||II. Methods||Page 2|
Four groups of students, 126 in all, took part in the study by filling out questionnaires: 74 students in an introductory psychology course at a prestigious liberal arts college; 22 students at that same college who had enrolled in a course on the psychology of war and peace; 19 students of nursing at a community college nearby; and 11 peace activists at another nearby community college. Students at the liberal arts college came mostly from families of relatively affluent professionals (teachers, doctors, lawyers, business executives, etc.), while students at the community colleges came mostly from working class families.
Questionnaires were administered on two occasions, one month apart. Of the 126 students who filled out the initial questionnaires, 114 completed the follow-up set of questionnaires a month later. In this way it was possible to measure peace activity that occurred after the students had already indicated their attitudes regarding activity and related factors. In other words, we could measure how much these factors predict future activity.
The initial questionnaires were administered in two parts: (I) a questionnaire with 16 questions designed to measure attitudes about the efficacy of peace activity, beliefs concerning human nature and war, feelings of anger about the arms race and threat of nuclear war, and normative attitudes about peace activity on the part of family, friends and school; and (2) an inventory of peace-related activities in which the students had been engaged up until that time ("past activity"). Table I shows 10 of the 16 questions on the questionnaire, which are the ones that are analyzed in the succeeding correlation matrices. Table II shows an activity inventory; although it is the follow-up inventory that is shown, it is similar in most respects to the initial inventory as well.
The follow-up questionnaires, administered one month later, asked the students for: 1) an inventory of peace-related activities in which the student had been engaged during the intervening month ("future activity"); 2) attitudes concerning several current events related to war and peace; and 3) a pair of questions concerning the relation of beliefs about human nature to activity. The inventory of peace-related activities is shown in Table 2. The questions about belief and activity were the following: a) " After you began to take part in peace activities, did you find yourself becoming more optimistic about human nature and human potential to create a peaceful world?"; and b) "Before you took part in activities for peace, were you discouraged from becoming active by a pessimistic view that humans are intrinsically violent and therefore wars are inevitable?" Only those students who indicated that they had taken part in peace activities were asked to answer the last two questions and only their responses were scored.
Activity scores were calculated as shown in Table II. A "past Activity" score was calculated by summing the scores (1-3) on individual questions in the initial activity inventory. A "future activity" score was calculated by summing the scores (1-3) on individual questions in the follow-up activity inventory as shown in Table 2. Future activity scores ranged from 1 to 27 (mean=5.2). A score of 1 was added to each person in order to avoid zero data.
Scores on other factors were calculated as follows. The answer to each question consisted of a check by one of the following: "very much," "somewhat," "very little," or "not at all" which were scored as 4, 3, 2 or 1 respectively. However, in order to ensure that most correlations would be positive, the scores for the belief question (see Table I) were scored in reverse, i.e. "very much" was scored as 1, "somewhat" as 2, etc. Two sets of highly correlated responses (r =.400 or more) were combined into composite scores as shown in the Table: belief about human nature; and attitude to peace activity. These two composite scores were measured as simple sums of the components; hence they had a range of 1-12 rather than 1-4 as in the case of other simple scores.
Data analysis was done by means of correlation and partial correlation using the computer program of SPSSX. A correlation matrix was calculated for all of the following seven items obtained in the initial set of questionnaires (n = 126): anger score; family normative attitude toward peace activity; friends normative attitude; school normative attitude; past activity score; belief composite score; and attitude composite score (Table III). Another set of correlations were calculated for each of the same seven items in relation to future activity score (n = 114), also shown in Table III. A partial correlation matrix was then calculated for all of the possible correlational pairings of the seven items obtained in the initial questionnaire (n = 126). Each correlation was partialed by the other five items. As shown in Table IV, of the resulting 21 correlations, 11 were significant at a level of p = 05. Finally, another set of partialed correlations was obtained for each of the seven items in relation of future activity, in which all other items except for past activity were partialed out (also shown in Table IV).
The purpose of the partial correlation data analysis was to determine the significance of various correlations over and above their shared relationship with other intervening variables. In other words, significant partial correlations suggest that there was a causal relationship that was not due to the intervention of any other factor measured in the study.
(End of Methods)