Chapter 17: Descriptive Statistics Answers to Review Questions 17.1.

What is the difference between descriptive statistics and inferential statistics? The field of descriptive statistics focuses on describing, summarizing, or explaining a set of data. Inferential statistics goes beyond the immediate data and infers the characteristics of populations based on samples. 17. . !ist the three steps in constructin" a fre#uenc$ distribution. 1. List each unique number in ascending order in column one. 2. ount the number of times each number listed occurs and place the results in column t!o. ". #$ptional% onstruct a third column by converting column t!o into percentages and placing them in column three. 17.%. What t$pes of "raphical representations of data were discussed in the chapter? 1. &ar graphs 2. 'istograms ". Line graphs (. )catter plots. 17.&. Which "raphical representation is used to e'a(ine the correlation between two #uantitative variables? * scatterplot #also called a scatter graph%. It is traditional to let the + axis #the horizontal axis% represent the independent,predictor variable and let the - axis #the vertical axis% represent the dependent,outcome variable. 17.). What is a (easure of central tendenc$* and what are the co((on (easures of central tendenc$? * measure of central tendency is the single numerical value considered most typical of the values of a quantitative variable. The most common measures of central tendency are the mode #i.e., the most frequently occurring number%, the median #i.e., the middle point or fiftieth percentile%, and the mean #i.e., the arithmetic average%. 17.+. When is the (edian preferred over the (ean? .hen the numbers are highly s/e!ed #i.e., non0normally distributed%. 17.7. ,f the (ean is (uch "reater than the (edian* are the data s-ewed to the ri"ht or s-ewed to the left? The t!o general rules are 1% If the mean is less than the median, the data are s/e!ed to the left, and 2% If the mean is greater than the median, the data are s/e!ed to the right. Therefore, if the mean is much greater than the median the data are probably s/e!ed to the right.

17... What is a (easure of variabilit$* and what are the co((on (easures of variabilit$? * measure of variability is a numerical index that provides information about ho! spread out or ho! much variation is present. 17./. 0ow are the variance and standard deviation (athe(aticall$ related? The standard deviation is simply the square root of the variance. 'ence, to get the standard deviation, first get the variance as sho!n in Table 11.(, put that obtained number into your calculator, and then hit the square root /ey # %.

17.11. ,f a set of data is nor(all$ distributed* how (an$ of the cases fall within one standard deviation? 0ow (an$ fall within two standard deviations? 0ow (an$ fall within three standard deviations? The ans!er is nicely summarized in the 234, 56, 55.1 percent rule.7 That is, if the data are normally distributed, 34 percent of the cases !ill fall !ithin one standard deviation, 56 percent of the cases !ill fall !ithin t!o standard deviations, and 55.1 percent of the cases !ill fall !ithin three standard deviations of the mean. 17.11. What is a (easure of relative standin"* and what are the co((on (easures of relative standin"? * measure of relative standing is a measure that provides information about !here a score falls in relation to the other scores in the distribution of data. )ome examples are percentile ranks and standard scores. 17.1 . 0ow do $ou calculate a 23score? )ubtract the mean from the ra! score that you are given and then divide that result by the standard deviation. 'ere is the formula8 z score ; 9a! score 0 :ean 0000000000000000000000 )tandard <eviation

17.1%. What are so(e of the different wa$s to e'a(ine the relationships a(on" variables? orrelation coefficients omparing group means )catter plots Line graphs ontingency tables 9egression analysis. 17.1&. ,f $ou calculate the percenta"es in a contin"enc$ table down* then should $ou (a-e $our co(parisons down the colu(ns or across the rows?

The rules are 1% If the percentages are calculated do!n the columns, compare across the ro!s, and 2% If the percentages are calculated across the ro!s, compare do!n the columns. Therefore, in the above question you !ould ma/e your comparisons across the ro!s. 17.1). What is the difference between si(ple re"ression and (ultiple re"ression? )imple regression is based on one quantitative dependent variable and one independent variable. $n the other hand, multiple regression is based on one quantitative dependent variable and more than one independent variable. #=ote that the purpose of regression analysis is to use independent variables in predicting or explaining dependent variables.% 17.1+. 0ow is the re"ression coefficient interpreted in si(ple re"ression? The basic or unstandardized regression coefficient is interpreted as the predicted change in - #i.e., the <>% given a one unit change in + #i.e., the I>%. It is in the same units as the dependent variable. =ote that there is another form of the regression coefficient that is important but not discussed in the chapter8 the standardized regression coefficient. The standardized coefficient varies from ?1.@@ to A1.@@ Bust li/e a simple correlation coefficientC If the regression coefficient is in standardized units, then in simple regression the regression coefficient is the same thing as the correlation coefficient. 17.17. 0ow is the re"ression coefficient interpreted in (ultiple re"ression? In this case the unstandardized multiple regression coefficient is interpreted as the predicted change in - #i.e., the <>% given a one unit change in + #i.e., the I>% !hile controlling for the other independent variables included in the equation. The regression coefficient in multiple regression is called the partial regression coefficient because the effects of the other independent variables have been statistically removed or ta/en out #2partialled out7% of the relationship. If the standardized partial regression coefficient is being used, the coefficients can be compared for an indicator of the relative importance of the independent variables #i.e., the coefficient !ith the largest absolute value is the most important variable, the second is the second most important, and so on.%