Dear all (following this…),
I'm stepping in after some time away. Likewise, I'm in Oslo with all the challenge of focus that implies today. So if I missed an aspect of the exchange, please be gracious in replying.
The question Jurgen asked (or was asked to address) is interesting. As Ken argues, it is potentially valuable. From my training, it is also nothing especially new — though the thematic itself is not something I've addressed or read about. That's familiar is the question, "how do a set of actors, in their activities, affect the conduct of others — or else the outcomes of a system" to put it one way. The "how" is a question of influence. It necessitates a theory of influence. That is, a statement about how actors, though their activities, might be able to affect the conduct of others.
On having a theory, it leads to questions of measures. "If this can happen in the way explained or proposed, is there any evidence that it has indeed done so." And finally, "how much" or "to what effect."
There is nothing that I have read here to suggest this is anything other than a very common kind of question one finds in empirical social research. Here are some examples of these types of questions I have indeed researched quite seriously:
1. How does the press (or media) influence foreign policy?
2. How does public opinion influence executive decisionmaking?
3. How does policy research affect policy conduct?
4. Does propaganda determine behavior?
5. What is it about composition that compels?
And there are limitless examples from other fields. But they share in common the question of influence by one set of actors on others given sets of constraints (legal, policy, institutional, etc).
Answering such question necessitates being very, very clear about what exactly the question is. Graduate students tend to start with vague questions that seem important. And they tend to be nervous about focusing too much because it seems to leave the "big ideas" behind. In fact, its the opposite. A dissertation advisor (worth his or her salt) will explain that clarity in questions makes possible precision in answers.
We are facing, here, rudimentary (which does not mean easy!) questions about research design. That is, "what answers that?"
If design — as an academic discipline, rather than a social practice — is not willing to engage in the rigors of social research necessary to answer its own questions, then it will be marginalized from the Academy. It is currently on the cusp of being received into it. When I read Victor's appeal to engage with literatures a few weeks ago, I heard a more fundamental appeal for design scholars to attend more systematically to the practices of the Academy in informing their debates on the basis of existing knowledge hard-won through lifetimes of scholastic work.
I'm not sure answering a question like Jurgen proposed would win anyone a Nobel, but it might earn someone a doctorate by making a new contribution to human knowledge. And that is no small thing. No small thing to do, and no small thing to achieve. And it remains a very worthy goal.
Derek
_________________
Dr. Derek B. Miller
Director
The Policy Lab
321 Columbus Ave.
Seventh Floor of the Electric Carriage House
Boston, MA 02116
United States of America
Phone
+1 617 440 4409
Twitter
@Policylabtweets
Web
www.thepolicylab.org (http://www.thepolicylab.org)
This e-mail includes proprietary and confidential information belonging to The Policy Lab, Ltd. All rights reserved.
On Monday, July 25, 2011 at 3:36 PM, Ken Friedman wrote:
> Dear Terry,
>
> As I noted in an earlier post, I can’t disagree.
>
> As I wrote to Jurgen, the evidence to date is only inferential.
>
> Now, it seemed to me that one could make a reasonable inferential
> statement based on the approach I suggested. I cannot argue this is a
> strong method, and I accept all of the issues you raise to the contrary.
> Even so, a partial or reasonable guess is some kind of approximation,
> and I’d propose this is better than nothing.
>
> To the contrary, one may argue that it is worse than nothing because of
> the possibility of a thorough refutation.
>
> We do have a reasonable basis for inference in stock market performance
> or sales figures of companies that use an integrated design process, but
> even these are subject to the arguments you raise.
>
> The difficult with nearly all micro-economic processes is that we
> cannot isolate them. Given the difficulty of isolating variables, it is
> equally difficult to demonstrate that design rather than any other
> factor makes the decisive difference. For that matter, the linkages
> between variables mean that any of the many factors you raise could
> derail the positive economic contribution of design.
>
> The question is therefore this:
>
> What sorts of approximations or measures can we fruitfully use to
> determine the economic contribution of design to products and services
> at the firm level of individual products and services?
