Toward a non-hierarchical structure for newswriting : the effects of compound sentences with semicolons on perceptions of readability and objectivity

The purpose of this study was to determine the effects of non-hierarchical, or coordinated, structure in newswriting on readers' perceptions of readability and objectivity. To operationalize non-hierarchical structure, a news story was written in three versions, all with identical content but with structural differences as follows: Version One (control)--journalism's traditional inverted pyramid model of organization with no semicolons; Version Two (experimental)--coordinated model of organization using compound sentences with semi-colons to juxtapose opposing arguments; Version Three (experimental)--coordinated model of organization identical to Version Two but with periods in place of semicolons between independent clauses to form separate simple sentences for juxtaposed, opposing arguments.

It was hypothesized that the coordinated versions would be perceived as more objective but less readable than and less preferable to the inverted pyramid version. Further, it was hypothesized that the coordinated version with semicolons would be perceived as more objective but less readable than and less preferable to the coordinated version with no semicolons.

An after-only laboratory experiment was conducted with 109 college students enrolled in a junior-level marketing class. Subjects were randomly but unobtrusively assigned to three groups for reading the three story versions.

Dependent measurement was via seven-step scales between twenty-one pairs of descriptions, opposites in meaning, that operationalized readability, objectivity, and preference.

One-way analysis of variance, coupled with Scheffé's multiple comparison test and the test of Least Significant Difference, showed significant differences between pairs of groups on the following measures: reading ease, simplicity, coherence, believability, and cumulative readability. To summarize findings, the coordinated version with semicolons was perceived as less readable and less believable than the other versions.

Though the data were insufficient for accepting the hypotheses on objectivity and on preference, findings did suggest that the semicolon alone caused lower scores on readability.

Both quantitative data and open-ended remarks requested from subjects suggest that the semicolon acts not only as a cue for complex-ity in writing that is cognitively simple, but also as a clue about the believability of a message.

Further research was recommended to test additional variables in the coordinated model that may impinge on readability and to explore more precisely the interaction of readability, objectivity, and believability as a function of newsreaders' fantasies about imagined writers based on stylistic cues.