Identifying and tracking research confidence trends based on information extracted from literature titles

The important impact of research confidence on research output has been confirmed in numerous studies. However, these studies have generally evaluated research confidence based on scales, with researchers subjectively reporting their research confidence levels. Accordingly, how can research confidence be evaluated through alternative indicators beyond subjective self-report measures? Using the title information from journals included in the Chinese Social Sciences Citation Index (CSSCI) as the primary dataset, the present study constructed judgement dimensions to assess research confidence and offers an analysis of changing trends in research confidence, as reflected in the title information of CSSCI papers. Based on interviews and the behavioural decision-making paradigm, the analysis drew on the title data from 5928 articles across 379 CSSCI journals from 1992 to 2017. In order to simulate these changes, this study adopts a hierarchical linear growth model. The results show that superficial words in a paper’s title can be a useful index for evaluating research confidence. The hierarchical linear growth model’s results reveal a consistent decline in the use of superficial title expressions by Chinese social science researchers over the past 26 years. Variations in this trend can be accounted for by factors such as journal category and discipline category. In conclusion, this study underscores the potential of literature title information as a reliable dimension for assessing research confidence. Applying this indicator, we discovered a significant transformation in the expression of research confidence among Chinese social science researchers over the last 26 years.