I guess it's something we've probably all thought about at some point - is the general quality of WH sites getting lower? Is the quality of the list getting diluted as the bar for inscription inevitably gets lower? What follows is a modest attempt to quantify these thoughts - does the data back up the claim?

To find out, I used the community ratings from our site. Because each year has such a variable number of sites inscribed, I summarised them into five year blocks (and a separate block for 1978/79). I also excluded sites with a 0 rating, since I don't think you can give 0 stars on here? Thus those sites haven't been visited and rated by a community member. And the results are honestly much as expected:

1978-1979 3.59
1980-1984 3.39
1985-1989 3.45
1990-1994 3.33
1995-1999 3.11
2000-2004 3.02
2005-2009 2.82
2010-2014 2.89
2015-2019 2.69
Overall Avg 3.13

As a side note, I've often thought that sites added this century are generally less interesting/deserving than sites added last century, so it's interesting to see that the data essentially bears this out: sites added last century are above the average, those inscribed this century are below the average. Obviously there are exceptions, but I've broken it out separately by year as well. Interesting to see that the lowest two rated years on record are 2018 and 2019 (the latter even in spite of Bagan rating in our top 10!), while the highest-rated year is the very first: 1978.

Here's the full data sheet for anyone interested:
https://docs.google.com/spreadsheets/d/1X25Gc6B3SUQlSaO17deUpiE0apty5vR9xMH26ocKIR0/edit?usp=sharing

joelonroad

I think the trend is visible even if you don't cluster in 5 year chunks. You can probably do a regression with inscription year and rating score and it will show a negative trend. Small caveat: Less rated sites get a larger rating reduction (uncertainty). Newer sites may have less visitors and less ratings as a consequence.

For me, it's also more a country issue. The first sites from a country tend to be better than the latter sites, simply because the truly great comes up first.

Other factors to note. Some countries
a. didn't get going at all until the scheme had been running for a few years - so their stars tend to appear in the middle
b, made use of their position as an early adopter and being on the WHS to get a number of quite weak sites inscribed e.g in 1979 Bulgaria got Boyana (2.68), Madara Rider (1.69), Ivanovo (2.71) and Kazanlak (2.2) inscribed. By 1985 Bulgaria had 9 inscribed WHS, only 1 of which (Rila) gets past a rating of 3 and includes our all time favourite Srebarna (0.91).

Nevertheless, the overall trend identified by joelontheroad is unarguable -albeit that it is perhpas slightly emphasised by the "recency" issue raised by Nan.

A simple sort of average ratings by inscription year shows a remarkable tendency for the 20th C inscriptions to have lower ratings. Though 1978 wasn't without its "weak" sites e.g Goree.
1978 4.05 
1982 3.85 
1990 3.83 
1984 3.82 
1987 3.79 
1989 3.72 
1979 3.67 
1981 3.65 
1992 3.65 
1986 3.6 
1991 3.54 
2002 3.52 
1988 3.49 
1985 3.48 
1980 3.43 
1997 3.4 
1993 3.39 
1983 3.36 
1994 3.36 
1995 3.35 
2013 3.31 
2003 3.29 
1996 3.28 
2016 3.26 
2001 3.24 
1999 3.19 
2005 3.19 
2000 3.17 
1998 3.15 
2011 3.14 
2004 3.06 
2010 3.06 
2008 3.05 
2007 3.01 
2014 3 
2012 2.96 
2017 2.92 
2015 2.88 
2006 2.8 
2009 2.8 
2019 2.79 
2018 2.72 

An interesting outlier is 2002: there were only 9 sites inscribed that year. With only 3 sites below 3 stars: Tokaji, Stralsund/Wismar and Paramaribo (I personally think the latter is worth more).