Sandbox: Difference between revisions

From ChanceWiki
Jump to navigation Jump to search
Line 4: Line 4:
by Jacob Harris, ''Source'' (opennews.org) 22 May 2014
by Jacob Harris, ''Source'' (opennews.org) 22 May 2014


By Jacob Hrris
Harris identifies 6 ways to make mistakes in reporting data
a
*Sloppy proxieson
Much attention is paid to the Commerce Departments monthly report of the number of jobs added or lost by the US economy.  But only rarely do reports in the popular press highlight the fact
*Dichotomizing
that these figures are based on survey data, and hence subject to sampling variability. 
*Correlation does not equal causation
*Ecological inference
*Geocoding
*Data naivete


Irwin's post here includes an animated simulation (screen capture show below) that indicates what 12 months of sample data might look like based on various assumptions about what the economy is really doing.
His principle example is a story that was widely covered in social media.  It concerned but the biggest is related to "geocoding"


<blockquote>
What happened here was that a large percentage of IP addresses could not be resolved to an address any more specific than “USA.” When that address was geocoded, it returned a point in the centroid of the continental United States, which placed it in the state of—you guessed it—Kansas!
</blockquote>
<center>[[File:JobReportSim.png | 500px]]<center>
<center>[[File:JobReportSim.png | 500px]]<center>


Submitted by Paul Alper
Submitted by Paul Alper

Revision as of 18:44, 2 June 2014

Politics and porn (What's the matter with Kansas?)

Distrust your data
by Jacob Harris, Source (opennews.org) 22 May 2014

Harris identifies 6 ways to make mistakes in reporting data

  • Sloppy proxieson
  • Dichotomizing
  • Correlation does not equal causation
  • Ecological inference
  • Geocoding
  • Data naivete

His principle example is a story that was widely covered in social media. It concerned but the biggest is related to "geocoding"

What happened here was that a large percentage of IP addresses could not be resolved to an address any more specific than “USA.” When that address was geocoded, it returned a point in the centroid of the continental United States, which placed it in the state of—you guessed it—Kansas!

JobReportSim.png

Submitted by Paul Alper