We began this chapter by studying how to depict the shape of the distribution of a data set. We learned that stem-and-leaf displays and histograms are useful graphics for portraying a data set's distribution. We also learned about some common population shapes. We saw that data sets often have shapes that are symmetrical, skewed with a tail to the right, or skewed with a tail to the left.

Next we presented and compared several measures of central tendency. We defined the population mean and we saw how to estimate the population mean by using a sample mean. We also defined the median and mode, and we compared the mean, median, and mode for symmetrical distributions and for distributions that are skewed to the right or left. We then studied measures of variation (or spread). We defined the range, variance, and standard deviation, and we saw how to estimate a population variance and standard deviation by using a sample. We learned that a good way to interpret the standard deviation when a population is (approximately) normally distributed is to use the empirical rule, and we applied this rule to assess process capability. We next studied Chebyshev's Theorem, which gives us intervals containing reasonably large fractions of the population units no matter what the population's shape might be. We also saw that, when a data set is highly skewed, it is best to use percentiles and quartiles to me sure variation, and we learned how to construct a box-and-whiskers plot by using the quartiles.

After learning how to measure and depict central tendency and variability, we presented several methods for portraying qualitative data. In particular, we used bar charts and pie charts for this purpose. We also discussed using a sample to estimate the proportion of population units that fall into a category of interest.

We concluded this chapter with several optional sections. First, we studied using scatter plots to examine relationships between variables. Next we discussed misleading graphs and statistics, and we explained some of the tactics that are commonly used to try to distort the truth. We also introduced the concept of a weighted mean and then explained how to compute descriptive statistics for grouped data. Finally, we showed how to calculate the geometric mean and demonstrated its interpretation.