ASA Connect

I would suggest the usual suspects (in alphanetical order):
Bertin, J. (1973). Semiologie Graphique. The Hague: Mouton-Gautier. 2nd Ed. (English translation done by William Berg & Howard Wainer and published as Semiology of Graphics, Madison, Wisconsin: University of Wisconsin Press, 1983.)
Bertin, J. (1977). La Graphique et Le Traitement Graphique de L'Information. Paris: Flammarion. (English translation done by William Berg, Paul Scott & Howard Wainer and published as . Graphics and the Graphical Analysis of Data. Berlin: De Gruyter, 1980. )
Cleveland, W. S. (1994). The elements of graphing data. Summit, NJ: Hobart Press.
Cleveland, W. S. (1994). Visualizing data. Summit, NJ: Hobart Press.
Tufte, E. R. (1983). The visual display of quantitative information. Cheshire, CT: Graphics Press.
Tufte, E. R. (1990). Envisioning information. Cheshire, CT: Graphics Press.
Tufte, E. R. (1996). Visual explanations. Cheshire, CT: Graphics Press .
Wainer, H. (1984). How to display data badly. The American Statistician, 38, 137-147.
Wainer, H. (1997). Visual Revelations: Graphical Tales of Fate and Deception from Napoleon Bonaparte to Ross Perot. New York: Copernicus Books, (reprinted in 2000, Hillsdale, NJ: Lawrence Erlbaum Associates).
Wainer, H. (2005). Graphic Discovery: A Trout in the Milk and Other Visual Adventures. Princeton, N.J.: Princeton University Press.
Wainer, H. (2009). Picturing the Uncertain World: How to Understand, Communicate and Control Uncertainty through Graphical Display. Princeton, NJ: Princeton University Press.
Wilkinson, L. (2005). The grammar of graphics. (2nd edition) New York: Springer-Verlag.

H

Howard Wainer
Distinguished Research Scientist
National Board of Medical Examiners
3750 Market Street
Philadelphia, PA 19104
Voice - 215-590-9861
email - hwainer@nbme.org

This email message and any attachments may contain privileged and/or confidential business information and are for the sole use of the intended recipient(s). Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please notify the sender immediately by reply email and destroy all copies of the original message and any attachments.


Original Message------

I am interested in finding authoritative treatments of graphic representation of data.  I am particularly interested in sources that handle scaling related misrepresentation of data to exaggerate or embellish perspectives.   Does the ASA have any position papers on this topic?   I would be interested in hearing any opinions that you may have but would appreciate your sharing your favorite treatments of the subject if you have any.

 

Thanks!

 

John Southard

Research Scientist

Advanced Energy Technology

 

GrafTech International Holdings Inc.

982 Keynote Circle

Brooklyn Heights, OH 44131

Office: (216) 676-2113

Cell: (216) 408-7463

http://www.graftech.com

 

 

 

 

Hardly authoritative, but there are some examples on my gallery of data
visualization, http://datavis.ca/gallery/

-Michael

--
Michael Friendly Email: friendly AT yorku DOT ca
Professor, Psychology Dept. & Chair, Quantitative Methods
York University Voice: 416 736-2100 x66249 Fax: 416 736-5814
4700 Keele Street Web:http://www.datavis.ca
Toronto, ONT M3J 1P3 CANADA


Original Message------

I am interested in finding authoritative treatments of graphic representation of data.  I am particularly interested in sources that handle scaling related misrepresentation of data to exaggerate or embellish perspectives.   Does the ASA have any position papers on this topic?   I would be interested in hearing any opinions that you may have but would appreciate your sharing your favorite treatments of the subject if you have any.

 

Thanks!

 

John Southard

Research Scientist

Advanced Energy Technology

 

GrafTech International Holdings Inc.

982 Keynote Circle

Brooklyn Heights, OH 44131

Office: (216) 676-2113

Cell: (216) 408-7463

http://www.graftech.com

 

 

 

 

I would add Naomi Robbins, Creating More Effective Graphs and any of the Stephen Few books.

------------------------------
Georgette Asherman

Original Message:
Sent: 11-18-2016 08:23
From: Michael Friendly
Subject: Best practices - Scaling and Graphing

Hardly authoritative, but there are some examples on my gallery of data
visualization, http://datavis.ca/gallery/

-Michael

--
Michael Friendly Email: friendly AT yorku DOT ca
Professor, Psychology Dept. & Chair, Quantitative Methods
York University Voice: 416 736-2100 x66249 Fax: 416 736-5814
4700 Keele Street Web:http://www.datavis.ca
Toronto, ONT M3J 1P3 CANADA

Thank you all for your responses.    I will see what I can turn up using the suggestions that have provided.   I see a lot of interesting content here.  

