Jump to content


Please note: You can easily log in to MPN using your Facebook account!

Decision Rules/Heuristics for Buying Keyboards


Moonglow

Recommended Posts

Hi All,

 

I teach a course entitled The Psychology of Consumer Behavior at Purdue University (Calumet Campus), and I thought Id share this consumer decision model with anyone considering the purchase of a keyboard, or otherwise interested in this sort of thing.

 

First of all, let me provide a little background. Ive been in the market for a new digital piano/performance controller for a while now, and Im having trouble deciding between several different brands/models. Consequently, I thought I would utilize a consumer decision heuristic to assist in the selection process. The model presented below illustrates the Compensatory Rule and falls within the larger category of multiattribute attitude models. It is called the compensatory rule because a relatively high rating on one attribute has the effect of compensating for a low rating on another attribute. While the model is cumbersome and requires some effort, it is favored by many in the consumer behavior research field in determining the best overall brand/model. The procedure involves listing the brands/models in question, calculating the product of the importance weight and attribute rating for each brand/model, then summing these values over all the attributes to arrive at a total score for each brand/model. When constructing the table it is important that the distribution of importance weights among the attributes total 100. Consider the following:

Attribute    Importance S90ES PC-2X  Nord Stage 
MIDI Control     25       8    10        5
Piano            20       9     8        7
Organ            15       3     6       10
Rhodes           15       9    10       10
Synth            10       8     5        6
Keybed Action    10       8     7        6
Weight/Size       5       6     7       10 

S90ES = 200+180+45+135+80+80+30 = 750

PC-2X = 250+160+90+150+50+70+35 = 805

Nord Stage = 125+140+150+150+60+60+50 = 735

 

So according to the compensatory rule, I should select the PC-2X!

 

Please note that other attributes can be included (e.g., price), but again, make sure the distribution of importance weights among the attributes total 100. Also, this post is NOT intended to stimulate debate among the synths.....Im just presenting the model/procedure. Besides, attributes and their rankings are unique to the individual consumer!

 

Admittedly, the compensatory rule may be viewed as overly analytical, clinical, sterile, etc. I realize that many product selections are made based in part on a vibe factor, and I am not suggesting the model replace this element.....although this certainly could be included as a ratable product attribute! But I have found this procedure fun to screw around with as a means of more systematically evaluating factors important to a purchase decision.....or if only as confirmation of decisions arrived at on a more visceral level. Enjoy!

"We don't stop playing because we grow old; we grow old because we stop playing."

- George Bernard Shaw

 

Link to comment
Share on other sites



  • Replies 6
  • Created
  • Last Reply

Mr. Glow,

As a mechancial engineer, I applaud any effort by humanity to quantify the abstract. I am tempted to share my mathematical treatment of wine quality/price ratio, but I don't want to be kicked off the Forum.

 

I like the model in that it can be personalized by each consumer through the relative weight of each atribute, and the score each choice receives (although this does inject subjectivity into the model, which is, sadly, unavoidable). Good luck with your purchase and thanks for sharing. :)

 

Regards,

Joe

Link to comment
Share on other sites

Let's see... I like like your sound and don't like mine. Therefore I should not choose the keyboard in front of me. But, if I get a keyboard like yours I will sound like you, therefore I should not choose the keyboard in front of you. Yet, immitation is the sincerest form of flattery but I am not that sincere, so I should not choose the keyboard in front of me...

I think I'll go in against a Sicilian when death is on the line. Play what you like.

Link to comment
Share on other sites

Originally posted by Moonglow:

Attribute    Importance S90ES PC-2X  Nord Stage 
MIDI Control     25       8    10        5
Piano            20       9     8        7
Organ            15       3     6       10
Rhodes           15       9    10       10
Synth            10       8     5        6
Keybed Action    10       8     7        6
Weight/Size       5       6     7       10 

S90ES = 200+180+45+135+80+80+30 = 750

PC-2X = 250+160+90+150+50+70+35 = 805

Nord Stage = 125+140+150+150+60+60+50 = 735

Add the free editor to the Kurz and that Synth number will increase to at least a 9. ;)
Link to comment
Share on other sites

The nifty thing about weighted categories is you can jimmy the weights to get any answer you want! ;)

 

BTW, the weights don't have to add up to 100. The number just make a little more intuitive sense if you do -- for some people, anyway. You can just as easily use a scale from 1 to 10 for the weights, and that way when you add a new criteria you don't have to adjust all the other weights.

 

The biggest number wins in any case.

 

Charts like this generally don't help me much. I remember when I worked at Ford they used a rating system like this to choose the next processor for engine control. Someone must have leaked the list of characteristics to Motorola, who then designed a processor that seemed to have each of the desired features (until you actually tried to program it, and realized it was a facade). The other vendor had a well thought out product but didn't hit some of the categories, and speaking English was weighted 10 out of a possible 10, so Moto got 100 points and the Japanese company got zero, and Motorola won. But the resulting product was ill-suited for the task and both Moto and Ford lost in the long run. They delayed introducing it (due to problems) and only used it for a fraction of the original intended run. Hopefully Motorola learned their lesson there.

 

Um, weren't we talking about synthesizers or something?

 

Cheers

Jeff

Link to comment
Share on other sites

Originally posted by learjeff:

The nifty thing about weighted categories is you can jimmy the weights to get any answer you want! ;)

 

BTW, the weights don't have to add up to 100. The number just make a little more intuitive sense if you do -- for some people, anyway. You can just as easily use a scale from 1 to 10 for the weights, and that way when you add a new criteria you don't have to adjust all the other weights.......

The compensatory model was originally developed with the importance weights totaling 100. While a 1-10 scale certainly offers enhanced practical utility of not having to readjust the importance weights when a new attribute is added to the list, it runs the potential risk of a consumer rating each importance attribute unrealistically high (e.g., rating all attributes 9 or 10) thus compromising the psychometric integrity of the instrument. Im also more comfortable with the weights totaling 100 from a theoretical perspective, as this encourages the consumer to think of the weights in terms of a hierarchical distribution of properties within a conceptual product framework. Also, the model is not intended for one to manipulate the ratings to produce a desirable or predetermined outcome, but rather to assist the consumer in the decision making process. That being said, I agree that in and of itself the model is of limited utility, as most consumers use combinations of decision strategies (e.g., compensatory, lexicographic, conjunctive) which are often sequential in nature and occur under a wide variety of contextual circumstances. But I would argue that these types of models retain their power in that they provide the consumer with a structured framework to critically evaluate their decisions.....and if adjusting the attribute ratings allow one to arrive at an enhanced understanding as to how these variables impact his/her decision, the model will have served a useful purpose.

"We don't stop playing because we grow old; we grow old because we stop playing."

- George Bernard Shaw

 

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...