Readability of Body Text in Computer Mediated Communication

Effects of Type Family, Size and Face

Abstract

This experimental design used 78 subjects to test readability of different type sizes and type faces in computer mediated or on-line environments.

The study found few differences for speed of reading between type sizes of fourteen, twelve and ten point type. Twelve point type had the highest rankings for both speed of reading and recall of material. Little difference was found between serif and sans serif faces except in ten point type where spped and recall were poorer with the serif type faces.

Joel Geske, Associate Professor

Greenlee School of Journalism and Communication

114 Hamilton Hall, Iowa State University

Ames IA 50014

e-mail: geske@iastate.edu

Phone: 515-294-4342

Readability of Body Text in

Computer Mediated Communication

Effects of Type Family, Size and Face

"I can make an ugly document very easily...I own a computer and layout software" (White, 1992). Unfortunately, while the quote may make one smile, low cost computer technology has allowed many individuals to become publishers over interactive computer networks such as the World Wide Web; many publishers with little or no comprehension of typography.

Numerous studies have been done determining both legibility and readability of type in printed materials in mass communication settings (Tinker, 1963, McVey & Weigeshaus, 1973 and Felici.) However, a great many of the classic studies in readability and legibility are from the 1920-1940 era. Starting in the 1980's with the advent of the personal computer and in particular in the 1990's with the increased usage of the Internet for transferring large amounts of information via the computer, it is necessary to see if many of the basic rules of legibility and readability of type hold true for computer monitor presentation.

Review of Literature

Good design for mass media is communicative design. It presents information in a clear, easy to use way. Type and good typography is an essential part of that design. Especially with interactive media, where the reader may jump from site to site, type can be used as a consistent design factor to give identity to a page.

Good design enhances the transfer of information, both in the style of the type and through the ability to read the type. The reader shouldn't have to strain or be aware of the act of reading. Type readability, particularly for longer pieces of text, is dependent on numerous factors. In general, type sizes over fourteen point size are reserved for headings, subheadings and other display purposes. Type for extended reading should generally not be smaller than nine point and not larger than fourteen point whether for print or computer monitor (Spiekermann, 1993; Lynch, 1994). Recent research indicates that nine and even ten point type has poor legibility on screen and should be used with caution (Geske, 1996).

Typography and design on the computer screen has always been difficult due to the low resolution of the screen compared to a printed page. When graphics based fonts first started appearing in graphical interfaces such as Macintosh and later Windows applications, they were almost always stored and displayed in bitmap format. A typical computer monitor has 72 pixel resolution (or 72 pixels by 72 pixels in each square inch of screen.) The minimum stroke width is one pixel wide or high. The next possible stroke width is double that or two pixels wide. Especially in smaller size fonts (such as those used for body copy) this allows for very little fine variation in displaying fonts.

More recently, outline font technology has come to the personal computer. Here the letter is outlined using a mathematical formula to create straight lines and curves. Once the outline is complete it is filled in solid. One problem, however, is that this method is much slower to produce a page full of text than bitmapped fonts. To solve this problem, once the font has been formed, it is converted to a bitmap form and saved or "cached" in the computer memory. This conversion is known as rasterization. Conversion takes some time, but once completed the display fonts can be as fast as the display of bitmapped fonts.

Unfortunately, in the rasterization process some rounding off of numbers occurs. Some letters, such as an H. which should have two vertical strokes of about the same width may end up with one a pixel larger. On large display type or when using a 300 dot per inch printer it may be hardly noticeable. On screen, however, with only a 72 dot resolution and when using small type faces the problem can be dramatic and severely affect readability (Petzgold.) Sans serif type faces, with less difference in stroke widths may have less problems than serif faces with pronounced differences in stroke widths.

To compound the problem, when the computer creates bold face type it generally doubles the vertical stroke width of the letter but leaves the horizontal stroke of the letter at the original height. This helps solve the problem of the centers of letters like "a" filling in by not adding additional pixels on the horizontal stroke, but alters the character and design of the type on a fundamental level. Doubling the vertical pixels also has the effect of extending the serifs and causing them to run together. Even sans serif types can have a problem with letters crowding or running together in a bold face.

Sans serif type faces, with less difference in stroke widths may have less problems. One type designer (Chuck Bigelow, designer of the Lucinda Family or type) recommends sans serif because "when printed, the serifs on typefaces are only a tiny percentage of the typeface design. But on-screen, in order to display the serifs using the limited number of available pixels, they take up a much bigger proportion of the information than they do in print. Serifs should be small things-but on screen they become big...noise or distracting chunks of interference (Will-Harris, 1996).

"Italic" type also has inherent problems on screen. In fact, rather than true italic, most monitors actually display an oblique or slanted form of the standard face. Again, especially in smaller font sizes that are using letter strokes only one pixel wide, this slanting letter can take on a very jagged and difficult to read appearance.

