What an interesting pile of rubbish.
Here's a simpler break down.
Imagine I found 10 countries that were recently at war with their neighbors out of 200 countries
Then I took a survey in those 10 countries stating that I wanted to know if they liked their neighbors and if they wanted them as neighbors anymore
But, this is the internet and surely there will be some backlash so for good measure and to make things legit I now go and add 5 countries that have had trouble 10 years prior to my data so as to not look like I have an agenda
All of the sudden the really high numbers start to look like real statistics because of course the median gets shifted and you've got a data-set that is advantageous to click-bait
Now, I have to of course talk about places that love their neighbors but my already burned mathematical wizardy tells me to not include all the countries I already did a survey in because god-forbid I find data in those countries to contradict my earlier data so I take the colored map and pick all the non-evaluated slots
What you end up with is a skewed data-set that looks sort of real so folks at a cocktail party can have one of two conversations
Did you hear, we're #2 in the list of countries with the least racism. I mean I tell my maid who moved here from her country that she's a worthless bag of dirt every day but thats just me telling her to do her job surely I am not racist.
We're #2? But I didn't know I was supposed to let the guys who tried to break into my house and shoot my parents just have their way and of course invite them over for biscuits the next day because that's the Christian thing to do.
In the end its all circumstantial and contextual. Of course some things are black and white as night and day but to think that the report above is a clear indication of racism at its peak in the developing world is myopic at best. It's here for sure but not like what's being reported above.