Skewness - measures the degree and direction of symmetry or asymmetry of the distribution. A normal or symmetrical distribution has a skewness of zero (0). But in the real world, normal distributions are hard to come by. Therefore, a distribution may be positively skewed (skew to the right; longer tail to the right; represented by a positive value) or negatively skewed (skew to the left; longer tail to the left; with a negative value). Kurtosis - measures how peaked a distribution is and the lightness or heaviness of the tails of the distribution. In other words, how much of the distribution is actually located in the tails? A normal distribution has a kurtosis value of zero (0) and is said to be mesokurtic. A positive kurtosis value means that the tails are heavier than a normal distribution and the distribution is said to be leptokurtic (with a higher, more acute "peak"). A negative kurtosis value means that the tails are lighter than a normal distribution and the distribution is said to be platykurtic (with a smaller, flatter "peak"). This short article contains simple descriptions of skewness and kurtosis, along with several other descriptive statistics: http://www.ehow.com/how_2242574_interpret-descriptive-statistics.html