Interval data: Real numbers, i.e. heights, weights, prices Box plot -Values are real numbers Outliers: -All calculations are valid less than 25th Percentile-Data may be treated as ordinal or nominal 1.5*interquatile Ordinal Data: order poor = 1, fair = 2, good = 3, very good = 4 or -Values must represent the ranked order of the data. Larger than 75th -Calculations based on an ordering process are valid. Percentile+1.5*interquatile -Data may be treated as nominal but not as interval. Nominal Data: Single = 1, Married = 2, Divorced = 3, Widowed = 4 -Values are the arbitrary numbers that represent categories. -Only calculations based on the frequencies of occurrence are valid. -Data may not be treated as ordinal or interval. -Bar Charts are often used to display frequencies Measures of Linear Relationship (strength&direction) -Pie Charts show relative frequencies -covariance(Sxy): when 2 variables move in the same direction, Relationship between Two Nominal Variables covariance will be large positive number, if in opposite direction it will be -Cross-Classification Table of Frequencies negative. -Row Relative Frequencies If no particular pattern, it will be small number -Graphing the Relationship between 2 Nominal Variables (Interpret) If the two variables are unrelated, the patterns exhibited in the bar charts should be approximately the same (EX) Do Male and Female American Voters Differ in Their Party Affiliation? (Interpret) there are substantial differences between the bar charts for men and women. We can conclude that gender and party affiliation are related. Specifically, men tend to identify themselves as independents, -coefficient of correlation(r): How strong is association between X and Y? whereas women support the Democratic party. .when close to 1: strongly positively related .when close to -1: strongly negatively related Building a Histogram (for Interval