Based on cointegration test, error correction model and Granger causality test， the empirical study on the construction investment and economic growth in China during 1995 to 2008 has been conducted. The study shows that there exists a longterm cointegration relationship between the construction investment and economic growth. The evidence shows that the construction investment is significantly positively related to economic growth. The study also shows that there exists twoway Granger causality between the construction investment and economic growth. Variance decomposition analysis indicates that the impact of construction investment on economic growth is much larger than that of economic growth on construction investment. The policy implication is that, economic growth could be effectively adjusted by controlling construction investment.