Finally, computer screens are luminous while most previous readability work was done on reflective or print material. with a luminous screen, the light tends to spread out and cause the letter strokes to lose a bit of weight. Because of these reasons, type faces must always be judged by their appearance on-screen, not by their esthetic appeal when printed.

It can be concluded that computer screen monitors present unique challenges to the typographer and designer. Many of the assumptions made for print legibility and readability may not hold true for on-screen type. The very nature of type construction and display on the screen bring into question legibility factors that may affect the readability and the comprehension of material presented on computer screens.

Research Question

As more traditional journalistic media look to computer mediated delivery systems, the question of how to provide large amounts of information in the most legible, readable form becomes an important factor. Today, with the increasing popularity of the Internet, books, magazines, newspapers, advertising and even videos appear on the computer. One fundamental question is to look at type readability on screen to determine type styles and type sizes that maximize readability and comprehension of the material being presented.

This study looks at type size and the use of bold type as basic factors affecting type readability. Plus, since the review of literature indicates there may be particular problems associated with serif type faces on screen, a comparison of serif and sans serif type in legibility and comprehension will be explored.

Hypotheses

Based on both previous research in print materials and common sense "rules of thumb" provided by typographers, the following hypotheses are proposed. The first three hypotheses deal with speed of reading material, while the second set of three deal with comprehension of material. The last two deal with the use of bold versus normal type.

Hypothesis One: Fourteen point type will result in faster reading times than twelve point type.

Hypothesis Two: Twelve point type will result in faster reading times than ten point type.

Hypothesis Three: Fourteen point type will result in faster reading times than ten point type.

With ease of reading, subjects should be able to comprehend the text I and process information more easily. Therefore:

Hypothesis Four: Recall of reading selection will be higher for text set in fourteen point type than text set in twelve point type.

Hypothesis Five: Recall of reading selection will be higher for text set in twelve point type than text set in ten point type.

Hypothesis Six: Recall of reading selection will be higher for text set in fourteen point type than text set in ten point type.

Previous research indicates that bold type is more legible on screen, therefore:

Hypothesis Seven: Bold face type will result in faster reading times than normal face type.

Hypothesis Eight: Recall of reading selection will be higher for text set in bold type than text set in normal type.

Methodology

For this experiment, paragraphs of copy were selected that were of a non-involving nature and all selected from the same book and author. Each paragraph had approximately 225 words and was measured and corrected to have a similar Flesch reading difficulty score for grade 7.5 (plus or minus 0.2) to provide similar reading material.

Readings were randomly assigned to test variables. Selections were randomly assigned reading order.

Reading material was incorporated into a Web site so subjects would read the material on screen. Screen windows were adjusted so line lengths were approximately 2 1/2 alphabets in length for the median type size. Black type on the default gray background was used for all samples.

Subjects controlled the start of testing by using the mouse to click on a button. When finished reading, subjects would click a button to take them to a blank screen. Subjects were given several examples to adjust to the procedure and methodology before the test began.

Subjects were measured (to the nearest second) for the amount of time it took to read each selection. Then students were asked to answer five simple multiple choice questions about the reading. Each question had five possible answers, including an option that the reading did not include this information. The test was to encourage subjects to read the material carefully for content and not to skim the reading material.

The multiple choice test should be viewed as a measure of short term memory of the reading.

Subjects were assigned to Form A or Form B as they entered the testing lab with Form A using a common serif type face, Palatino, and Form B using a common sans serif type face, Helvetica.

Subjects chosen were students at a major midwestern university in the United States. College students were chosen for several reasons. First, they are one of the largest and fastest growing areas of computer mediated communication is the Internet. One of the most comprehensive surveys on Internet use still shows students making up over half of the total users (even with the increased number of commercial users). College students in general are literate, familiar with computers and are at an age where there are minimal uncorrected vision problems. This should be viewed as an optimal audience, however, with differences expected with less literate subjects or at different age groups.

A total of 78 subjects were recruited for the experiment. Over ten majors were represented in areas of business, communication, design, human performance and consumer sciences among others. Mean age was 21.2 . Gender for the respondents was 56% female and 44% male.

Findings

Speed of Reading

Hypothesis One: Fourteen point type will result in faster reading tines than twelve point type.

Type Face	14 pt Time	12 pt Time	t-test of means (p=0.05)
Serif	77.4 sec.	74.0 sec.	not significant
Sans Serif	81.6 sec.	74.8 sec.	not significant

Mean time for reading a selection in fourteen point serif type was 77.4 seconds while mean time for twelve point serif type was actually faster at 74.0 seconds. However, a t-test of means shows no significant difference in reading speed.

For sans serif type, mean time for reading a selection in fourteen point type was 81.6 seconds while mean time for twelve point sans serif type was actually faster at 74.8. A t-test of means shows no significant difference in reading speed.

