LAUSR

We report three studies in which we applied a value dictionary to narratives. Our objective was to test a theory-driven value dictionary for extracting valuable information from autobiographical and narrative texts. In Studies 1 (N = 106) and 2 (N = 152), participants wrote short autobiographical narratives and in Study 3 (N = 150), participants wrote narratives based on ambiguous stimuli. Participants in all three studies also completed the Portrait Value Questionnaire as a self-report measure of values. Overall, our results demonstrate that it is possible to extract value-relevant information from these narratives. Extracted values from autobiographical narratives showed average correlations of 0.07 (Study 1) and 0.12 (Study 2) with self-reports compared to an average correlation of 0.01 for the extracted values from implicit motive tasks (Study 3). The correlations with self-reports were in line with previous validation studies. The most salient values in narratives diverged somewhat, with a stronger emphasis on achievement values compared to self-reports, probably due to the nature of salient episodes within one's life that require demonstrating success according to social standards. Benevolence values were consistently most important in both self-ratings and text-based scoring. The value structure emerging from narratives diverged from the theoretically predicted structure, yet broad personally vs. socially focused value dimensions were qualitatively discernible. We highlight opportunities and challenges for future value research using autobiographical stories.