Ch3: Visual Displays of Data

biased scale lie

uses scaling to skew results

sneaky sample lie

occurs when participants are preselected or self selected to provide data

interpolation lie

assumes that a variable between two data points follows the same pattern

scatterplots

• depicts relation between two continuous variables

• observe every data point

• frequently used to depict correlations

• could tell no relation, linear relation, or nonlinear relation

linear relation

the relation between variables is best described by a straight line

nonlinear relation

the relation between two variables is described by a line that breaks or curves in some way

• illustrated as a inverted U-curve

line graph

used to show relation between two scale variables

shows changes in the values on the y-axis

line of best fit

graph based on a scatterplot, represents the predicted y score for each x value

time plot

plots scale variable on the y-axis as it changes over increment of time labeled on x-axis

bar graph

independent variable is nominal/ordinal, dependent variable is a scale

bars do not touch each other

x-axis indicates discrete levels of a nominal/ordinal variable

y-axis may represent counts, percentages, or continuous variables

how to choose a graph

one scale variable (with frequencies): histogram or frequency polygon

one scale independent and dependent variable: scatterplot or line graph

one nominal or ordinal independent variable and one scale dependent variable: bar graph

two or more nominal or ordinal independent variables and one interval dependent variable: bar graph