Terry,
There are three problems in your arguments about the book industry.
The first two problems are statistical. First, regression analysis requires that you can account for the variables. This is not practical for books. While it is possible to quantify some aspects of any one book, you can’t quantify them all, and you certainly cannot control enough variables to determine the specific contribution of a cover design. Even if you could, however, regression analysis shows correlation, not causality.
But there is a third problem in your argument. This problem is empirical, not statistical. No one person makes decisions for a book cover, and most of the people who make those decisions work for publishing firms as salaried employees. Many people make these decisions: editorial directors, editors, marketing directors, marketing managers, art directors, design directors, as well as external designers and illustrators. Designers and illustrators often work to a brief — they execute the instructions of the entire group of decision makers. All are involved in deciding on books covers. Only at a small press does one person alone decide on a cover.
As David Sless stated, you don’t seem to know much about book publishing. You have a small press where you make all the decisions and sell your books by mail order. People apparently order your books sight unseen. It is a long way from small press mail order publishing to the kinds of book publishing that Martin Salisbury and the others are talking about.
It’s not simply your statistics that are mistaken. It is your understanding of the publishing industry.
If you’d like to know how the publishing industry works, read Marshall Lee (2009). Lee describes the process by which editorial, marketing, and design departments work together to plan, design, and produce book covers and dust jackets.
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
—
Reference
Lee, Marshall. 2009. Bookmaking. Editing, Design, Production. Third Edition, Revised. New York: W. W. Norton.
--
> On 2015Apr11, at 19:07, Terence Love <[log in to unmask]> wrote:
—snip—
> Regression analysis?
>
> To say it again, and contrary to your post, it IS “trivially easy to assess whether sales have increased by 20% per month for the book with the new cover.”
>
> The statistical challenge is to assess how much of that change is due to the change of cover.
>
> Advertising design faces much the same problems in assessing the value of changes, and quantitative analytical methods are well established to do so. In fact, to make it simpler you might regard the book cover as advertising.
>
> I see no obvious reason why similar quantitative methods cannot be used in the case of graphic design to establish the value of the designers contribution? If there is a fundamental reason why such analyses cannot apply, please let me know.
—snip—
—
And Terry Love wrote:
—snip—
With current arrangements, the graphic designer bears little or no financial responsibility for the design outcomes which they alone make the decisions for. This is a situation of moral hazard and is a situation usually to be avoided.
—snip—
Either the designer's work contributes to good sale outcomes, in which case the designer can be held responsible, or the book cover design doesn’t significantly contribute to good sales outcomes in which case why employ a book cover designer?.
As I wrote to Ken, quantitative analytical methods are available that would appear to provide the basis for analysing the value of the contribution of visual designers to outcomes. These are used as a matter of course in advertising and print media analysis assessing the value of different aspects of print media communication.
—snip—
—
-----------------------------------------------------------------
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
-----------------------------------------------------------------
|