ACES 2012 | New Orleans

Online editing study update

An update of a study first presented at the ACES conference in 2011 reinforces the finding that routine editing has an overall positive effect on reader perceptions of a story’s professionalism, organization and writing quality.

Fred Vultee of Wayne State University presented the study update Friday, April 13, at the 16th national conference of the American Copy Editors Society in New Orleans. Vultee’s research was sponsored by ACES.

The study tried to present a realistic view of news as readers see it today, Vultee said.

The process, Vultee said, involved choosing eight brief news articles from online news operations, the broadcast sides of news partnerships and similar sites to represent a “publish first” mentality; those articles were then edited, so that each story had edited and unedited versions. Participants read all eight articles, four edited and four unedited, and answered a set of questions after each article. Those responses produced a set of measures representing professionalism, organization, writing quality and value.

Overall, the study found, editing produces statistically significant and moderately strong effects on perceptions of all those variables. Not every outcome or every story is improved, but editing makes enough articles better enough to produce a measurable effect.

Relevant new findings include:

● Perceptions of value are most strongly associated with improvements in the perceived
professionalism of an article. Editing can make an article seem better written or better organized, but professionalism seems to most clearly make “stories like this” worth paying for.

● Editing is not a matter of preaching to a partisan choir. People who see themselves as
politically similar to the news media see the least difference between unedited and edited stories. People who rank their political views as to the right or the left of the media consistently see unedited stories as worse and edited stories as better than their counterparts in the middle. Mitt Romney’s comment at the meeting of the American Society of News Editors this month – “I find myself missing the presence of editors to exercise quality control” – could find an echo with both his supporters and his opponents.

● Differences in group measurements sometimes hide important parallels in effects. Some
audience subgroups give consistently higher marks to unedited copy. Whatever their starting point, editing often improves their perceptions as much as it does for the whole audience.

The study was conducted from May 2011 to December 2011 at Wayne State University in Detroit. It used 119 participants from a recently established extra-credit pool that mostly draws on a campuswide public speaking course.

Participants in the study reflect the diversity of this urban research university. Ages ranged from 17 to 58, with an average age of about 24 and a median age of 21. About 61 percent were women and 39 percent men. Forty-two percent described themselves as white and 37 percent as black, with South Asians (11 percent) accounting for the largest proportion of other ethnicities. Most (77 percent) listed English as their first language, but 18 other languages were spoken as a first or second language at home.

“The Internet” – distinguished from the websites of TV stations or newspapers – is the main source of news for most (57 percent). Given recent indications that innovations like the smartphone and the tablet computer could be adding to, rather than replacing, news use while
raising the profile of legacy media platforms, it is especially important to look at how news reaches the audience that journalism will have, not the mythical audiences it wishes it still had.

Vultee said that while student-only samples are often criticized as nonrepresentative, and it is true that students – like most population groups – do not make decisions in the same way that military and political elites do. But it has also been shown that in “multivariate” tests – comparing a media effect across levels of media use, for example – student samples perform much like general samples.

Comments are closed.