My specific concerns are when we scale y-axis data in a way that shows that change is occurring relative to the x axis. It may help to visually indicate that change is occurring but if we select endpoints to bracket the change itself too closely the nature of the delta can become very much exaggerated. To me this seems inappropriate enough when the scale of the axis is included, but you will sometime see that there is no scale and you are only given a trend.

That we can measure a change and display a change at some scale does not necessarily imply the inherent relevance of the change particularly.  I think this is the graphical equivalent of the notion that a variable may be statistically significant but not necessarily practically significant for some given context.   Taking advantage of scale is just one way that such interpretation of change can be used to create misleading representations of reality.

John Southard  

------------------------------
John Southard
Research Scientist
GrafTech International

Original Message:
Sent: 11-21-2016 06:57
From: Georgette Asherman
Subject: Best practices - Scaling and Graphing

I would add Naomi Robbins, Creating More Effective Graphs and any of the Stephen Few books.

------------------------------
Georgette Asherman

Original Message:
Sent: 11-18-2016 08:23
From: Michael Friendly
Subject: Best practices - Scaling and Graphing

Hardly authoritative, but there are some examples on my gallery of data
visualization, http://datavis.ca/gallery/

-Michael

--
Michael Friendly Email: friendly AT yorku DOT ca
Professor, Psychology Dept. & Chair, Quantitative Methods
York University Voice: 416 736-2100 x66249 Fax: 416 736-5814
4700 Keele Street Web:http://www.datavis.ca
Toronto, ONT M3J 1P3 CANADA

Let me offer an opinion on your problem:
1. The ratio between the scales of the X and Y axes is arbitrary, and so you can make things look any way you want. Expand Y and a steep change can look like nothing. Shrink Y and a minor change can look momentous.
I discuss this in Chapter 1 of my book Visual Revelations. Note that Playfair did this in his skyrocketing national debt plot.

2. A fact -- if you make the Y scale too large you lose the ability to see things that are happening. If you make it too small the user can regraph in another scale if they wish. Thus making the latter error a venial sin whereas the former can be mortal.
3. Bill Cleveland proposed a sensible rule that he noted had been followed by masters of graphics in the past; make the ratio of the scales (the aspect ratio) such that the data follow a 45 degree angle. My exerience suggests that this is wise advice indeed.

H

Howard Wainer
Distinguished Research Scientist
National Board of Medical Examiners
3750 Market Street
Philadelphia, PA 19104
Voice - 215-590-9861
email - hwainer@nbme.org

This email message and any attachments may contain privileged and/or confidential business information and are for the sole use of the intended recipient(s). Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please notify the sender immediately by reply email and destroy all copies of the original message and any attachments.


Original Message------

Thank you all for your responses.    I will see what I can turn up using the suggestions that have provided.   I see a lot of interesting content here.  

My specific concerns are when we scale y-axis data in a way that shows that change is occurring relative to the x axis. It may help to visually indicate that change is occurring but if we select endpoints to bracket the change itself too closely the nature of the delta can become very much exaggerated. To me this seems inappropriate enough when the scale of the axis is included, but you will sometime see that there is no scale and you are only given a trend.

That we can measure a change and display a change at some scale does not necessarily imply the inherent relevance of the change particularly.  I think this is the graphical equivalent of the notion that a variable may be statistically significant but not necessarily practically significant for some given context.   Taking advantage of scale is just one way that such interpretation of change can be used to create misleading representations of reality.

John Southard  

------------------------------
John Southard
Research Scientist
GrafTech International
------------------------------

Original Message------

Thank you all for your responses.    I will see what I can turn up using the suggestions that have provided.   I see a lot of interesting content here.  

My specific concerns are when we scale y-axis data in a way that shows that change is occurring relative to the x axis. It may help to visually indicate that change is occurring but if we select endpoints to bracket the change itself too closely the nature of the delta can become very much exaggerated. To me this seems inappropriate enough when the scale of the axis is included, but you will sometime see that there is no scale and you are only given a trend.

That we can measure a change and display a change at some scale does not necessarily imply the inherent relevance of the change particularly.  I think this is the graphical equivalent of the notion that a variable may be statistically significant but not necessarily practically significant for some given context.   Taking advantage of scale is just one way that such interpretation of change can be used to create misleading representations of reality.

John Southard  

------------------------------
John Southard
Research Scientist
GrafTech International
------------------------------