Chapter 3 Visual Displays of Data

The Power of Graphs "The Most Misleading Graph Ever Published" Graphs can be misleading. - As critical thinkers, we want to know whether

(1) a sample represents a population, (2)- how the variables were actually measured, (3)- whether a graph tells an accurate data story.

The Power of Graphs

- "The Most Misleading Graph Ever Published" -Data Ethics: How to Lie with Graphs

The Power of Graphs -Data Ethics: How to Lie with Graphs

- Ethical researchers develop graphs that clarify, rather than confuse. - So avoid misleading tricks, including the following: - biased-scale lie. - sneaky-sample lie - inter-polation lie - extra-polation lie - inaccurate values lie.

Types of Chartjunk:

- Moiré vibrations: - Grids: - Ducks:

How to Build a Graph - Choosing the Appropriate Type of Graph AND How to Read a Graph:

- One scale variable: (X-axis) (with frequencies): -Two Scale variables: One scale (independent) variable on the x-axis and one scale (dependent) variable on the y-axis: -Two variables but : One nominal or ordinal independent variable(Y-axis) and one scale dependent variable:(X-axis).

-Two Scale variables: One scale (independent) variable on the x-axis and one scale (dependent) variable on the y-axis:

- Scatterplot and - Line Graph

· Graphs with one nominal (or sometimes ordinal) independent variable and a scale dependent variable: (REMEMER INDEPENDENT VARIABLE can have several LEVELS) Meaning: Two Scale varaible: One norminal/ordnal variable(Y) and One scale variable.(X)

-Bar Graphs -Pareto graph -Pictorial Graphs - Pie Charts

-Two variables but : One nominal or ordinal independent variable(Y-axis) and one scale dependent variable:(X-axis).

-Bar graph or Pareto chart >or PIie Chart

- One scale variable: (X-axis) (with frequencies):

-Histogram or - Dot plot

The Power of Graphs - Common Types of Graphs:

-Scatterplots -Line Graphs -Bar Graphs -Pareto graph -Pictorial Graphs -Pie Charts

Pie Charts

A ____ is a graph in the shape of a circle, with a slice for every level (category) of the independent variable. _________ uses slices = proportion of variable There are numerous problematic examples of ________. Data can almost always be presented more clearly in a table or bar graph than in a __________.

Pictorial Graphs

A ______ is a visual depiction of data typically used for an independent variable with very few levels (categories) and a scale dependent variable. _______ uses visual depiction that uses pictures or symbols _______ usually used when there are few categories listed on the x-axis and a scale variable on the y-axis. Eye-catching __________________are far more common in the popular media than in research journals. They tend to direct attention to the clever artwork rather than to the story that the data tell.

Scatterplots,

A _______ is a graph that depicts (shows) the relation between two scale variables. _____ shows the relationship between two scale variables using a dot for each data point ______ are graph(s) that depict the relationship between two scale variables _____Shows every data point via Linear relationships or • Nonlinear relationships Great for identifying potential outliers

Line Graphs

A _____________ is used to illustrate (draws out) the relation between two scale variables. 1) - One type of ________ is based on a scatterplot and allows us to construct a line of best fit that represents the predicted y score for each x value. (Scatterplot with a line). ____ are useful for illustrating trends ______ are used to clarify the trend in a scatterplot, thus we can add a line of best fit -Where as ! 2) - a second type of line graph allows us to visualize changes in the values on the y-axis over time it is called .(Time Series Plot). _ · (Review Line of Best fit: The line of best fit allows us to make predictions about a person's value on the y variable from his or her value on the x variable.)

Nonlinear relationships

A________ relation between variables means that the relation between variables is best described by a line that breaks(not continuous can dip) or curves(U shape or upsidedown U shape) in some way. _______ doesn't appear to have strong linear association but there is a relationship between the variables) Seen in a Scatterplot graph.

Bar Graphs

A____________ is a visual depiction of data in which the independent variable is nominal or ordinal and the dependent variable is scale. The height of each bar typically represents the average value of the dependent variable for each category.

How to Build a Graph - Guidelines for Creating a Graph

Avoid: Chartjunk • Avoid chartjunk—any unnecessary aspect of a graph that detracts from its clarity.

defaults

Computer _________ : are the options that the software designer has pre-selected; they are the built-in decisions that the software will implement if you do not instruct it otherwise. · Several computer-generated graphing programs have ___________ that correspond to many—but not all—of these guidelines.

Pareto graph

__________ a type of bar graph in which the categories along the x-axis are ordered from highest bar on the left to lowest bar on the right. Key: Order along the X-axis. As well ordered from highest to lowest As well a type of bar graph.

What is the purpose of a graph?

The purpose of a graph is to reveal and clarify relations between variables. -"Graphs are powerful because they can display the relation between two or more variables in just one image."

