Methodology
Preparing our statistical reports, we have made significant efforts to ensure the accuracy of
our data and the validity of our statistical analysis process. The methodology of our data
gathering and analysis is described below:
Data Collection
We used a large sample of over 10 million, randomly selected, social networking profiles from our
freshly crawled people search index to create the data-set.
Note:
The only exception is the data for Facebook, in this case we did not have enough data available
so we had to rely on data provided to advertisers.
Age Data
Since some social networks block the age data for teenagers, and age data for very old users is mostly
incorrect, we limited our age calculation to ages 16-60.
Size Calculation
We calculated the size of each social networking site based on the number of actual profiles we indexed;
hence, inactive/deleted profiles were not included in this calculation.
Note:
The only exception is the data for Facebook, in this case we did not have enough data available so
we had to rely on data provided to advertisers.
Trend Calculation
The trends were calculated by sorting the (equally distributed) member profiles by the profile creation
dates and splitting the data into 20 consecutive groups, the changes between these groups represent the
trend data in our reports.
It is important to note that since different social networking sites were formed in different dates and
their growth rate is not always consistent, trend data do not represent points in time and only presents
the general trend overtime.
Geographical Grouping
The grouping into the different world regions and sub regions is based on the
Composition of macro geographical regions and geographical sub-regions
as defined by the United Nations Statistics Division.
For additional inquiries about this data please contact us via: stats
pipl.com
