Laserfiche WebLink
<br />20 <br /> <br />Four separate parts of the sample are examined by discriminant <br />analysis: 1) survivors and the non-survivors; 2) groups that took some <br />action and those that took no action; 3) warned and non-warned groups; <br />and 4) local and non-local groups. <br />The variables on which the different categories were expected to <br />differ are those which can be measured on an ordinal scale and which are <br />known for the entire set of groups involved in 'the analysis: 1) geographic <br />location in the canyon; 2) number of people in a group; 3) origin, whether <br />the people were indoors or out; 4) knowledge of the Big Thompson Canyon; <br />5) action taken; 6) whether the group survived; and 7) whether a group <br />was warned. The last four variables listed were excluded when used as <br />the dependent variable in the analysis. Appendix E is a listing of these <br />variables and their range of values. <br /> <br />BEHAVIOR PATTERNS <br /> <br />A total of 112 groups were included for analysis: 58 survivor <br />and 54 non-survivor groups of 270 persons and 129 persons, respectively. <br />A total of 399 persons were involved in the survey. Table 4 shows the <br />range of behavior patterns found in the canyon during the flood, and the <br />origin and destination of those who survived and those who did not survive. <br />The results of the Inter-Faith survey (1976) sent to 600 flood <br />victims following the flood showed that 33% of the full or part-time <br />population in the canyon was 60 years old. One-fourth of those who died <br />were over 61 years old. Thirty-five percent of those who died were full- <br />time residents, 17% seasonal residents, 26% tourists, and 21% from parts <br />of Colorado other than the Big Thompson Canyon and the nearby towns of <br />