The conventional practice, prior to the advent of word processor, was to use double space after a period for sentence separation. However, today’s leading word processors seem to contradict this position by having regulated the sentence separation to single space. The advent of proportionate type fonts, as compared to the mono-space typewriter characters, appeared to offer better readability. The use of proportionate type fonts has been carried over to all digital publishing such as Web pages, on-line documents and standardized tests, such as the Graduate Record Examination (GRE). Empirical studies are needed to determine the effect of the new text formatting, as compared with the conventional practice, on reading achievement.
The purpose of this study is to measure the effect of sentence separation using single or double space on time in an on-screen reading task.
…Methodology: College students (n = 66) from a variety of classes, namely Instructional Technology, Business Information Systems, and Technical Communication, from 3 large Southern universities in the US participated in the study. Permission was duly obtained from the Institutional Review Board for Human Subjects for all 3 universities.
…The test passages and comprehension questions used in this study were obtained, with permission from ETS, from the 1997 GRE Practice General Test. Each passage contained an approximation of 164 words. They were first formatted using MS Word using single space and double space for sentence separation, and then a screen-capture of the passage was obtained to preserve cross-platform display consistency between Mac and PCs.
…Results: The Cronbach’s reliability alpha (α) for the sample was found to be 0.6. Figure 4 presented the means of the reading speed for individual passages, as well as the grand means of the two groups. The grand means of double spaced passages was reported to be 65.61 seconds, whereas for the single spaced passages, it was 59.80 seconds.
…Figure 5 presented a clear picture of the reading speed by the two groups, measured in seconds. It can be seen that the ‘double space group’ consistently took longer time to complete the on-screen reading task than the ‘single space’ group.
Paired sample t-test between the two groups (single space versus double space) was carried out using SPSS and the two-tailed statistical-significance level was found to be 0.226 (ie. p > 0.05 at the 95% confidence level).