Scribd
Upload
4 views

Chapter 03

The document discusses the uses and potential misuses of graphs in data presentation, highlighting techniques that can mislead viewers, such as biased scales and inaccurate values. It also outlines common types of graphs, including scatterplots, line graphs, and bar graphs, along with guidance on when to use each type based on the nature of the data. Additionally, it provides instructions for creating these graphs using SPSS software.

Uploaded by

Diego Bado
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
4 views

Chapter 03

The document discusses the uses and potential misuses of graphs in data presentation, highlighting techniques that can mislead viewers, such as biased scales and inaccurate values. It also outlines common types of graphs, including scatterplots, line graphs, and bar graphs, along with guidance on when to use each type based on the nature of the data. Additionally, it provides instructions for creating these graphs using SPSS software.

Uploaded by

Diego Bado
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 41

Visual Displays of

Data

Chapter 3
Uses of Graphs
• Positive and negative uses
– Can accurately and succinctly present information
– Can reveal/conceal complicated data
• Graphing in the information age: a critical skill
– http://www.datavis.ca/gallery/
How to Mislead
with Graphs

“The Most Misleading


Graph Ever Published”

The Cost and Quality


of Higher Education
“The Most Misleading Graph Ever Published”

The Cost and Quality of Higher Education

• So why bad?
– The times are not the same for both lines (going
across)
– The scales are not the same
• So the starting point is also a problem
– Rankings … down = BETTER, so why did they graph
ranks that way?
Techniques for Misleading
• The false face validity lie
– Method seems to represent what it says, but does not actually.
• e.g., using yelling as a measure of aggression
• The biased scale lie
– Scaling to skew the results, or leading questions
– i.e. goodreads book scale (did not like, it was ok, liked it, really
liked it, it was amazing)
• The sneaky sample lie
– When participants are preselected or self selected to provide
data.
– Like rate my professor!
More Techniques

• The interpolation lie


– Assumes that a value between 2 data points
follows the same pattern
• The extrapolation lie
– Assumes knowledge outside of the study
• The inaccurate values lie
– Uses scaling to distort portions of the data
• The outright lie:
– Making up data!
The Inaccurate Value Lie
Which of the two graphs is misleading?
Common Types of Graphs
• Scatterplots
– Graphs that depict the relation between two scale
variables
– Range-frame (!!) – x and y axis show the range of
the variable
• Eep?
– Observing every data point
• Linear relationships – relationship between variables is
a straight line
• Nonlinear relationships – relationship between
variables is curved
– Performance & Anxiety
Scatterplot of Hours Studied and
Statistics Grade
SPSS Scatterplots
• Graphs > chart builder
– Hit ok if you get a define message
SPSS Scatterplots
• Click scatter / dot on the left hand side
• DOUBLE click or drag the first box (top left) to
the design window
SPSS Scatterplots
SPSS Scatterplots
• Drag a variable into X and a variable into Y
SPSS Scatterplots
SPSS Scatterplots
• Hit ok!
• (please note that the picture it creates for you
IS NOT the real scatterplot).
Line Graphs
• Line graphs are best for scale variables
(remember this is interval/ratio)
– Especially useful for trends over time
• Line of best fit
• Time series plot
The Line of Best Fit
SPSS Line of Best Fit
• Double click on your scatterplot and new
window will pop up
SPSS Line of Best Fit
• Hit close and then also close the pop up
window to get the line in the output file.
Weekly
Newspaper
Circulation
SPSS Line Graphs
• Graphs > chart builder
• Click line on the left
• Select the first box (double click or drag and
drop)
• Drag one variable in X and one in Y
Bar Graphs
• When the independent variable is nominal or
ordinal and the dependent variable is scale
– Meaning X = groups, Y = interval/ratio
– Each bar usually represents and AVERAGE score.
• Pareto chart: bar graph in which categories along
the x-axis are ordered from highest to lowest
– (I have never heard this term before, so if you see a
question on it in the homework let me know so I can
take it out).
The Flexibility of the Bar
Graph
Bar Graphs Highlight Differences
Between Means
Deceiving with the Scale
SPSS Bar Graphs
• Graphs > chart builder
• Click bar on the left.
• Select the first option (double click or drag and
drop).
• Put a categorical variable in X, a continuous
variable in Y.
Two others
• Pictorial graphs
– Meh?
• Pie Charts
– I think you are all smart enough to know what a
pie chart is … as well as they are not really used in
research papers (frequency charts!)
• So we are going to leave these out.
Frequency Polygons / Histograms
• Graphs > chart builder
• Select histogram
• Histogram = first box, frequency polygon =
third box
• This is different: only put something in X…
Check Your Learning
• What is the best type of graph to depict the
following:
– Depression levels and stress levels for 150
university students. Is depression related to
stress?
– Mean years of education for six regions of the
United States. Are education levels higher in some
regions than in others?
Choosing the Graph Based on
Variables
1. One scale variable (with frequencies): histogram or
frequency polygon
2. One scale independent and one scale dependent
variable: scatterplot or line graph
3. One nominal or ordinal independent and one scale
dependent variable: bar graph
4. Two+ nominal or ordinal independent and one
interval dependent variable: bar graph

You might also like