Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

Determining attributes for exploring public acceptance of a type of park and relative importances of attributes

I've started a project using conjoint analysis to investigate public acceptance of a type of wild-looking, ecologically-oriented urban parks with remnant brownfield qualities preserved (these parks developed on brownfields), and also, to investigate the relative importances of the features of this park type set by the public.

But I have a question in determining the attributes for constructing scenarios. I think attributes should be characteristics of this urban park type that the public deem important (I will investigate this through focus groups with participants. My hypotheses of these characteristics are: wild-looking appearance, remnant brownfield qualities, rich ecological value). Are these attributes appropriate for understanding public acceptance of this urban park type? If I want to know this acceptance situation, should attributes come from a set of parameters determining general park characteristics rather than specifc features of this urban park type? And to see what degree the public preference are consonant with features of the specific urban park type built on brownfields? I'm quite confused about this. Any help is appreciated! Thanks!
asked Nov 10, 2019 by Xinlei Hu (160 points)

1 Answer

0 votes

It really depends on what choice you are trying to model.  If it is the choice of this kind of park among all competing parks, then I think you might want the attributes and levels that cover all the relevant types of parks.  

If you really only care about modeling the importance of the features of this kind of urban parks, then you might want to limit your focus to the attributes and levels that describe this kind of urban park and to screen your respondents so they are people interested in this kind of urban park.  

It's possible you want to do both - to see how different urban parks built on brownfields compete with one another and with  different types of parks.  In this case you might want a model with park-type specific attributes and levels (i.e. alternative specific attributes and levels) or you may want only to vary the characteristics of this type of urban park and handle all the other kinds of parks with alternative specific constants.  

In short, there are a lot of ways to structure this kind of experiments, depending on what goals you have for the project.
answered Nov 10, 2019 by Keith Chrzan Platinum Sawtooth Software, Inc. (85,325 points)
Hi Keith!
Thank you very much for your quick response. It is very helpful. Regarding the sole aim of understanding the relative importance of the features of this novel park, I'm just wondering is the "feature" represent an attribute or a level? I find there are some obscure thoughts determining the attribute. For example, regarding this park type, generally, "wild-looking" "rich biodiversity" "less facilities" are distinct features, which might influence people's willingness to visit them. But "less facilities" seems like a "level" for constructing scenarios... I'm just wondering the "quality" constituting a specific "product," i.e., in this case, the novel park type, not a general "product," i.e., general parks, is an attribute or a level? Thanks!
Typically the feature would be the attribute and in your case the two levels might be "absent" and "present. "
Ok I see! Many thanks for your reply!