Dear Terry,
Your latest post runs off in several directions. I do not have the time to explain point by point what is wrong. One issue alone will do.
You made the claim that it is “trivially easy to assess whether sales have increased by 20% per month for the book with the new cover.” Now, you ask for “evidence that there is some feature that makes the proposal impossible (as distinct from difficult in your eyes).”
There are three issues here.
The first issue is whether it is in theory possible “to assess whether sales have increased by 20% per month for the book with the new cover.” In my view, it is not. I have explained why. You disagree. To offer greater detail takes a paper, not a post. You have not shown that this is in theory possible — you merely claim it is with lots of talk about why you believe this to be so.
The second issue is the difference between trivially easy projects and difficult projects. Even if it were merely difficult “to assess whether sales have increased by 20% per month for the book with the new cover,” there is a significant difference between “trivially easy” assessment and difficult assessment.
Even if difficult assessment were possible, the third issue is cost. Complex processes such as the rework analysis of large infrastructure projects involve so much cost per project that they justify the cost of the analysis. And despite the fact that rework analysis is difficult and costly, it is possible to attribute COSTS on infrastructure engineering. This is not the same problem as assessing “whether sales have increased by 20% per month for the book with the new cover.”
If you cannot demonstrate whether the cover specifically accounts for an increase in sales as distinct from other factors, you cannot determine the amount. Regression analysis will not give this information for the reasons I explained before.
If this really is trivially easy, it’s up to you to demonstrate how to do this rather than simply repeating your assertion.
So far, you have only demonstrated the gaps in your knowledge of the publishing industry. You are making claims about a situation in the real world: the question of attributing an increase in sales to the book cover. You are making the claim that based on this attribution, it is possible to pay a designer or penalise the designer by withholding pay. Even without executive decision makers, the “designers” of a commercial book cover generally include art directors, designers, illustrators, and possibly typographers. Which one do you plan to penalise? And how do you plan to penalise them if they all work as salaried employees of the publisher? If you actually could demonstrate the exact amount of increase or decrease in books sales that each executive or employee or consultant contributes, do you plan to penalise or reward everyone in the process of book production and sales based on their specific contribution to profit, loss, or failure to realise expected profit?
Those of us who have actually been involved in commercial publishing or communication research argue that it is impossible to determine the increase in book sales that can be attributed to the cover as distinct from other factors.
There is one simple way to show that we’re wrong. Simply demonstrate precisely how it is “trivially easy to assess whether sales have increased by 20% per month for the book with the new cover.” If it is trivially easy, you should be able to show us rather than explaining that we really don’t understand your examples because we’re focused on the real world properties of two men and four digging a hole based on health and safety requirements or sheep jumping over a fence.
Everyone can solve trivial hypothetical problems, and no one confuses algebra problems with the real world properties of the creatures in the example. Your claims with respect to regression analysis and book covers operate at an order of difficulty far greater than these trivial examples.
But you are making a claim about the real world, and not a claim about hypotheticals. It is up to you to provide evidence for your claims.
Yours,
Ken
Ken Friedman, PhD, DSc (hc), FDRS | Chair Professor of Design Innovation Studies | College of Design and Innovation | Tongji University | Shanghai, China ||| University Distinguished Professor | Centre for Design Innovation | Swinburne University of Technology | Melbourne, Australia
-----------------------------------------------------------------
PhD-Design mailing list <[log in to unmask]>
Discussion of PhD studies and related research in Design
Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design
-----------------------------------------------------------------
|
