.png?height=120&name=logo_matecat_big_white%20(1).png){kind=logo}
What is it for?
The Quality Report is the new feature that replaces the Editing Log.
The Quality Report provides a detailed view of your performance, both in terms of quality and productivity. It also provides a history of changes made throughout the document, suggestions received and QA issues.
Who is the Quality Report for?
Translators and revisers
If you are a translator, the Quality Report in Matecat is your go-to page to keep track of your productivity and, above all, improve your work.
Thanks to an industry-standard analytical approach to calculating the score, you will always know precisely the quality level of your jobs. You can also easily retrieve all flagged issues and learn at a glance how to improve.
Project managers
If you are a project manager, you can use the Quality Report in Matecat to keep track of quality for your projects.
The score is updated in real-time and you can easily identify at a glance all flagged issues, in order to take action.
The indications about Post-Editing Effort (PEE) and Time to Edit (TTE) provide insights about productivity and the actual effort it took to complete the job.
NOTE: Time-to-edit is calculated by adding up the time a translator spends on each segment. If a translator goes back to the same segment several times (opening and closing the segment), the times for each edit are added together.
For the calculation of the Post-Editing Effort, on the other hand, some segments are excluded:
- Segments that have a time-to-edit of over 25 seconds per word, and
- Segments that have a time-to-edit below 0.5 seconds per word.
How is the quality score calculated?
Matecat uses an analytical approach to assess quality based on data provided by the reviser.
There are basically two categories of Quality Score: Pass and Fail.
The numeral value of the score is calculated using this formula:
(total error points * 1000) / reviewed words
NOTE: reviewed words take into account segments that have been approved, excluding ICE matches.
On the other hand, ICE matches that have been modified for any reason are included in the reviewed words count.
The quality score is then compared to a maximum amount of tolerated error points for 1,000 words to determine the final quality.
Looking at the example above, we can notice that there are 0,5 error points on a total of 63 reviewed words.
If we apply the formula we obtain
(0,5*1000)/63 = 7.94
QR in detail
The Quality Report in Matecat is divided into three main sections:
- QR Job summary – Here you can find job data, such as final score, average post-editing effort and time to edit, as well as total weighted words and reviewed words.
- Issue grid – This is a detailed breakdown of all the issues found and the error points generated.
- Segment details – In this section, you can find the history and segment metadata, such as suggestion received, difference view between last translation and suggestion, and between last revision and last translation (if applicable).
You can also see the specific post-editing effort and time to edit, status, QA issues, speed per word, and word count for each segment, and filter the segments per status, issue category, and severity.
If the segment has been commented on through the highlight mode during the revision, you can read the comment by hovering the mouse over the bubble icon in the QA row. You will see the highlighted text and the comment left by the reviewer.
If you need, you can download the QR in CSV format.
