Skewness and kurtosis describe the shape of a data distribution. Skewness uses the third power to measure asymmetry around the mean, identifying whether the tails are shifted to the left or right. Kurtosis, using the fourth power, measures the "sharpness" of the peaks and the weight of the tails, giving more importance to extreme values. Unlike skewness, which is directional, kurtosis focuses on the magnitude of the deviations.