1. Background Information

Authors: James B. Weaver, Christopher J. Porter and Margaret E. Evans

Title: Patterns in Foreign News Coverage on U.S. Network TV: A 10 ó Year Analysis

Source: Journalism Quarterly Vol. 61:2, 356 - 363

Year: 1984 (Summer)

2. Abstract

Earlier studies have lent support to allegations by representatives of Third World countries, in particular, that Western news coverage of their countries is disproportionate, with too little overall coverage and too much concentration on sensational and bizarre items. In recent years there has been some indication that coverage of foreign in the US news has increased. This study was designed to determine: 1) whether there was a change in the volume of foreign news content presented on the evening newscasts of the three US commercial networks over the period 1972 - 1981 and, if so, what factors accounted for such changes; 2) whether there was a change in the amount of time devoted to specific geographic origins and specific topics of foreign news and, if so, why.

A content analysis of Television News Index and Abstracts from 1977 - 1981 compared with the results of an earlier analysis of the preceding 5-year period showed considerable variation in the volume of foreign news presented, but not a significant increase from the first to the second five-year period. Also, a decrease of foreign news coverage in times of exceptional domestic news events was found. The analysis also yielded a significant variation with respect to geographic origin and topical content of the news stories.

3. Null hypothesis, ALPHAgif- level and sample size per group

H01: There would be no difference with respect to the volume of foreign news content presented on the three US commercial TV networks between the period of 1972 - 76 and the period 1977 - 81.

H02: There would be no difference in the amount of time devoted to specific geographic origins of foreign news presented on the three commercial US TV networks between the periods of 1972 - 76 and 1977 - 81.

H03: There would be no difference in the amount of time devoted to specific topics of foreign news presented on the three commercial US TV networks between the periods of 1972 - 76 and 1977 - 81.

An alphagif-level was not explicitly stated in advance.

Number of observations: 180 newscasts (1977 ó 1981); 685 foreign news items (ABC, 228; CBS, 236; NBC, 221).

4. Independent and dependent variables

Independent variables: It cannot be said with certainty what the IV is because the issue of determination or causality is not addressed in the research design.

Dependent variable: Amount of newscast time devoted to foreign news item reported from abroad.

5. and 6. Instrument and procedure¹

First an "operational replication" of an earlier content analysis covering the news 1972 - 76 was conducted for the period 1977 - 81. The data were gathered from the Television News Index and Abstracts. This source had been shown to be a reliable data base describing international affairs coverage on network television's evening news broadcasts. This source seems to be the most appropriate basis for such an analysis, because often the news programs analyzed are not available any more. And if they were available, it would be almost impossible to view and analyze such an enormous amount of material.

One day per month was randomly selected from the 60 months in the 5 -year period to construct 12 composite weeks. Each newscast was broken down into stories. A story was coded for date of broadcast, network, time, geographic origin and topic category. The number of coders is not reported, but a test of coder reliability is said to have produced a reliability coefficient of 96.7% using Holsti's method. Once the data were collected they were combined with those from the earlier study(l972 - 76) to allow for an analysis over a ten - year period.

While criteria for reliability seems to be fulfilled, validity is problematic as in most content analyses. It does not seem satisfactory in terms of external validity to use seconds of newscast time as a measure of the attention paid to the problems of foreign countries. And such a measure has next to no explanatory value.

7. Statistical analysis and conclusion.

Because the data collected were not expected to satisfy the criteria of normality of distribution and homogeneity of variance, nonparametric statistical procedures were used. Kruskal-Wallis ANOVA by ranks showed considerable variation in the volume of news presented over the ten-year period(H=24.88, p 2~ .005). A contrast between the 1972-76 and the 1977-81 sample periods, however, was not significant. It is not apparent from the article how this contrast was calculated. An (apparently post hoc) comparison between years with exceptional domestic news events and years without such events revealed a significant difference at the .05 - level. Significant changes were found with respect to volume of foreign news from different geographical areas(H=18.42, p < .005.): significant declines for news from Western Europe and Asia, significant increases for news originating in the Middle East and other areas. There were also significant shifts in the attention given to particular foreign news topics.The findings of this study reject the allegations of some that the volume of foreign news presented over the three commercial US television networks has significantly increased towards the end of the 1970s. Moreover it was shown that variations in the volume of foreign news coverage can be more parsimoniously explained as negatively related to the amount of news devoted to domestic events(e.g. election years, "Watergate" years). The reverse is not true, namely that significant foreign events reduce the amount of domestic news presented. A significant increase was found in news items that relate back to domestic interest, which leads the authors to the suggestion that "much so-called 'foreign' news is in fact news of the U.S. from a foreign dateline." With regard to topics attention to not specifically crisis- or conflict-oriented news items increased, which could be seen as an indication of a less sensationalistic orientation of foreign news.

8. Possible improvements of the study.
The study is primarily descriptive and has little explanatory power. If I were to do such a study I would try to use content analysis only as one of several measures relating to foreign news coverage. I would combine it with a survey assessing changes in the awareness of the American people with regard to news, ideally a panel or a cohort study.I would also propose some changes with respect to the content analysis: I would analyze not only amount of time devoted to a particular news item, but also if they deal actually with problems of the foreign countries or problems that are primarily of interest to the US. In addition it may be of interest to compare news about foreign countries with domestic new in terms what they focus on, as well as the news coverage in foreign countries with that of the US.

1. In describing a content analysis it seems less artificial to separate 5. and 6.