In this paper, we develop a new way of measuring the herding behavior of market participants and test herding behavior among investors in Chinese IPO firms, compared with prior methods of herding measure developed by Christie and Huang (1995), Chang, Chen, and Khorana (2000), and Hwang and Salmon (2009). Our proposed new non-parametric herding measure, cross sectional time dispersion (CSTD), is defined differently as dispersion in IPO issuance timing, compared with traditional definitions of herding as dispersion in IPO returns or risk measure. Traditional CSSD, CSAD and beta herding measures do not provide statistically significant or consistent relationship between the herding measures and the IPO firms' initial or long-term returns. In contrast, the new measure of time herding, CSTD clearly and consistently indicates that investors are affected more by the herding behaviors of IPO firms than by those of investors in the IPO markets in China.