The Effect of Content, Style, and Color of Picture Prompts on Narrative Writing: An Analysis of Fifth and Eighth Grade Students' Writing
Current assessment practices for writing are moving away from the traditional objective test and towards performance-based assessment. The use of picture prompts to elicit writing samples is a common practice but it adds a level of complexity to the writing process. Writing tasks which use pictures to elicit writing samples require the writer to interpret the picture, create meaning from the picture, and then transfer the visual information into a verbal mode of expression.
The purpose of this study is to examine the characteristics of style and color of picture prompts while holding content constant. Four independent variables were investigated: style, (photograph, drawing), color (color, black and white), content (delivery man with a box, cliff rescue), and grade-level (fifth, eighth). Ratings of the students' stories served as the dependent variable. Each story was scored by two raters on three dimensions: narrative, descriptive, and events. These scores were added together to obtain a total score. The overall design was a four factor repeated measures ANOVA with grade level, style, and color as between subject factors and content as a within subject factor. A total of six ANOVAs were conducted, one each for the total score, narrative component, descriptive component, events component, prior events item, and after events item.
Results of the ANOVA for total scores indicate that the main effect for content was significant, as was the content by style interaction. The main effects for color and style were not significant, nor did these factors yield significant interactions. Similar results were obtained in the analysis of the narrative and descriptive components. The main effect of content was not significant on the event component due to reversal of mean scores on the two items comprising this component. When ANOVAs were conducted on the two items comprising this component, the main effect for content was significant for both items. Also of interest is that the main effect for grade level was significant for the total scores, descriptive component, events component, and prior events, but was not significant for the narrative component or after events item.