16 years
I believe your math is right. and it seems a shame that a good hostel can suffer so much from one bad review.
I agree that the ratings are good for a reference and hopefully, thoughtful people look more closely at the comments as i think that is more telling than the actual ratings.
For instance, one of our hostels is known as a 'party' hostel. if you're looking for a place to hang out and have a late night, you'll be really happy there. if you are looking for a smaller, quiet hostel we're not your best choice. So someone who wants the second type can give a bad rating that someone who wants the first type would give a great rating.
So i think in the end, the comments, taken with a grain of salt of course, are the best guide.
I also like that hostelworld has added the 'type' and 'country' of traveller. I think there is a big difference in rating based on being a 'novice' or an 'experienced' traveller. And from the management side, knowing what country someone is from is helpful too. That way we can see if we are having a pattern of problems with people from certain countries. If so, we can then work on finding out what isn't working and try to fix it. Sometimes it's as simple as translating something better or having more people on staff who speak a certain language.
Mathematically the disadvantage of a percentage presentation of data is always affected by extreme numbers. I still do not find percentages a good way of measurement but...you know how information moves around these days...it HAS to be fast and visually attracttive.
Al this comes with our territory, unfortunately.
Nice comment on the yes or no.
Log in to join discussion