Investor sentiment becomes more and more popular to be investigated in the financial market. This study aims to analyze the relationship between investor sentiment and market returns in Chinese market. We divide our work into two dimensions. In the first part, we focus on solving the question of how to measure investor sentiment in China. Two direct sentiment proxies are found as the representative sentiment indexes. In addition, this study uses the method of Baker and Wurgler (2006) for reference, employing the technique of PCA to construct four composite sentiment indexes. So, totally, six sentiment indexes are introduced to represent the Chinese investor sentiment. Then, in the second part, we test the efficiency and effectiveness of each index of studying the relationship between the investor sentiment and the market return as well as the cross-section portfolio returns from 16 different firm characteristics by carrying out the correlation analysis, sorting skills, and both the uni-variable and multi-variable time series regressions. We find that investor sentiment do affect Chinese stock market. Conditional on positive/negative sentiment market, the cross-sectional returns of different firm characteristics show different patterns. When in high sentiment period, companies with younger age, higher dividend payout, more tangible assets, high growth rate and low financial burden usually have lower return. No obvious trend or totally opposite phenomena could be seen when sentiment is low.