Based on the evidence, Hypothesis One is rejected; fourteen point type is not faster to read on screen than twelve point type.

Hypothesis Two: Twelve point type will result in faster reading times than ten point type.

Type Face	12 pt Time	10 pt Time	t-test of means (p=0.05)
Serif	74.0 sec.	83.9 sec.	significant
Sans Serif	74.8 sec.	81.9 sec.	not significant

Mean time for reading a selection in twelve point serif type was 74.0 seconds while mean time for ten point serif type was 83.9 seconds. t-test of means shows this to be a significant difference in reading speed (p=0.026).

For sans serif type, mean time for reading a selection in twelve point type was 74.8 seconds while mean time for ten point sans serif type was slower at 81.9 seconds. A t-test of means shows no significant difference in reading speed.

Based on the evidence, Hypothesis Two is accepted for serif type and rejected for sans serif type.

Hypothesis Three: Fourteen point type will result in faster reading times than ten point type.

Type Face	14 pt Time	10 pt Time	t-test of means (p=0.05)
Serif	77.4 sec.	83.9 sec.	not significant
Sans Serif	81.6 sec.	81.9 sec.	not significant

Mean time for reading a selection in fourteen point serif type was 77.4 seconds while mean time for ten point serif type was 83.9 seconds. A t-test of means shows no significant difference in reading speed.

For sans serif type, mean time for reading a selection in fourteen point type was 81.6 seconds while mean time for ten point sans serif type was just slightly slower at 81.9 seconds. A t-test of means shows no significant difference in reading speed.

Based on the evidence, Hypothesis Three is rejected; fourteen point type is not faster to read on screen than ten point type.

Serif vs. Sans Serif

Comparing the serif face and sans serif face in equal sizes shows no significant difference in reading time between the 41 subjects reading serif versus the 37 subjects reading sans serif type. >

Type Face	14 pt Time	12 pt Time	10 pt Time
Serif	77.4 sec.	74.0 sec.	83.9 sec.
Sans Serif	81.6 sec.	74.8 sec.	81.9 sec.
t-test of means (p=0.05)	not sig.	not sig.	not sig.

In all three point sizes, there appears to be no significant difference in reading times for serif vs. sans serif type.

Recall & Comprehension

Speed of reading is only one consideration. It is possible that if subjects find the text hard to read, they will skim through the text or give up on reading it. Therefore, it is important to test for recall of knowledge about the reading as well.

Hypothesis Four: Recall of reading selection will be higher for text set in fourteen point type than text set in twelve point type.

Type Face	14 pt Recall (average on 5 item test)	12 pt Recall (average on 5 item test)	t-test of means (p=0.05)
Serif	4.42	4.85	sig. (p=0.0031)
Sans Serif	4.43	4.62	not significant

Based on the evidence, Hypothesis Four is rejected for serif type and for sans serif type. In fact, there is a significant difference in the opposite direction for serif type with twelve point type having higher recall.

Hypothesis Five: Recall of reading selection will be higher for text set in twelve point type than text set in ten point type.

Type Face	12 pt Recall (average on 5 item test)	10 pt Recall (average on 5 item test)	t-test of means (p=0.05)
Serif	4.85	3.98	sig. (p=0.0000)
Sans Serif	4.62	4.22	sig. (p=0.015)

Based on the evidence, Hypothesis Five is accepted; material set in twelve point type has a higher recall rate than material set in ten point type.

Hypothesis Six: Recall of reading selection will be higher for text set in fourteen point type than text set in ten point type.

Type Face	14 pt Recall (average on 5 item test)	10 pt Recall (average on 5 item test)	t-test of means (p=0.05)
Serif	4.42	3.98	sig. (p=0.0.33)
Sans Serif	4.43	4.22	not significant

Hypothesis Seven: Bold face type will result in faster reading times normal face type.

Type Face	12 Point Bold	12 Point Normal	t-test of means (p=0.05)
Serif	71.7 sec.	74.0 sec.	not significant
Sans Serif	75.1	74.8	not significant

Based on the evidence, Hypothesis Seven is rejected. Bold face type is not faster to read on screen than normal type.

Hypothesis Eight: Recall of reading selection will be higher for text set in bold type than text set in normal type.

Type Face	12 Point Bold	12 Point Normal	t-test of means (p=0.05)
Serif	4.20	4.85	significant (p=0.0001)
Sans Serif	4.27	4.62	not significant

Based on the evidence, Hypothesis Eight is rejected for both serif type and sans serif type. For serif type, the normal face results in a statistically better recall and for sans serif type the normal face is also better, although not significant at the 0.05 level.

Qualitative Measures

In addition to the quantitative measures for speed and recall, subjects were asked at the end of the study to rank different type sizes and styles on a five point Likert scale with 5 being very easy to read and 1 being very difficult to read.