Bar graph vs Pareto chart vs r PIie Chart

We use a ____________If there are two or more nominal or ordinal independent variables and one scale dependent variable. Where as We use __________ if the independent variable has many levels ( which categories along the x-axis are ordered from highest to lowest ) Where as ______ are least used and they represent percentages).

Ducks:

______ are features of the data that have been dressed up to be something other than merely data (e.g., three-dimensional bars). Unnecessary pictures and fancy fonts would also fall into this category. Keyword: data dressed up as something else.

- One scale variable: (X-axis) (with frequencies): -Histogram vs - Dot plot

_______ a graph that depicts one scale variable, with the values of the variable on the x-axis (horizontal axis) and frequencies on the y-axis (vertical axis)._ ________are excellent for quickly seeing the shape of a distribution of data. Depending on the data that we are working with, we might group sets of values into intervals. where as ______________a graph that displays each data point in a sample, with the range of scores along the horizontal axis and a dot for each data point above the appropriate value. These _______ show us the shape of our data while still letting us observe the individual data points.

- inaccurate values lie.

________ scaling that distorts portions of the data _____________ is a lie that tells the truth in one part of the data but visually distorts it in another place. Exp: Notice in the figure above that just four stick figures are meant to represent more than 43,000 nurses in the first three columns. But when only a few thousand more nurses are included, as in the last column, 40 stick figures are used. The proportional change in the number of stick figures is much larger than the proportional change in the size of the data. Keyword: distortion. Distortion of data in a graph

- biased-scale lie.

__________ pushes people toward a certain responses or biased interpretations. Exp: reviewers use five stars to indicate that a restaurant's food, service, and ambience are "almost perfect" and one star means "good." There's no option for bad! Apparently, you can't buy a bad meal in New York City if a New York magazine reviewer has eaten there.

- extra-polation lie

___________ - Implying (or allowing people to assume) that a pattern will continue beyond the range of the data _________is a lie that assumes that values beyond the data points will continue indefinitely. Exp: 4. In 1976, The Complete CB Handbook assumed that U.S. elementary schools would soon have to teach students how to communicate with CB radios, their popularity then growing exponentially. They are now used mostly by long-distance truckers. What happened? Mobile phones. Do not assume that a pattern will continue indefinitely. Keyword: beyond the range of the data!

bubble graph

___________ is a graph that resembles a scatterplot, but the dots are replaced by bubbles that can represent additional variables through their color and size.

Char-tjunk

____________ is any unnecessary information or feature in a graph that detracts from a viewer's ability to understand the data. Avoid________ it—any unnecessary aspect of a graph that detracts from its clarity.

- sneaky-sample lie

____________ select participants purposefully for bias or allow self-selection. Exp: You might pick up some useful information from Web sites that rate professors, but be cautious. The students most likely to supply ratings are those who strongly dislike or strongly approve of a particular professor. A self-selected sample means that the information might not apply to you.

Grids:

____________ that takes the form of a background pattern, almost like graph paper, on which the data representations, such as bars, are superimposed . Keyword background and pattern?/

Line Graphs - Time Series Plot(aka Time plot)

_____________ : is a graph that plots a scale variable on the y-axis as it changes over an increment of time (e.g., days, years) labeled on the x-axis Scale variable/ a time One type of a Line Graph

- inter-polation lie

______________ assuming that some value between the data points lies on a straight line between those data points. ________ Assumes a straight line between data points Exp: Canada had its lowest rate of break-ins (property crime) since the 1970s (see the figure below on the left), but you cannot assume a gradual decline over 30 years. In the years leading up to 1991, there was a dramatic increase in property crime. Make sure that a reasonable number of in-between data points have been reported. key word between not outside ????

Scatterplot vs Line Graphs

______________depict the relationship between two scale variables using a dot for each data point • Shows every data point Shows both: Linear relationships • Nonlinear relationships - Great for identifying potential outliers Where as. __________________ shows the relationship between two scale variables using a line. Useful for illustrating trends of dot plots as well Can be used as : Time series plot which is a graph that plots a scale variable on the y-axis as it changes over an increment of time (e.g., days, years) labeled on the x-axis So good for a scale variable over time: Exp Temperature rising over a period of one week:

Moiré vibrations :

___________any visual patterns that create a distracting impression of vibration and movement. (avoid patterns : use shades of gray instead of patterns.) Keyword : impression of vibration!

Line Graphs -Line of Best fit:

________the line of best fit allows us to make predictions about a person's value on the y variable from his or her value on the x variable.) - illustrating trends One type of a Line Graph

Linear relationship(s) or Linear relation : a______

a___________ relation between variables is best described by a straight line. seen in Scatterplots graph

· Types of graphs that have two scale variables

o -Scatterplots, - Line Graphs