|
ABSTRACT Nationwide Newspaper Coverage of Music Censorship:
A nationwide study of music censorship compared hypotheses linking different city characteristics and nationwide newspaper coverage, using the “community structure” approach to research, as developed in nationwide studies by Pollock and others (1977, 1978, 1994-2002). A national cross section sample of 14 city newspapers was drawn using the Lexis-Nexis, Dialog Select, and Newsbank databases, sampling all articles 150 words or more in length between 1981 and 2001, resulting in 144 articles. The articles were coded for both prominence and article direction (favorable, unfavorable, or neutral), then combined to calculate a single score “Media Vector” for each newspaper. The scores ranged from 0.284 to -0.401, with positive scores favoring and negative scores opposing music censorship. Pearson Correlations linking city characteristics and Media Vectors supported both “stakeholder” and “violated buffer” hypotheses, falling into four clusters: partisanship stakeholders (% Voting Democratic - r =-.518, p = .029, % Voting Republican - r = .621, p = .009); belief system stakeholders (devotional reading – r =.673 p = .004); lifecycle position stakeholders (families with children under 7 – r = .519 p = .029); and “violated buffer”, associating privilege with opposition to censorship (4 Years college – r = -.505 p = .033). Regression analysis revealed that four variables account for 92% of the variance in newspaper coverage: % Republican voters and % Democratic voters (accounting for 68% of the variance), number of FM Radio stations, and number of Music Dealers. Paper presented at the annual conference of the International Communication Association, San Diego, May, 2003 |