Penn State Study Examines Font Legibility

The Larson Transportation Institute at Penn State University conducted a study on font legibility through a grant from Gemini Inc. (Cannon Falls, MN), a manufacturer of dimensional letters. The following is the Executive Summary from the report. For information about the full report, contact Philip Garvey at

Background and objectives

The enormous font selection available for on-premise signs fosters creativity, but also limitations, because of the unknown of a given font’s legibility at various distances. Although a small number of studies have broached this topic, this research effort is intended as the first to address the visibility of a large set of existing on-premise fonts.

Laboratory experiment


The study was conducted at Penn State. Sixty-four signs were tested, using 34 unique fonts. The fonts were displayed as scale-sized, one-word, on-premise signs on a high-resolution computer monitor. Sixty-four subjects from 19 to 87 years of age participated. The legibility distance of each font was determined, and the effects of age, uppercase vs. lowercase, serif vs. sans serif, word-superiority, and art/word combination were evaluated.

Age group effect

The subjects were divided into three age groups. Not surprisingly, and consistent with earlier research, the young group and the middle age group were both able to read the signs from further away than the more elderly group.

Font effect

Large differences in font legibility were found. Gill Sans uppercase provided the best legibility, while Mistral lowercase had the worst. Also, simply choosing a font with a 5-ft./in. or larger ratio of distance to letter height insures better legibility, both statistically and practically.

Case effect

For all 31 fonts presented in both upper- and lower-case conditions, the upper-case words were more legible. In 22 of those cases, that difference was statistically significant.

Serif vs. sans-serif effect

There was no statistical difference between the serif and the san-serif fonts when shown in uppercase. A statistically significant effect was found in the lower-case analysis; however the difference was not practically significant.

Font family effect

Several fonts tested had more than one “weight,” such as bold or condensed.  The upper- and lower-case fonts were analyzed separately with the following results:

  • For upper and lower case, Times Bold was significantly more legible than Times New Roman.
  • For upper and lower case, Optima Bold was significantly more legible than Optima.
  • For upper and lower case, Garamond Bold was significantly more legible than Adobe Garamond.
  • For upper case, Helvetica was significantly more legible than Helvetica Bold, Helvetica Light and Helvetica Medium Condensed. Also, Helvetica Bold was more legible than Helvetica Light. For lower case, Helvetica was more legible than Helvetica Bold and Helvetica Light.

Word superiority

The 64 words showed large differences in legibility. The most legible word was Sunday, which was more than twice as legible as the least legible word, Crawfordsville.

Words and art

All of the signs tested included words and a graphic element. In 10 instances, the graphic directly related to the word (e.g., a drawing of a flower and the word “Flower”). This relationship minimally impacted sign readability.


This research sought to determine the relative legibility distances of a large set of fonts used on on-premise signs.  It allows users to compare legibility distances and make an informed decision about which font to use on their signs. Several results are clear:

  • Although font selection can significantly impact on-premise-sign legibility, many fonts have equivalent legibility.
  • Case (upper vs. lower) sometimes, but not always, can greatly impact sign legibility. Upper case often performs significantly better than lower case, at least under this study’s conditions.
  • The choice of serif vs. sans serif alone doesn’t measurably affect legibility.
  • Font weight can dramatically impact the distance from which a sign can be read. Fonts from the same family (e.g., Times) can have dissimilar legibility.
  • Word selection can greatly impact sign legibility. Not surprisingly, simpler, shorter words can be read at greater distances, regardless of font.
  • Matching a word to an image or graphic on a sign doesn’t, generally, positively impact legibility.