Testing the effect of sentence headlines in teaching slides

For the teaching slides in engineering and science classes, instructors often follow the defaults of Microsoft PowerPoint and choose a single word or short phrase as the headline. This paper challenges that practice with experimental evidence showing that a different design using a succinct sentence headline to identify the main assertion of a slide leads to statistically significant increases in the transfer and retention of that assertion. The experimental tests occurred in a large, lecture-based geoscience course that typically had 200 students per section. For the study, the same instructor, during five class periods, used about 100 slides with mostly phrase headlines to communicate the information to two sections of students and then used the same number of slides with succinct sentence headlines to communicate the same information to two additional sections. In the slide transformations, other changes occurred such as typographical changes and conversions of bullet lists to more visual evidence. However, for the fifteen slide transformations considered in this study, the principal change was the conversion of a traditional headline to a succinct sentence headline. For example, in one transformation, the phrase headline Placer Deposits in the original slide was changed to Placer deposits arise from erosion of lode deposits in the transformed slide. When asked to recall the main assertions of slides, the students in the sections taught with the sentence-headline slides had significantly higher recall. For the fifteen questions in the study, the average score for the students viewing the sentence-headline slides was 79% correct, while the average for the students viewing the traditional slides was only 69% correct. A chi-square analysis shows that this difference is statistically significant at the 0.001 significance level. On seven of the fifteen questions, the students in the section with the sentenceheadline slides achieved statistically significant higher scores (three at the 0.001 significance level, three at the 0.005 significance level, and one at the 0.025 significance level), while on only two questions did these same students achieve lower test scores that were statistically significant (both at the 0.01 significance level). In this classroom situation, all four sections of students not only viewed the slides during class, but also had access to the slides as notes after the presentation. The results of these tests have implications in the way that educators design not only their teaching slides, but also their research slides.