The fuzzy c-means (FCM) algorithm is a popular fuzzy clustering method. It is known that an appropriate assignment to feature weights can improve the performance of FCM. In this paper, we use the bootstrap method proposed by Efron [Efron, B., 1979. Bootstrap methods: Another look at the jackknife. Ann. Statist. 7, 1–26] to select feature weights based on statistical variations in the data. It is simple to compute and interpret for feature-weights selection. Compared with the feature weights proposed by Wang et al. [Wang, X.Z., Wang, Y.D., Wang, L.J., 2004. Improving fuzzy c-means clustering based on feature-weight learning. Pattern Recognition Lett. 25, 1123–1132], Modha and Spangler [Modha, D.S., Spangler, W.S., 2003. Feature weighting in k-means clustering. Machine Learn. 52, 217–237], Pal et al. [Pal, S.K., De, R.K., Basak, J., 2000. Unsupervised feature evaluation: A neuro-fuzzy approach. IEEE Trans. Neural Networks 11, 366–376] and Basak et al. [Basak, J., De, R.K., Pal, S.K., 1998. Unsupervised feature selection using a neuro-fuzzy approach. Pattern Recognition Lett. 19, 997–1006] we find that the proposed method provides a better clustering performance for Iris data and several simulated datasets based on error rate criterion and also performs well in color image segmentation according to Liu and Yang’s [Liu, J., Yang, Y.H., 1994. Multiresolution color image segmentation technique. IEEE Trans. Pattern Anal. Machine Intell. 16, 689–700] evaluation function.