fltis study investigated readers’ reactions to vertical spacing of text presented on a computer screen. Results showed that text width, font size, and seveml other variables were related to judgments of vertical spacing. INTRODUCTION The development of software to supprt collab orative work activity increases the amount of time people may spend interacting with information presentedas text on a computer screen, SUCce.ssfidadoption of such software places a premium on the appearance of text on the screen. Extensive mearch on factors affecting appearance printed material has been carried out (Tinker, 1%3), but less has been done for text presented on computer screens (e.g. Gould, Alfaro, Finn, Haupt, & Minute, 1987; Holleran, ]992). One variable which may have significant effects on Iegibitity and appearance is the vertical spacing of text. The distance from the baseline of one line of text to ano[her, called the line height, substantially affects both the overall darkness of the page (“color”) as well as the eye’s ability to correctly position itself as text is read. Extra space between lines may help the eye scan lines accurately during reading and kxae the beginning of the next line after the previous is finished. For example, Kruk & Muter ( 1984) found that subjects read text on a computer screen over 1070 more slowly with singlespaced compared with double-spaced text. However, the space between lines is ncx independent of other typographic factors, such as the height of the characters, the “x-heigh{” of the characters, the presence or absence of serifs, and the length of the line. There is little empirical parametric evidence for the relationship of these variables with one another, ahhough Bouma (1980, cited in van Nes, 1986) has suggested that the ratio of the ljne height to the line width should be greater than .033. Grabinger & Amadeo (1988) also found that subjects rated text displayed on computer screens as less easy to study when the lines were single spaced and also when they were longer. The purpose of the present study was to evaluate users’ preferences for various levels of vertical spacing of text presented on a computer screen. Specifically, the relationship of vertical spacing with several other variables, including typeface, character size, and width of tex~ was also to be investigated. STUDY 1 The ob@t of the f~t study was to investigate preferences for vertical spacing of text presented in a normal, comfortable size on a computer screen. Additionally, the relationship of these preferences to two different typefaoes-one with serifs and one without—was also to be investigated. It was hypothesized that subjects would prefer vertical spacing in which the line heightiine width ratio was greater than .033, and that the pattern of preferences would be similar for both typefaces. Ten subjects, all employees of AppIe Computer, Participate in the research. Each subject was seated in front of a computer screen and presented with five windows containing the same paragraph of text presenttxt in a 12 point font (either Geneva or Palatine). The text was left-justified and the width of the single column was set to 6.5 inches. In each window the line height was set to 13, 14, 15, 16, or 17 pixels (representing height/width ratio of .28, .30, .32, .34, and .36 respectively). All text was displayed on an Apple 21” RGB monitor with a rescdution of 79 pixels per inch. In this rejwm all measurements reported in inches are actualJy based on a “ruler” in a word processor which assumes there are 72 pixels per inch, and so are actually only 9]% of the statd value. Subjects were asked to amange the windows in an order determined by the attractiveness of the display and how much they liked it. These rankings were recorded for later analysis. Once each subject finished with the text in the frost typeface a separate presentation of text in the other font was made. Results Separate analyses of variance for the rankings within each font were performed. Separate analyses of variance for both the Pdatino and Geneva faces revealed significant effects for vertical spacing, F(4,36) = 27.00, p <.001, and F(4,36) = 9.52, p < .001, respectively. Pairwise r-tests among mwts within each font Were dso perfO~ed. Mean mnkings for each font are displayed in Table 1; means with common letters are not significantly different from one another within that font at the p e .05 level.
[1]
van Fl Floris Nes,et al.
Space, colour and typography on visual display terminals
,
1986
.
[2]
Douglas Amedeo,et al.
CRT Text layout: perceptions of viewers
,
1988
.
[3]
John D. Gould,et al.
Reading from CRT Displays Can Be as Fast as Reading from Paper
,
1987
.
[4]
H Herman Bouma,et al.
Visual reading processes and the quality of text displays
,
1980
.
[5]
Patrick A. Holleran.
An assessment of font preferences for screen-based text display
,
1993
.
[6]
M. Tinker.
Legibility of print
,
1963
.
[7]
Paul Muter,et al.
Reading of Continuous Text on Video Screens
,
1984
.