reBlog from openparachute.wordpress.com: Open Parachute
Click on a quote to select it...
There’s a school of thought that the only reliable way to rank blogs is by using actual sitemeter data for visits. Unfortunately, most blogs don’t make this information public. Perhaps if more did bloggers could compare their statistics with those for other sites or have a listed ranking. This would help their interpretation.
I have managed to identify 37 New Zealand blogs with public sitemeter stats, or who self report this information – mostly from Tumeke’s blog ranking reports (see nz blogosphere rankings : March 2009). So how do these sites rank amongst themselves? Have a look at the table below for the April ‘09 data. And for those interested in comparing sitemeter data with the traditional blog ranking indices, and with the rankings of Tumeke, Halfdone and Open Parachute – look at my analysis below the table.
Update: Homepaddock has passed on (see Naked Stats) that her stats are also public so we now have 38 blogs on the list.
Tumeke bases his ranking score on sitemeter data for visits, or (in most cases) an estimate of these. His estimate is based on Alexa Rank data (3 month average). So I have used a correlation analysis to judge how well Alexa Rank, and other common indices, predict the visit numbers for April ‘09.
The results are summarised in the table below. These suggest that Alexa Rank and Technorati Rank explain 40% and 48% of the variation in visit numbers rerspectively. Both are better predictors than Alexa NZ Rank. However Google Reader Subscription Counts predicts equally well as Technorati Rank, and better than Alexa Rank. Unfortunately, fewer blogs are ranked by Technorati than Alexa which creates problems.
When the raw data is ranked (1-37, or 1-32 for Tehcnorati) for each blog I found Alexa Rank, Techorati Rank and Google Reader Subscription Counts would be equally predictive (about 50%) for ranking visits.
Alexa Rank40%*49%Alexa NZ Rank35%*40%Technorati Rank (32 sites)48% *50%Google reader48%50%
So how effective are the blog ranking systems used by Tumeke, HalfDone and Open Parachute in predicting ranking of sitemeter visit numbers? I have given estimates of the effectiveness for each system in the table below.
Although Tumeke’s data for April ‘09 are not yet available it would clearly give the best estimate because it uses actual, or estimated, visit numbers. These have a big influence because they contribute the largest amount (80% on average) to the final score (ranking score = actual or estimated visits + Technorati Authority + average weekly posts + average weekly comments ).
However, the fact that visit numbers are estimated from Alexa Rank for most blogs means the actual predictive power of the final score is far lower than estimated from these 37 sites alone (54% compared with 79% for March data using the 28 sites for which sitemeter data was available).
Halfdone’s ranking score uses the formula Score = (Alexa Rank x Alexa NZ Rank)/Technorati Authority/1000.
The predictive power of Open Parachute’s aggregate ranking of the 4 ranking indices only is slightly better than the total aggregate ranking (11 indices).
SystemPredictive powerTumeke (stats)*79%Tumeke (Estimated)**54%HalfDone52%Open Parachute (total rank)60%Open Parachute (page rank)63%
* March data using sitemeter data ** March data using Alexa Rank to estimate visits
I think there would be a lot of value in a blog ranking based on actual visit counts. The list above may encourage more NZ bloggers to allow public access to their stats.
On the other hand. I think there is value in having some idea of ranking based on other criteria such as links and RSS subscription counts. Different bloggers may be interested in different criteria for judging their performance.
openparachute.wordpress.com, Open Parachute, May 2009