Tester rating system: How am I rated as a tester?

All testers are rated against other active members of the community based on their quality and activity level (tester rating calculations are performed on a daily basis). There are a total of five tiers for the rated tester: Gold, Silver, Bronze, Proven, and Rated tier (if you do not have a badge/tier, this simply means that you have not filed enough "activity" points - this threshold equates to approximately five "Exceptionally Valuable" bugs). Testers can be Gold in Functional testing, and be lower or even unrated in testing types they are less interested in.

Detailed Ratings per Testing Type
Instead of earning one overall rating, testers earn one rating per testing type, based on their activity level and quality of work for test cycles of that testing type. The thresholds for tester ratings differ across testing types. For Functional, the breakdown of ratings among the tester population is as follows: Gold represents the top 4% of rated testers in the community; Silver represents the next 5%; Bronze represents the next 6%; Proven represents the next 35%. This breakdown may vary from each testing type, please refer to your tester snapshot for more details.

Additionally, Gold/Silver/Bronze testers receive higher payout rates than other members of the testing community: Gold testers receive a 10% premium on all approved reports, while Silver and Bronze testers receive a 5% and 2.5% premium, respectively. Payout increases are determined by the rating that applies to the test cycle's testing type. For example: If a tester is a Silver Security tester but a Rated Usability tester, he gets 5% extra for Security bugs but not Usability.
Here is more information about the parameters that drive your rating:

Activity Level

Lifetime participation level: # reported bugs, # approved bugs, etc. (see “Quality of Participation” below)
Recent participation level (prior 3,6,12 months)
Reliability: Reporting bugs and test cases for those in which the Test Cycle Agreement was checked and vice versa
Declining a test cycle has higher positive weight than Accepting a test cycle without submitting at least one report

Quality of Participation

Approval percentage now takes into account positive weight towards Exceptionally Valuable and Very Valuable approved reports. For example: a tester with an average (75%) bug approval percentage, but finds mostly “Exceptionally” valuable bugs will be rated higher than a tester with a very high (95%) bug approval percentage but finds mostly “Somewhat” valuable bugs.
Approval percentage for all types of reports (surveys, test cases, usability reports, etc) that also take into account tiered value.
Rejected disputes have more negative weight than an initial rejected report.
Accuracy of your initial bug report/severity classification. If your bug is reclassified or more information is requested it will hold more negative weight.