Interval/Ratio Variables-freuquency table-histogram-median, mean, standard deviation-deviation is the basis for assessing variability for finding the SD. Can be positive (when response is above the mean) and negative (if below the mean)-skews and outliers-skew- look at the median and then the mean• The “1.5 IQR rule”– An observation/value will be considered unusually low if it is less than a “lower boundary” =Q1-(1.5∙IQR)– An observation/value will be considered unusually high if it is greater than an “upper boundary” =Q3+(1.5∙IQR)• The “3 standard deviation rule”– An observation/value will be considered unusually low if it is less than a “lower boundary” = x -(3∙S_x)– An observation/value will be considered unusually high if it is greater than an “upper boundary” = x +(3∙S_x)• Then do outliers of lower and upper and you compare them to the minimum and maximum. • -to do outliers you do count if (and then “<the lower boundary”) and ">then the upper bounder"•Comparing an interval/ratio variable between two groups -frequency ditributions for each group-determine median, mean, and SD-make comparisons for each one. (find differences and compare, not median)-compare skew and relative presense of outliers • Mean Interpretation:– If the difference is positive, it indicates the first of your two means is larger– If the difference is negative, it indicates the second of your two means is larger– The general template for interpreting this new statistic is: “The average (title of interval/ratio variable) for (group 1) is (value of the difference between means, without the positive/negative sign) (“lower” if the sign is negative; “higher” if the sign is positive) than the average (title of interval/ratio variable) for (group 2).”• Pearson’s correlation is a single number that tells us something about thestrength and direction of an association between two interval/ratio variables–Direction:•Negative values indicate that as one variable increases in value the other decreases•Positive values indicate that as one variable increases in value so does the other–The correlation can be as low as -1.0000 and as high as +1.0000•± 1.0000 à perfect (very strong) association•± 0.6667 à strong association•± 0.3333 à moderate association•± 0.1000 à weak association•0.0000 à no associationWhen you are making a prediction you take the intercept + the slope * the predicted number-extrapolation•Template for interpreting the partial slope for nominal/ordinal variables: “The mean (title of dependent variable) for (focal group) is (value of the partial slope, without the positive/negative sign) (“lower” if the sign is negative; “higher” if the sign is positive) than the mean (title of dependent variable) for (reference group).”•Template for interpreting the partial slope for interval/ratio variables: “If (title of the X variable {label of what slope you’re looking at} ) increases by (what a 1 unit means for X), we predict that (title of the variable we’ve labeled Y [dependent variable] ) will (“decrease” if the sign is negative; “increase” if the sign is positive) by (value of the slope without the positive/negative sign).”Scatterplot-the variable we think causes the change in the other variable (independentvariable) is labeled X and dependent variable