This study is an attempt to measure information versus propaganda in Swedish news paper editorials dealing with the question of Swedish nuclear armament. In order to categorize the content a schema was developed identifying the following content ele ments: 1) the possible defense alternatives; 2) the situation categorized into sixteen exclusive situation variables which may be affected by the choice of defense alternatives; 3) evaluations of alternatives or situations, dichotomized into good or bad; 4) general value concepts, called 'ideologies.' The sample consisted of four papers, three of which endorsed Swedish nuclear armament. The fourth did not take an explicit position. Propaganda was defined as selectivity in the presentation of factual decision premises, i.e. the descriptive statements about the situation variables. (The more propagandistic argumentation is thus uneven, with high frequencies of statements for some variables, and low for others.) We expected to find the argumentation of the uncommitted paper less propagandistic than the other three papers, but data gave no support for this hypo thesis.
A closer study of the evaluations made of consequences of different alternatives dis closed two facts: 1) the uncommitted paper could be considered as an opponent of Swedish nuclear armament, as it tended to evaluate consequences of a Swedish nuclear armament negatively and consequences of other alternatives positively. 2) By dividing the material into three time periods, it was found that all the papers became totally one-sided in their evaluations during the last time period. That is, they exclusively made positive evaluations of their endorsed alternative and negative evaluations of other alternatives.
Considering the fact that the 'uncommitted' paper could be classified as an opponent of Swedish nuclear armament, we investigated to see if we could find a tendency toward polarization. Did those who endorsed Swedish nuclear armament tend to stress some situation variables, while the paper that was against this alternative tended to stress other variables? A very slight tendency in this direction could be found. Because the sample was very small, no generalizations can be made regarding the Swedish press in general.
1966. Vol. 3, no 1, 75-88 p.