Twitter is a social networking and microblogging service that enables its users to send and read messages known as tweets. Tweets are text-based posts of up to 140 characters displayed on the author’s profile page and delivered to the author’s subscribers who are known as followers.
Ranking is a part of our everyday life. It is even more common in this part of the world, where there are several awards given out during lavish ceremonies to recognize those who have excelled in their endeavours. We have music awards, sport awards, journalism awards etc. The awards not only serve as an endorsement of excellence but also motivate those awarded to do better and for the lower ranks to work harder so they could also be recognized in future.
Since the web is not yet mainstream in this parts, there are not yet many web awards and thus the web-savvy have had to make do with mere rankings. The problem with such rankings is that they are generated by computer algorithms which are fed very limited inputs. The results are thus often heavily skewed.
A case in point are Twitter Rankings. I have a draft blog post titled “Top 10 Twitter Users in Ghana” in my blog for several months now. It has remain unpublished because it has been very challenging finding that web tool that could be used objectively to rank the Twitter users in Ghana. It is still possible for me to combine the results from the most credible Twitter ranking websites but I would need to come-up with a formula that can aggregate the factors and give an objective ranking. All these is a lot of work.
Some of the factors limiting an objective Twitter Ranking include:
One of the best means to classify Twitter rankings is by ranking users according to their country, region/state or city. Example: Top 10 Twitter Users in South Africa, Top 20 Twitter Users in Accra etc. It is thus not surprising that many of the Twitter ranking websites include this factor. The challenge here however is, not everyone enters their real geographical location. Not a few Twitter users enter fanciful information as their geographical information. Examples include: somewhere in cyberspace, everywhere, and some enter geographic coordinates. It is thus impossible to accurately compile the full list of Twitter users in a particular location, for statistical reasons.
Another challenge with compiling statistics on Twitter users is that many of the websites that offer Twitter usage information are opt-in. Twitter users themselves have to sign-up on these websites for their stats to be factored into the overall ranking. Since they do not cover every Twitter user by default, their results are skewed. Most Twitter Ranking websites are culpable in this area.
Measuring Real-Life Influence
Tweets from a Head of State or an internationally recognized entertainer for example, would be far more influential for obvious reasons but I do not know of any Twitter Ranking website that includes a Twitter user’s real life influence.
My next blog post would highlight some Twitter User Ranking websites.
I agree that its quite an arduous task trying to rank correctly influential twitter users. My interaction on Twitter has shown that a lot of Nigerians for example who I believe are tweeting via blackberry don’t include their real location; but they ‘loads’ of followers and tweets from their end; hence I see most of these rankings as skewed like you rightly said.
Nice article and I believe, you were pushed to write this because, you still think it’s all skewed information. It’s your opinion and I just have to respect that.. Love you style, man 😉
Well spoken, i once wrote blog post a bout Top ten social media kings in Kenya, http://www.kachwanya.com/2009/08/07/top-ten-kenyan-social-media-kings/. In that i had to use a combination of many social networks, blogs and websites to come up with the list and still at the end after reading various comments, i realized that is hard to achieve accurate ranking in the usage of the social media. Yes you can tell the influencers around but you can’t measure how much or how high one is above the rest.