Translation quality measurement2

1,184
-1

Published on

Published in: Education
0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
1,184
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
47
Comments
0
Likes
2
Embeds 0
No embeds

No notes for slide

Translation quality measurement2

  1. 1. Translation Quality Measurement
  2. 2. Biographical Notes on the Authors <ul><li>Riccardo Schiaffino </li></ul><ul><li>Riccardo Schiaffino worked as translator, translation manager and special software translation project lead for a major software company. As a translation manager, Riccardo worked on the improvement of translation quality and on translation quality metrics and tools. He holds an MA degree in Translation, and has been working in translations for over 18 years, first in Italy and then in the U.S. Riccardo is ATA accredited. Contact: [email_address] . </li></ul><ul><li>Franco Pietro Zearo </li></ul><ul><li>Franco Pietro Zearo is a project manager with Lionbridge Technologies in Boulder, Colorado. He holds a degree in translation from the Advanced School of Modern Languages for Translators and Interpreters at the University of Trieste, Italy, and earned an MBA from the University of Phoenix. Before joining Lionbridge in 1996, he worked as a freelance technical translator in Italian, English, and Russian. At Lionbridge, he has held positions in translation, localization analysis, presales, and cultural and globalization consulting. He has been responsible for translation quality on numerous projects for many Fortune 500 clients. In his previous role as senior technical translator, he helped define best practices for the translation department. Contact: [email_address] . </li></ul>
  3. 3. Overview <ul><ul><li>Technical translation and quality </li></ul></ul><ul><ul><ul><li>Translation quality initiatives </li></ul></ul></ul><ul><ul><ul><li>Quality Control vs. Quality Assurance </li></ul></ul></ul><ul><ul><li>Our proposal for quality assurance </li></ul></ul><ul><ul><ul><li>Checklists </li></ul></ul></ul><ul><ul><ul><li>Sampling techniques </li></ul></ul></ul><ul><ul><li>Conclusions </li></ul></ul><ul><ul><ul><li>Importance of cost/benefit factors </li></ul></ul></ul>
  4. 4. Overview <ul><li>Measuring Quality </li></ul><ul><li>Translation Quality Assessment </li></ul><ul><li>Quality Assurance Forms </li></ul><ul><li>Error Categories </li></ul><ul><li>Sampling </li></ul><ul><li>Translation Quality Index </li></ul><ul><li>Questions and Answers </li></ul>
  5. 5. Overview <ul><li>Why Is Quality Measurement Important? </li></ul><ul><li>How to Set Up a Quality Measurement System </li></ul><ul><li>Demo of a Translation Quality Measurement Tool Prototype </li></ul><ul><li>Practical Recommendations </li></ul><ul><li>Questions & Answers </li></ul>
  6. 6. Our Definition of Quality <ul><li>Functional approach to quality </li></ul><ul><ul><li>Different views of translation lead to:  Different concepts of quality  Different assessments </li></ul></ul><ul><li>Quality is defined as meeting the needs and expectations of the customer or user. </li></ul>
  7. 7. Our Definition of Quality <ul><li>Functional approach to quality </li></ul><ul><li>Quality is defined as consistently meeting the needs and expectations of the customer or user </li></ul>
  8. 8. Correct Translation <ul><li>A correct translation is a translation with no errors or where total error points result in a Translation Quality Index above the desired threshold </li></ul>
  9. 9. <ul><ul><li>Conformance to specifications </li></ul></ul><ul><ul><li>Customer’s vs. One’s own </li></ul></ul><ul><ul><li>Fitness for use </li></ul></ul><ul><ul><li>How well the translation performs its intended purpose </li></ul></ul><ul><ul><li>Value ( = quality & price) </li></ul></ul><ul><ul><li> How well the translation performs its intended purpose at a price customers are willing to pay </li></ul></ul><ul><ul><li>Support </li></ul></ul><ul><ul><li>E.g.: Printing, testing </li></ul></ul><ul><ul><li>Psychological impressions </li></ul></ul><ul><ul><li> E.g.: In-country translators; certification </li></ul></ul>Customer-driven Considerations
  10. 10. Customer-driven Considerations
  11. 11. Importance of Quality <ul><li>Quality as a Competitive Weapon </li></ul><ul><ul><li>Good Quality  Higher Profits Good quality of translation (product) and service (process) can pay off in higher profits </li></ul></ul><ul><ul><li>Improving on quality can reduce costs and speed up time-to-market </li></ul></ul>
  12. 12. Why is Quality Measurement Important? <ul><li>You can’t manage what you can’t measure </li></ul><ul><ul><li>It is difficult to improve something if you cannot measure it. </li></ul></ul><ul><ul><li>Such measurement should be repeatable and objective . </li></ul></ul><ul><ul><li>Different persons should arrive at similar assessment for the same piece of translation. </li></ul></ul>
