Effective Presentations

1
Effective Presentations

• Louis Woynarowski
• CHAS/GPHAP Administrator
• lwoynaro_at_uchicago.edu

2
What is a Presentation?

• The purpose of a presentation is the transfer of
  information, whether for didactic or persuasive
  purposes.
• The presenters obligation is to provide
  sufficient information to
• advance the discourse,
• not distort or exaggerate the data,
• make the data and statistics coherent, and
• map or describe complicated scenarios clearly.

3
General Presentation Guidelines

• Know your material practice your presentation
  to increase your confidence.
• Include citations where appropriate (this aids
  credibility).
• Use complete sentences and avoid abbreviations.
  Abbreviations lead to forgetfulness and possibly
  a loss of confidence during the presentation.
• Induce the audience to think about your material
  not your PowerPoint skills.
• Promote your topic information that you do not
  cover is NOT your presentation so why dwell on
  it?
• Use a spelling checker and still proofread!

4
My Research Topic

• My proposed research will increase pubic
  awareness of a grave issue.
• My research will cover a lot of areas
• The issue and its consequences
• The people and there problems
• Cures

5
The Use of Outlines

• An outline by itself is not a presentation
  outlines do not show the relationships,
  assumptions or the analytic structure of the
  reasoning between the outline items.
• Do not compromise the amount of information that
  is needed to promote your topic.
• The hierarchical outline structure of PowerPoint
  unfortunately promotes maintaining an outline
  structure and the use of bullets lists which are
  a poor information transfer mechanism.
• The notion that outlines are easier to follow
  becomes nonsensical after several slides of
  unrelated bullets lists with ensuing levels of
  indentation.

6
Corporate Plan for Fiscal 2007

• Increase market share by 25.
• Increase profit by 30.
• Increase new-product introductions to ten a year.

Tufte, The Cognitive Style of PowerPoint, 2004,
p. 6.
7
Graphical Notions

• Graphics are not just for presentations. Seeing
  the problem space or data set often aids our
  understanding of the material and encourages new
  ways to view the problem and/or data. This is a
  tool!
• One glance, a lot of information. Show a lot of
  data in a relatively small space to reveal
  relationships.
• There needs be no relationship between the
  density of information, especially graphic, and
  the presenters vocalizations, which only need to
  highlight important features.

8
Content-free abstract diagram
Source Harold Pollack
9
Figure 1
outcome
Admit to NICU
Offer NICU Services
?
Do Not Admit to NICU
outcome
Hospital A
No NICU
outcome
outcome
Admit to NICU
Offer NICU Services
?
Do Not Admit to NICU
outcome
Hospital B
Patient
outcome
No NICU
Admit to NICU
Offer NICU Services
outcome
?
Do Not Admit to NICU
Hospital C
outcome
No NICU
outcome
Mortality Model
Patient Flow Model
Entry Model
NICU Admission Model
Several levels of NICU offerings will be
measured. This Figure is simplified to ignore
transfers. The analysis will adjust for transfers.
Source Harold Pollack
10
Graphical Excellence

• Graphic presentation of data is used to promote
  the communication of complex quantitative,
  multivariate and/or relational ideas. Often,
  scenarios that are hard to explain verbally are
  easier to explain graphically.
• The need to disclose complete information and to
  provide correct source citations is required for
  both graphical and textual information.
• The best choice of a graphic to be used is one
  which provides the greatest amount of information
  at a glance.
• The graphic used should promote thinking about
  the information imparted and not the design or
  the technology used for the graphic.

11
Napoleons March to Moscow
Charles Minard, Carte Figurative des pertes
successives en hommes de lArmée Françasise dans
la campagne de Russie 1812-1813, 1869.
12
Data Mapping
Example of a Southside Cluster With no Major
Player Grocery Coverage
Dr. John Snow, Cholera Epidemic in London, 1854.
Source Mari Gallagher, Chain Reaction Income,
Race, and Access To Chicagos Major Player
Grocers, The Partnership for New Communities,
Chicago, 2005.
13
Graphing Data

• Data graphing is an important tool to guide our
  understanding of the data and allow us to view
  the data in different ways.
• By applying different graphing techniques to the
  same set of data and by asking ourselves
  different questions of the data we can, perhaps,
  uncover aspects of our research we had not
  anticipated.
• The same techniques can be used in a presentation
  to encourage understanding and constructive
  feed-back.

