Re: Who is in favour of empirical research approach?

Thanks to everyone for some interesting stuff to consider over the  
weekend.

I've recently been running something that I guess you could call an "A/ 
B test" on two different ways of presenting the same data. The idea  
was to ask people to make decisions on the basis of the data presented.

My research seemed to show that people made better decisions with  
their preferred design. (happy to post a reference when I get it  
published).

As in much design, I suspect, there are rarely absolutes in  
information design. If a designer was to tell me that a certain design  
was "better", I'd be tempted to ask "better for what?". People are  
interesting things, and their reasons for preferring different things  
are complex.

I think one of the strengths of empirical research in design, when  
conducted properly, is to find out not only which of two alternatives  
people prefer, but to find which people are likely to prefer which,  
and perhaps even suggest why. Remember, empirical research includes  
qualitative, such as interviews and focus groups.

It might not be very useful to find which of two designs someone  
prefers, but finding which of two designs is generally preferred by,  
say, females in the 30 - 40 age group in socio-economic group 3, and  
we might be getting somewhere. Add in something like: which design do  
they make quicker decisions with (a quantitative metric), and we start  
to get somewhere useful.

The designer can really only offer their own views. And while their  
experience makes this valuable, research methods, used responsibly and  
correctly, can give them some very useful feedback from those that  
actually use those designs.

Cheers,

--Will

...............................................

Will Stahl-Timmins B.A., M.A.
PhD Researcher: Information Graphics in Health Technology Assessment.

T: +44 (0) 1392 726 081
M: +44 (0) 7941 865 196
E: [email protected]

www.pms.ac.uk/infographics/
www.pms.ac.uk/pentag/
www.willstahl.com

PenTAG (Peninsula Technology Assessment Group)
Peninsula College of Medicine and Dentistry
Veysey Building
Salmon Pool Lane
Exeter
EX2 4SG

On 13 Mar 2010, at 17:56, Gunnar Swanson wrote:

> I fall somewhere in the middle of this. My clients are often small  
> enough that the cost of finding out much about what would be  
> effective would exceed the value of knowing.
>
> My experience with "researchers" has not all been good. For example,  
> my students are currently working on a project with a "research"  
> project here at East Carolina University. I believe that I  
> understand the research in their field better than they do. They  
> hide behind being "scientific" even when they wander away from  
> anything backed by actual research. When pressed for reasons for  
> their decisions and demands they resort to "we like it" even though  
> they are working with my students because they say that research  
> shows that designers who are closer to the target audience are more  
> effective. They are among the least strategic people I have ever met.
>
> Most of the time when marketing people have told me "research says x  
> is better" I have discovered that they either misunderstand the  
> research or are speaking in third hand generalities learned in class  
> somewhere. It is easy to understand designers taking the position  
> that research is an impediment to good design.
>
> That said, I strongly believe that certain (not all) research can  
> inspire creative and effective design and move it forward. Even if  
> that were not true, the research faction has a more effective bit of  
> rhetoric than the traditional design faction so clients will believe  
> them, thus research is inevitable.
>
> If the research is controlled by designers, it will be inspirational  
> at best and harmless at worst. If it is controlled by others, it  
> will be a hindrance to design unless we educate researchers. One way  
> or another, designers need to participate in design-related  
> research, if only for self-preservation.
>
> As much as I am not a Jakob Nielsen fan, he is right about how much  
> information you can get out of very little testing. It is worthwhile  
> to see if people understand what we assume they understand. That  
> doesn't lead me to believe that A/B testing every design decision or  
> trying to find or follow a set of rules for design is a worthwhile  
> future.
>
> A lot of worthwhile empirical information can be cheap and casual,  
> comparable to making dummies and quick prototypes of various sorts.  
> I try to get my students to think in terms of what they could learn  
> that would be helpful as designers. I discourage them from asking  
> friends and roommates questions like "Which of these do you like?"  
> and instead try to get them to be at once more specific and more  
> general--"What do you think the coffee in each of these tastes  
> like?" or "Tell me what you feel about the company behind each of  
> these."
>
> Gunnar
>
> ----------
>
> Gunnar Swanson Design Office
>
>
> 1901 East 6th Street
> Greenville NC 27858
> USA
>
> [email protected]
> +1 252 258 7006
>
> all new website:
> http://www.gunnarswanson.com
>
> On Mar 12, 2010, at 1:57 PM, Jose Marconi Bezerra de Souza wrote:
>
>> Randal,
>>
>> I appreciate your sincerity.
>> I think that there are many others who think just like you.
>>
>> However, how do you test if your design ideas are really working?
>>
>> José.
>>
>> On 12 March 2010 15:22, Randal <[email protected]> wrote:
>> At 2:35 AM -0700 3/8/10, [email protected] wrote:
>> >Jose's title was not clearly formulated in line with his intention:
>> >"Who is in favour of empirical research approach?" I doubt  
>> somebody in
>> >this group could not be in favour.
>> >
>>
>> Personally, as a designer, I have never ever seen ANY empirical  
>> research results that didn't either:
>>
>> 1. Merely state the obvious.
>>
>> 2. Indicate preference for a design direction that any half decent  
>> designer could have directed you to in five minutes.
>>
>> 3. Create results that are essentially useless or at worst  
>> completely misleading.
>>
>> Empirical design at its best only prevents really bad and stupid  
>> design decisions at about ten times the cost of a good designer who  
>> could do exactly the same thing.
>>
>> -- Randal
>>
>
>
> [email protected]
> +1 252 258 7006
>
>
>
>
>
> <ATT00001..txt>

___________________________________________________________________

Use the following address to post a message to all subscribers: 
 [email protected]

To subscribe, unsubscribe or change your options, visit:
 http://list.InformationDesign.org/mailman/listinfo/infodesign-cafe

For all Information Design matters:
 http://InformationDesign.org

Problems? Write to:
 [email protected]
___________________________________________________________________