I'm actually having a busy day in the office, but working with this data is a nice distraction from working with other data.

I hope he doesn't mind but I just did a quick comparison of our scoring approaches and just in case you were interested here are the distributions.

As Solivagant stated the data shows he is more likley to start from a default average score.

It looks like Solivagant is more likley to award top marks than I am (or as is much more likely, he has just been to many more World Class places than me.)

Likewise I'm much more likely to award lower marks, again this could be a result of Solivagnat having a middling default score, but me starting from 0 and working up, or that may just be that a higher percentage of my site visits are made up of dross on shorter trips from Western Europe?

I used 3 stars for an "average" WHS

I have not rated all my visited WHS yet, find it very hard to do for sites that I visited long ago. I have kept my own rating (1-10) since 2011, so am comfortable on those.

Not at all - it is interesting to compare the results of different approaches by different people both of whom were "trying" to be "objective"! I think it is worth our while seeing these different approaches and results before confirming all our own assignments - we may well decide to stick with our own assessments (and have the right to) but at least we will have had the chance to "calibrate"!
As you say - there may be reasons for different extreme numbers which are related to number/continent of WHS seen by each of us. Regarding my very high median figure - that is likely, at least in part, to be because my assignments are still "Work In progress" - I set up the M.O yesterday and and have done a single trawl through together with a few targeted comparisons. I would expect a number of the 2.5 ratings to "migrate" before they are finalised (both up and down). I am hoping to go through "subjects" like e.g Wooden Churches, Castles, Cathedrals, Colonial Towns, European Cities, Palaces, Gardens, African Cultural Landscapes etc in order to see if a bit more differentiation is justified. But I suspect that I will never totally escape the result of deciding that an "average" WHS deserves an "average" rating.,


And you don't feel the need to change those "initial assessments" from a relatively long time ago in the light of your subsequent travels and "learning"??

No, maybe because it does not feel like a long time ago. And when I decide on a rating, I always look at the others that got the same rating before and see "if it feels right" / is in about the same league as others with the same rate (using 1-10, with 0.5's too, of course is much more specific than the 5 stars).

As he has a couple of reviews up at the moment I wanted a quick hat tip to YURI SAMOZVANOV, who has come up with a very neat way of rating sites based on three aspects:

UN ique
ES sential
CO st-effective

Chapeau, I wish I had thought of the UNESCO scoring system.

Just posted my ratings. I hope it worked, because "your rating" still shows the "community rating".
I'm quite satified with the distribution of my score. My average score is 2.75 (based on 223 votes, I did not rate 4 of my visited sites). Roughly estimated my ratings are a bit lower than the average ratings for most of the sites. The largest discrepancies to the community ratings seems to be for Gothic Cathedrals (lower than the community) and Modern Architecture (higher than the community).

Nationalism bias: my average rating for the German WHS is 2.66 (37 visited sites, ranging from 1 to 4.5 stars) and for Austria is 2.77.


Also support for Greenwich (4 stars)


I've changed my mind on this point, voted for Graz (slightly higher than the community) and Vienna (slightly lower).

I think a bug appeared on my ratings, because when i click now on each WHS, the number of stars of my rating and the average rating are exactly the same.

Yes, this is a bug that has now been reported several times. Will be solved in the next release within a few days. But your actual scores have been saved so no worries.

To test the bug fix yourself, you can go to the UAT site (disabled). It also contains the bulk rating feature as part of the user checklist.

Looking at graphs, I think I will program a separate page to show the ratings in a table form (site, country, user rating, site rating) with the distribution shown as graph.

Interesting indeed, meltwaterfalls. Since I'm terrible with numbers, is it too much work for you to do the same graph concept to compare my ratings with yours. Just curious!

Ups...i didn't read read well the previous comments, sorry

Will be happy to do it, though I would need some easy way to get all of your ratings or a summary of them (Solivagant published his breakdown higher up in the thread).

If we can get that somehow I would be delighted to have a look at them.

One thing I am surprised with mine, is how even the distribution is, it is entirely by accident rather than design.

meltwaterfalls
Normal distribution is quite normal ;) As stated I plan to provide a separate page for your ratings with the distribution. With a bit of skill I may even allow you to select two users to compare to.

Thinking about it it's two pages:
* Site: Show user distribution for a site.
* Users: Show user ratings for all sites. Optional parameter to pick a reference set (all users, specific user).

Let's see if i can do it in a day. Already found a javascript bib.

Very true, I'm very happy to see I'm normal though :)

I would be very interested in having that analysis available, but to be honest I would be delighted with any data I can dive into and make pretty pictures with.

And distribution for a country?? You mentioned that above and it would seem to be of interest and uses similar logic and formatting. Also Natural and Cultural and Mixed? And "Category" would show the most/least appreciated "types" of sites! But "no rush" wedonot even have many ratings yet!!