>
> This is genuinely problematic. Your objections are reasonable, but
> until we can conduct firm level case studies of the kind you propose, I
> can’t see a better way forward than approximation and reasoned
> inference.
>
> Despite this, I’m going to paste your full critique below on two
> principles. The first is that you are right. The second is that these
> objections may help us to identify ways to overcome the barriers to
> solving this problem.
>
> Yours,
>
> Ken
>
> Professor Ken Friedman, PhD, DSc (hc), FDRS | University Distinguished
> Professor | Dean, Faculty of Design | Swinburne University of Technology
> | Melbourne, Australia
>
>
> Terry Love wrote:
>
> —snip—
>
> My experience is that things are not as simple as you suggest in
> relation to distilling the value of design (below). What you described
> was one of the pathways that I explored 5 years ago researching the
> value of design. My experience was that the approach doesn’t work for
> multiple reasons. When value is accrued it’s not possible to easily
> identify whether the increase in value was due to:
>
> a) Management deciding to commit to a new product (i.e. outcomes are
> primarily due to management decisions rather than the detail of design
> decisions)
>
> b) New technology being available that enables new features to be added
> (i.e. the increase in value was primarily due to engineering)
>
> c) Change in attractiveness and sales increases are primarily due to
> success of previous product (s) and evolution of consumer sentiment
> (which is not necessarily itself primarily due to design activity)
>
> d) Improvements in firms efforts at consumer retention (viral marketing
> etc. leading to improved sales outcomes independent of product design)
>
> e) Good marketing research identifying features that customers require
> (i.e. most of the conceptual design decisions are done by marketing
> researchers rather than designers)
>
> f) Good advertising
>
> g) Improved sales channels (more and better retail outlets increase
> sales independently of quality of product design)
>
> h) Improved manufacturing enabling lower pricing (increased sales can
> be primarily due to better price/value relationship)
>
> i) Improved supply channels (sales of many electronics devices are
> limited by distribution and manufacturing - e.g. Kindle and cars)
>
> j) Decisions by management about the structure of firms’ innovation
> processes (think Nokia - firms that have manger KPIs that insist on a
> flow rate of new products can result of a flow of less attractive of
> failed products, independent of design activity - hence value is
> primarily echoing organisaitonal decisions rather than design)
>
> k) Access to sound information and research data (quality of products
> and sales volume can be primarily due to better access to data rather
> than design activity)
>
> l) Type of product stream. Product streams and platforms are more or
> less sensitive to good design. For example, sales rates of extreme
> outdoor clothing are relatively independent of products being
> competently designed to satisfy the needs of extreme outdoor conditions.
>
>
> m) Type of market. Some markets respond more or less to improved design
> outcomes, and consumers’ responses are typically variable, over short
> and long time frames. Changes in the value of a product (or implied
> value of design) can be more a matter of incidental market changes than
> design
> activity.
>
> n) Errors, luck and increased value outcomes that happen in spite of or
> accidentally alongside design activity.
>
> The latter is interesting in terms of one defining characteristic of
> design activity, that has been poorly addressed in the design research
> and design literature - the ability to predict the behaviour of
> outcomes.
>
> Put simply, if designers do not have a justifiable process to predict
> the behaviour of outcomes (including changes in sales, changes in the
> world, product behaviour, changes in consumer behaviours, future changes
> in society etc) then it is hard to justify that it is designers and
> design activity that have achieved increases in value.
>
> All of the above issues are found even in apparently ‘simple’
> situations such as a product relaunch. They confound reasoning and
> analysis about the economic contribution of design based on publicly
> reported sales data. Put simply, I suggest that unless in single
> specific cases, the process and activity events, values and outcomes
> were drilled down and explained at a very detailed level, it would be
> possible in each case to come to an equivalently sound argument that any
> value improvement was due to other factors than design activity.
>
> Perhaps someone would put up a case to test?
>
> —snip—
|