14
Graphic Misuse

• Graphical devices are very powerful tools and
  easy to misuse.
• The use of chartjunk and clipart, for example,
  plays at a basic psychological level to take our
  mind away from the actual data presented.
• We are somewhat on guard when looking at what we
  know or suspect to be an advertisement, but tend
  not to have the same defense mechanism in place
  in business, research or educational settings.
• The argument for a right cause is not a valid
  justification for misrepresenting the data.

15
Confusing Numerical Data with Graphic Area
Source Los Angeles Times, August 5, 1979, p. 3.
16
Proper Distortions or Adjustments

• To properly compare data sets when an important
  attribute is quite different in each set, an
  adjustment in the attribute is quite helpful.
• Age-adjustment facilitates the comparison of two
  populations with different age curves.
• Inflation-adjustments facilitate understanding an
  implicit or explicit monetary time line.
• Such adjustments can be applied over a wide
  variety of attributes, age, size, area, etc.
• Always adjust for inflation and think in terms of
  per-capita numbers.
• When an adjustment is used, say so!

17
Data Table
Age-Adjusted Male Mortality Rates by
Race/Ethnicity, 2001
Source National Center for Health Statistical
Reports, Vol. 52 No. 3, Sept. 18, 2003, table 1.
18
Age-Adjusted Mortality Rates by Race/Ethnicity
and Gender, 2001
Deaths per 100,000 Persons
Source National Center for Health Statistical
Reports, Vol. 52 No. 3, Sept. 18, 2003, tables 1
2.
19
Age-Adjusted Mortality Rates by Race/Ethnicity
and Gender, 2001
M
Deaths per 100,000 Persons
M
F
M
M
F
M
F
F
F
Source National Center for Health Statistical
Reports, Vol. 52 No. 3, Sept. 18, 2003, tables 1
2.
20
Time-Series Graphs
Source Harold Pollack
21
The essential question of quantitative thinking
is Compared to What?
Connecticut Traffic Deaths, Before (1955) and
After (1956) Stricter Enforcement by the
Police Against Cars Exceeding Speed
Limit Synthesized data used
Sure looks like a decrease to me!
320
310
300
290
280
1956
1955
Adapted from Donald T. Campbell an H. Laurence
Ross, The Connecticut Crackdown On Speeding
Time Series Data in Quasi-Experimental
Analysis, Tufte, ed., The Qualitative Analysis
of Social Problems, 1970, p. 113.
22
Expanded Year Comparison
Connecticut Traffic Deaths, 1951-1959 Synthesized
data used
Treatment
Source Ibid. p.115.
23
Finally, a Multivariate Comparison
Traffic Deaths per 100,000 Persons in
Connecticut, Massachusetts, Rhode Island, And New
York, 1951-1959 Synthesized data used
N.J.
N.Y.
CONN.
MASS.
R.I.
Source Ibid., p. 118.
24
Scatterplots

• Scatterplots are useful for examining data sets
  over two or more related variables and for
  examining different set of data over those
  variables. For example
• Data correlation
• Positive or negative relationships between
  variables
• Non-linear patterns
• Spread of data
• outliers

25
Scatterplots
Positive or direct relationship
Negative or inverse relationship
Examine the outlier
Very low or zero correlation
http//www.statcan.ca/english/edu/power/ch9/scatte
rgraphs/scatter.htm
26
Scatterplots
No relationship
Random
Source Ibid.
27
How Do We Value a Presentation?

• Presentations largely stand or fall depending on
  the quality, relevance and integrity of the
  content. Edward Tufte, The Cognitive Style of
  PowerPoint, 2003.
• The same best practices used to create a
  presentation can and should be used to critique
  and experience a presentation.
• A good presentation lives on after the
  presentation event. Unfortunately, a bad
  presentation lives on also.