Seite - 202 - in Critical Issues in Science, Technology and Society Studies - Conference Proceedings of the 17th STS Conference Graz 2018

Fig. 3b. Increase in likelihood-to-eat score. Fourteen selected agents. Dashed red line shows gradual increase. Results Below, we present observed changes in eater-noneater counts as generated by the models. We conducted fifteen runs per condition. In order to test increase among eaters, we employed a Wilcoxon matched-pairs signed-rank test (including an alpha correction). We conceptualized each model world state as a variable with two observations: a before and an after. Each procedure (pre-, social facilitation, post-) was a unit/case. Table 1 presents some summarized data analysis from the earlier study. Table 1. Results of earlier study. Mean used as parameter We ran the same test for the same conditions during the current iteration. Results are presented in Table 2 below. All conditions showed a significant increase of eaters. None of the 95% confidence intervals per condition contained zero. While increase in eaters is recorded in both cases, it is notable that increases in the beta- weighted conditions are more modest than those of the mean-weighted conditions. The former 202