This study surveys domestic and international research on investor sentiment indices and describes the proxy variables that have been discussed in the behavioral finance field for constructing investor sentiment indices, while investigating investor sentiment indices applicable to the Korean stock market. Specifically, among the discussed investor sentiment indices, the study examines those with high explanatory power for individual firm-level stock returns in Korea. Using manufacturing firms listed on the Korea Exchange (KOSPI) from 2010 to 2017, the study compares and analyzes the explanatory power of single-variable investor sentiment indices based on individual investors' buy-sell trading volume imbalance and listed stock turnover ratio against a multi-variable investor sentiment index constructed using the methodology of Baker and Wurgler (2006, 2007). The empirical results show that even after controlling for firm-characteristic variables that affect individual stock returns, the investor sentiment index generated using multiple variables has higher explanatory power than the single-variable index, and the same results were obtained after controlling for risk factors and option investor sentiment indices. The findings of this study imply that multi-variable investor sentiment indices have greater usefulness than conventional single-variable indices in terms of appropriately reflecting investor sentiment for individual firms in the Korean market.