.org/10.1016/j.visres.2005.06.013
Lowry, M., & Abrams, G. (1989). Venetian printing: Nicolas Jenson and the rise of the Roman letterform. Kristensen.
Melton, J. (2020) The Serif-Less Letters of John Soane. In: Pen, print and communication in the eighteenth century. (pp. 215-228) Liverpool University Press. ISBN 978-1-789-62230-0