The article, “Information Recall of Four Elements Among Young Newspaper Readers,” written by Wayne Wanta and Jay Remy (Newspaper Research Journal 16 [2], 1992), examines the ability of high school students to process and recall information contained in four different print-newspaper elements. The elements were: in-story texts, graphics, index boxes, and pullout quotes.
I believe the purpose of this study was to identify how to relay information in a more efficient manner as newspapers began to identify a young demographic as their future target audience. To me, this seemed like a very practical study because it was conducted at a time when newspapers started to change their looks in order to attract new readers. Some of the biggest changes that newspapers were making were the inclusion of more advanced graphics. The hope was that this design revolution would especially appeal to young readers through the increased usage of visual elements. If young readers were processing information from certain design elements and are recalling this information in a more efficient manner, then they may develop a reading habit as they mature.
According to Wanta and Remy (1992), “When young readers process information more efficiently, their satisfaction with the reading experience is higher. Increasing reader satisfaction translates into stronger circulation figures because satisfied readers are more likely to return to newspapers for additional information in the future” (p. 113). The primary research questions that Wanta and Remy utilized were: “Do graphics enhance readers’ ability to recall information contained within them by combining textual information with visuals?” and “How does the processing of graphics information compare to the recall of information contained in other elements on the newspaper page?”
Wanta and Remy surveyed a total of 204 high school students from nine Illinois high schools. The students were, “mostly 18-year-old seniors” (p. 116), and came from a variety of backgrounds, an advantage which was enhanced by the use of a wide range of high schools.
Each participant read the front page from one of 20 selected newspapers. The newspapers were selected based on two criteria: First, because they received a large number of awards for page designs as determined by the Society of Newspaper Design. Second, some newspapers were selected because they were judged as top newspapers in the country. The January 10, 1991, edition of each newspaper was used in the study. The front page of each newspaper had to do with the Gulf War, and each newspaper used different elements to convey their respective messages. “The study was conducted in April and May of 1991, or two to three months after the Gulf War and a few months after the newspapers were published” (Wanta, W. & Remy, J., 1992, p. 118).
Each subject within each participating class was randomly assigned to one of the 20 newspaper treatments. Subjects were instructed to read the front page of the newspapers for 20 minutes, then answer a quiz dealing with information contained in the stories and other elements on the page. Since each newspaper ran different elements on its pages, only a few newspapers used story texts, graphics, pullout quotes and inside boxes. However, every newspaper in the study used at least two of the elements. Therefore, to examine the subjects’ ability to recall information in each of the four design elements, a series of T-Tests examined differences in knowledge recall scores across two elements at a time.
One potential issue with the study was how educated each of the participants was on the Gulf War. If any of the participants were avid readers of stories dealing with the Gulf War, then they may have been able to recall information about the war without thinking about the newspapers they had just read. However, Wanta and Remy noted that, “as an added precaution, a question dealing with how often subjects typically read a newspaper was used as a control. Since none of the results changed substantially, the control was not used in the final analysis” (1992, pg. 119).
Critique
Control is what distinguishes scientific research from a common idea. It rules out all other possible explanations to the results of the study. While I enjoyed reading the results of the research, I did not find it to be rigorous in all aspects of the study.
Wanta and Remy focused on their study through two research questions: “Do graphics enhance readers’ ability to recall information contained within them by combining textual information with visuals?” and “How does the processing of graphics information compare to the recall of information contained in other elements on the newspaper page?” I thought their data produced some interesting findings, however I also feel like these two questions could be their own research study.
Wanta and Remy did seem to be very well versed in the previous research that had been conducted up to that point. Their literature review was very thorough and answered many of the questions I had in my mind about the pre-existing data they were referring to earlier in the article.
I did not feel that the method of which they utilized to select participants for the study was completely sound. Wanta and Remy claim that the students that made up their sample came from a variety of backgrounds. They said the participants came from a wide-range of high schools, with the respective enrollments ranging from 1,276 to 85 total students. I was not impressed by these numbers. I graduated from a moderately sized high school of a little under 2,500 students. My high school graduating class was about 600 students. Near my school, there were both an alternative high school and a private school, each with an enrollment of under 200 students. While I obviously do not have any hard data with me, I am sure that each of these schools would produce significantly different results if this study were recreated. The article also fails to mention what the final demographics were of the study. Upon completion of the article, I do not know anything about the participating sample, other than they were mostly 18-year-old high school seniors. Wanta and Remy only identified two schools they used, not giving us much of a clue if they were small schools in rural Illinois, or inner-city schools.
The use of language was a strong point of this article. I felt that Wanta and Remy utilized strong language that was easy to understand. Key terms were clearly defined and they used little to no jargon. I think someone with just a casual interest in media-based research could read this article and understand each of Wanta and Remy’s findings and conclusions.
Precision
I was a little disappointed by the seemingly little effort put into controlling the error term or the denominator of the statistic. I do not feel that Wanta and Remy were very precise with their experiments. The term “precision” mean those who are performing the experiment cut down on the sources of variation among factors other than experimental manipulations. We also know that any study that controls only by randomization is likely to have low precision, unless it boasts an unusually large number of research subjects.
One short section in the article reads, “Randomization also should have lessened any potential effect that different reading habits of subjects may have had upon the dependent variables. However, as an added precaution, a question dealing with how often subjects typically read a newspaper was used as a control. Since none of the results changed substantially, the control was not used in the final analysis” (1992, 119). I was shocked that the greatest effort put into controlling the experiment was not even used. It seems like Wanta and Remy were relying heavily on randomization, and since they did not have an abnormally large amount of participants, I am led to believe that this study has low precision. If I were recreating this experiment, I would certainly employ greater measures of control to ensure my data was both precise and valid.
Generalizability
Generalizability refers to whether the results of a research study are generalizable to the population of interest. How the participants in a study were selected has a lot to do with how generalizable the results are. Results that lack generalizability are misleading.
As I previously mentioned, I do not feel the class in question is clearly defined. While I think it was wise to target a future target audience to identify their reading habits, I am not sure the data this experiment produced is valid outside of the Illinois communities in which the study took place. This was one aspect of the study that I did not find very generalizable.
One thing I found very pleasing that Wanta and Remy did in this study was the wide-array of newspapers used. Publications from Atlanta all the way to Seattle were used, which helps with the generalizability of the study. Wanta and Remy also used the publications from the same date for each of the newspapers.
Overall, I felt this was a good study that produced some valuable data and information for newspaper editors across the country. Without being too critical, I do feel like it would be wise to recreate this experiment in order to increase its level of precision, which I felt was one of the study’s biggest weaknesses. If I were an editor of a local newspaper, I would follow the blueprint that Wanta and Remy created with this study and specifically test my target audience(s) in order to determine the reading habits of my readers.
Hayden Coombs 