Kirsten’s answer to “Why are 50% of the reviews from the past couple of days less than three stars, yet this book has an…” > Likes and Comments
6 likes · Like
Today, I looked at the text only new reviews and 50% of them were 3 stars or less whereas in the overall tallies, only 20% of the reviews are 3 stars or less. Like the other two methods I used, this also suggests that most of the reviews are fake.
Could you explain more how the ratio of upvotes to the number of reviews in each star category indicates that 80% of the reviews are bots?
I didn't know there were fake reviews on here. I guess I never really thought about it...
Carrie, I didn't claim that 80% are bots, I claimed that they were probably fake or paid for. I used three independent methods to draw this conclusion.
But I still don't understand how the ratio of upvotes to the number of reviews in each star category leads to the conclusion that these reviews are fake or paid for.
That is one of three independent methods. Alone it wouldn't mean that much, but since result matches that of the other two methods, it is meaningful. I was a physicist for 20 years, so this is automatic for me.
back to top
date
newest »
newest »
message 1:
by
Kirsten
(new)
May 07, 2021 06:13AM
Today, I looked at the text only new reviews and 50% of them were 3 stars or less whereas in the overall tallies, only 20% of the reviews are 3 stars or less. Like the other two methods I used, this also suggests that most of the reviews are fake.
reply
|
flag
Could you explain more how the ratio of upvotes to the number of reviews in each star category indicates that 80% of the reviews are bots?
I didn't know there were fake reviews on here. I guess I never really thought about it...
Carrie, I didn't claim that 80% are bots, I claimed that they were probably fake or paid for. I used three independent methods to draw this conclusion.
But I still don't understand how the ratio of upvotes to the number of reviews in each star category leads to the conclusion that these reviews are fake or paid for.
That is one of three independent methods. Alone it wouldn't mean that much, but since result matches that of the other two methods, it is meaningful. I was a physicist for 20 years, so this is automatic for me.