In all cases, the mean ranking of normal type faces was higher than bold and subjects showed a clear preference for the larger size type faces.

Earlier work indicated that italic type faces were much more difficult to read and italic type faces were not tested in the timed readability study. Twelve point type was selected to compare for subject preference between normal, bold and italic faces. Subjects preferred the normal face with a mean score of 4.268. Bold scored a mean of 4.000 with italic at 2.07.

Conclusions

This study indicates that body type size, ranging from ten to fourteen point does not make a great deal of difference in speed of reading. Type in the 12 point range appears to be the best choice for reading on computer monitor screens. Larger size, fourteen point type, while preferred subjectively by users was actually read more slowly than twelve point type on the computer screen. In either case, these type faces are somewhat larger than would normally be used for body copy in normal newspaper, magazine or other print materials.

Twelve point type was significantly faster to read than ten point in serif type faces but showed no significant difference in sans serif. Based on the literature, this would seem likely as at smaller sizes the serifs can begin to cause legibility problems. It appears the boundary line may be that below twelve point type, serifs should be avoided as they hinder the reading process.

However, size of type did make a significant difference in the comprehension of material as measured by short term recall. Subjects reading twelve point type scored higher on recall tests than subjects reading fourteen or ten point type. Twelve point scores were significantly higher than ten point scores for both serif and sans serif faces. For serif faces, twelve point was significantly better than fourteen point and fourteen point was significantly better than ten point. Sans serif type showed the same trends, but not at significant differences. Again, the evidence would suggest that twelve point type is the best choice for body text for recall of material as well as for speed of reading.

This evidence is contrary to what would be expected through the literature or "common sense" rules of typography. One would expect that with the low legibility level of type on screen, that the larger size would improve readability and comprehension. In fact, the attitude rankings on a five point scale showed readers preferred fourteen point type over twelve by a full point on the ranking scale and preferred twelve point type over ten point type by nearly two points on the scale. However, that visual preference is at the expense of both speed and comprehension.

Similar results were obtained for bold faces, with fourteen point preferred, followed by twelve point followed by ten point. While earlier studies indicated that bold type helped increase legibility of the type, subjects preferred the normal faces.

As expected from earlier studies, subjects found italic type much more difficult to read scoring it nearly two points lower than the other choices.

Serif and sans serif type faces scored very similar in reading speed in all cases and show no significant advantage in readability using one or the other. Ten point serif type did show significantly poorer scores in some instances and should be used with caution if at all.

In general, this research shows that some of the common sense typographic "rules" and traditions are not supported by the evidence. Larger type, while preferred by the reader, is not better for speed of reading or comprehension of material.

Bold type, while shown in improve legibility in earlier research, is generally not preferred by subjects, does not increase readability and may decrease recall.

Overall findings indicate that optimal type for computer mediated communications should stay in the twelve point type range in a normal type face. The choice of serif or sans serif appears to make little difference in twelve and fourteen point sizes. However, at ten point size, the serif face tested begins to show significant problems in both speed of reading and recall.

Sources

Felici, J. (1987). Desktop Publishing Skills: A Primer for Typesetting with Computers and Laser Printers. Reading MA: Addison-Wesley.

Felici, J. The Desktop Style Guide. Bantam, ITC.

Felici, J. (1996, March). A Bitmap a Day Keeps the Eyedoctor Away [Internet Document]. Internet locator: http:/ /www.will-haris.com/ typofeli.htm.

Geske, J. (1996, August). "Legibility of Sans Serif Type for Use as Body Copy in Computer Mediated Communication". Association for Education in Journalism and Mass Communication (AEJMC) annual conference, Anaheim CA.

Lynch, P. (1994). Visual Design for the User Interface. Journal of Biocommunications, Vol. 21, no. 2. p. 6-15.

Matrix News. (1995). Second TIC/MIDS Internet Demographic Survey, [Internet Document]. Internet locator: http:www.mids.org.1995.

McVey, G.F., & Weigeshaus, C.J. (1973) Studies in Legibility. Rochester, NY: The Center for Visual Literacy, University of Rochester.

Petzold, C. (1992, November 24). Using text and fonts under the OS/2 2.0 Presentation Manager. PC Magazine, Vol. 11, no. 20. p. 453ff.

Spiekermann, E and Ginger, E. (1993). Stop Stealing Sheep and Find Out How Type Works. Mountain Mlew CA: Adobe Press.

Tinke K. (1996, March/April). Taking It In. Adobe Magazine, pp. 40-45

Tinker, M.A. (1963). Legibility of Print. Ames IA: Iowa State University Press

Will-Harris, D. (1996, March). The Best Faces For the Screen. [Internet Document] . Internet locator: http: / /www.will-haris.com/ typoscrn.htm.

White, A. (1992). Type in Use. New York NY: Design Press.