Monday, April 07, 2014

Problem with Future Orientation Index


Today, I stumbled upon the so called 'future orientation index' by chance, and retrieved a short 2012 paper which described it.

Upon browsing the paper by Preis et al (DOI:10.1038/srep00350, also available freely via Pubmed database), I found that the idea of using each country's queries of the years in arabic number, e.g. '2011, 2012', for future prospect is clever.

However, I noticed that it is without flaw: the authors disregarded cultural background of several countries. I should first say that I am not interested in trying to get the ranking of my country to be any higher, but as someone used to work as a reseacher, I could not avoid looking for ways to improve data quality.

Thailand, for example, is a Buddhist country, and people use mainly the Buddhist calendar (CE + 543) in their day-to-day lives. Not only that, Thais also use Thai numerals often interchangeably to Arabic numerals. If people from Thailand conducted searches for events or trends in their queries for next year's, they are also likely to search using the number representing Buddhist year, and might as well using Thai numerals when they are too lazy to swith the keyboard language.
Likewise, people in muslim countries could also be conducting searches using number for islamic year.

That's my 2 cents.

No comments: