Nip Impressions logo
Sat, Oct 22, 2016 19:30
Click here for Pulp & Paper Radio International
Subscription Central
Items just for you
New publication added! Advertising Arguments 2015 book
My Profile
Management Side
Technical Side

Responses to last week's column

Last week we asked,

When looking at others' charts and graphs, do you ever feel the whole story is not being told?

90% of you said YES!

Then we asked, "What do you do to satisfy yourself that data presented to you is presented honestly and accurately?"

>I remember when I was a Co-op student 35 years ago that my plant manager told me to always remember, "Liars figure and figures lie!

>some thinking

>a) that there is actually a cause/effect relationship going on b) a statistical analysis will validate that the relationship is more than just experimental error or biased sampling (Excel allows that to be easily displayed, too) c) the data is worth drawing a bunch of graphs when one well-done one is enough.

>I ask direct questions about what is not shown.

>Ask detailed questions with follow-up questions as needed - has led to alot of embarrasing moments for general line management consultants, e.g. McKinsey, Accenture, etc. advising paper companies. The old Texan expression "All hat and no cattle" comes to mind.

>Pull the raw data, if available and analyze it myself.

>Check scale and ask about what is left out.

This week's quiz is here.


Want to get a jump on the surveys? Follow me on twitter, where they are posted early. You can do this here. (note: all respondents are confidential; the software is programmed in such a way that neither we nor anyone else can determine who responded)


Want to be heard on other subjects? Be sure to watch for "Paperitalo Second Tuesday Surveys."


Printer-friendly format

Powered by Bondware
News Publishing Software

The browser you are using is outdated!

You may not be getting all you can out of your browsing experience
and may be open to security risks!

Consider upgrading to the latest version of your browser or choose on below: