Should point out Neutral was only introduced in May 2013.
The neutral rating was definitely introduced in January 2014 (apart from my "elephant's memory" here
are the proofs). Perhaps though a few members asked you to change the rating of some past reviews into neutral and that is why in the database it appears that the earliest Neutral entry was made prior to January 2014 (May 2013 that is) leading you to believe that the rating was introduced since then.
I doubt that lately there is a significant statistical increase in the number of negatives. But to find out for sure you need to compare the last month and the penultimate one (or the average since Jan 2014 excluding last month).
19% negatives in past 30 days
24% negatives since 2010
13% neutrals in past 30 days
8% neutrals since 2010
What the above statistic says is this:
1. average of 13% neutral for last month is roughly the same as the average post-2014 (assumption). As the neutrals posted pre-2014 were virtually zero (apart from few converted ones) an overall average of 8% means that the reviews (regardless of rating) posted post-2014 account for 8/13=61.5% of the total.
2. The negatives, post-2014 became 19%. To find out how much it was pre-2014 you solve 0.385x+0.615*19=24 which gives x=32. This means roughly 32% of the reviews pre-2014 were negative
.So the introduction of the neutral rating, which takes a 13% of the total each month since then, caused the percentage of negatives to decrease by the same amount (from 32% to 19%) whilst the percentage of positives remained unchanged.It should be noted here that we are talking about percentages. The absolute number of negatives has been increased (but so has that of the positives and neutrals) due to the increase in the number of reviews. This year we will surpass 3000 reviews by May while it took until August to achieve the feat last year.