Property talk:Has condition quality
Contents
Meaning of numerical ratings[edit]
When users create condition reports, one of the most important pieces of information is how "good" the conditions are. Sites like Candition simply have a numerical rating between 1 and 5. At RopeWiki, we would like to also include an adjective or phrase to elaborate on what the numbers mean. Ideally, we would follow some existing standard like whatever is de-facto in place at Candition. But we may not want to actually do that because they might be rating something different than we want to rate and/or we might want the ratings to be on a different scale. This is the breakdown of Candition ratings reported:
So, there are two questions we must answer:
1: What should quality ratings rate?[edit]
The are two main answers to this question.
A user's overall experience in the canyon[edit]
I think Candition contributors are using this rating paradigm since there are many more 4 and 5 ratings than the lower ratings. Advantages:
- Our ratings can be interchangeable with Candition's ratings -- users of both sites won't have to figure out a different rating system for each site
- First-time visitors can provide a useful report (this wouldn't be possible with the second option since a first-time visitor has no idea what "typical" is for a given canyon)
- No need for people to figure out what "typical" means for any given canyon (typical for the time of year, typical for the time that people usually descend the canyon, typical for optimal conditions, etc)
Condition of the canyon relative to "typical"[edit]
Advantages:
- The condition of a canyon is separable from its underlying quality
- This makes condition quality a search term for finding canyons that are unusually good, rather than canyons are are just good period (abnormally good conditions would pop out over standard conditions in Eaton, even though the latter may be overall more enjoyable). This is probably more useful for locals and less useful for visitors.
2: Which adjectives should we assign to the numerical ratings?[edit]
The answer to this question is, of course, highly dependent on the answer to the first question. Here are some other suggestions that might help you make your suggestion:
1 | 2 | 3 | 4 | 5 |
---|---|---|---|---|
terrible | poor | typical | good | great |
boring | fine | average | great | exceptional |
terrible | poor | normal | good | excellent |
poor | ok | average | better then normal | best conditions possible |
not worth doing | ok | normal | unusually good | exceptional |
not worth doing | ok | worthwhile | great | among the best |
Responses[edit]
Edit this section to add your response, and add four tildes (~) at the end of your comment to sign it.
I think that, between letting first-time visitors adequately rate its current quality and compatibility with Candition, rating the overall experience is probably the better choice. With that, I recommend the adjectives poor/ok/good/great/exceptional. I interpret Candition's adjectives to be something like horrible/poor/ok/good/great, and we could map the numbers accordingly when linking to Candition reports from RopeWiki. --Bjp (talk) 19:07, 23 June 2014 (PDT)
- I am okay with the above (Bjp 201406231907) except for the following: can you think of examples of 'exceptional' cases? Given the definitions of the words 'poor' and 'exceptional' one would expect the above to result in a rather left skewed distribution. Would that be the intent? —Bahman (talk) 09:50, 24 June 2014 (PDT)
- Exceptional case for LSA. I'd expect the distribution to be identical to the Candition distribution above, except shifted left by one number (with 1 & 2 merged together), and "5" column split into two columns because I expect the 5's on Candition to be a combination of "great" and "exceptional". So, I expect the distribution to be right-skewed, but (hopefully) not as much as Candition. There should be fewer 5's than 4's in my proposed system. --Bjp (talk) 15:15, 24 June 2014 (PDT)