  13. 13. Why is Quality Measurement Important? <ul><ul><li>It is difficult to improve something if you cannot measure it. </li></ul></ul><ul><ul><li>Such measurement should be repeatable and objective . </li></ul></ul><ul><ul><li>Different evaluators should arrive at similar assessment for the same piece of translation. </li></ul></ul>
  14. 14. Why is Quality Measurement Important? <ul><li>It is difficult to improve something if you cannot measure it </li></ul><ul><li>Metrics provide: </li></ul><ul><ul><li>A way to objectively quantify a process </li></ul></ul><ul><ul><li>A means to reduce the cost of poor quality </li></ul></ul><ul><ul><li>A means to increase customer satisfaction </li></ul></ul><ul><ul><li>An opportunity for benchmarking </li></ul></ul><ul><ul><li>Competitive advantages </li></ul></ul>
  15. 15. “ You cannot measure quality” <ul><li>This is not true: </li></ul><ul><li>There are certain components of translation quality that will always remain subjective . </li></ul><ul><li>However , </li></ul><ul><li>There are other elements that can be objectively measured. </li></ul><ul><li>By concentrating of these, we believe we can achieve a satisfactory measurement of translation quality. </li></ul>
  16. 16. Who Benefits from Reliable Translation Quality Measurement? <ul><li>Professional Translators </li></ul><ul><li>Translation Companies and In-House Translation Departments </li></ul><ul><li>Translation Customers and Users </li></ul>
  17. 17. Why Do We Make Errors? <ul><li>The reasons behind the errors are separate from the measurement of the errors: Studying why errors happen is important, but it pertains more to quality control and improvement than to quality assurance </li></ul><ul><li>E.g., capitalization errors due to the &quot;Autocorrect&quot; (mis)feature of MS Word (e.g., HBsAg &quot;corrected&quot; to HbsAg) </li></ul>
  18. 18. QC vs QA <ul><li>Quality Control (QC) </li></ul><ul><ul><li>Quality verification over the whole text. </li></ul></ul><ul><ul><li>Example : editing. </li></ul></ul><ul><li>Quality Assurance (QA) </li></ul><ul><ul><li>Sampling techniques, control of quality over a (statistically significant) sample of the whole text. </li></ul></ul><ul><ul><li>Example: quality measurement. </li></ul></ul>
  19. 19. QC vs QA <ul><li>Quality Control (QC) </li></ul><ul><ul><li>Quality verification over the whole text. </li></ul></ul><ul><ul><li>Example : Editing. </li></ul></ul><ul><li>Quality Assurance (QA) </li></ul><ul><ul><li>Sampling techniques, control of quality over a (statistically significant) sample of the whole text. </li></ul></ul><ul><ul><li>Appropriate use: Q uality measurement. </li></ul></ul>
  20. 20. QC v QA <ul><li>Quality Control (QC) </li></ul><ul><ul><li>Quality verification over the whole text. </li></ul></ul><ul><ul><li>Example: Editing </li></ul></ul><ul><li>Quality Assurance (QA) </li></ul><ul><ul><li>Sampling techniques, control of quality over a (statistically significant) sample of the whole text. </li></ul></ul><ul><ul><li>Appropriate use: Quality measurement </li></ul></ul>
  21. 21. Translation Quality Factors
  22. 22. Inspection Points Key Principle: Reject “defective material” at its lowest value Proof Edit Translation SL Content Development (GIGO) $ Value of Service Stages of Production
  23. 23. Inspection Points Key Principle: Reject “defective material” at its lowest value Proof Edit Translation SL Content Development (GIGO) $ Value of Service Stages of Production
  24. 24. Cost/Benefit Analysis <ul><li>Quality measurements are a tool to determine the optimal level of quality. They could help us identify a cut-off point. </li></ul>
  25. 25. Ideas from other disciplines <ul><li>Software project management techniques </li></ul><ul><li>W. Edwards Deming and other quality assurance experts </li></ul>
  26. 26. When we study translation quality, we can focus on different things: <ul><li>The translated text (the “ product ”) </li></ul>The translation process (the “ process ”) The translator
  27. 27. Product & Process Assessment <ul><ul><li>Translation quality assessment must apply to both : </li></ul></ul><ul><ul><ul><li>The translated text (the “ product ”) </li></ul></ul></ul><ul><ul><ul><li>The translation process (the “ process ”) </li></ul></ul></ul>
  28. 28. Product & Process Assessment <ul><ul><li>Translation quality assessment must apply to both : </li></ul></ul><ul><ul><ul><li>The translated text (the “ product ”) </li></ul></ul></ul><ul><ul><ul><li>The translation process (the “ process ”) </li></ul></ul></ul>
  29. 29. Translation Quality Initiatives <ul><li>The translated text </li></ul>The translation process The translator SAE J2450 LISA QA DIN 2345 ISO 900x UNI EN 10754 EUATC ASTM ATA and other translators’ certification initiatives
  30. 30. <ul><ul><li>ISO 9002 </li></ul></ul><ul><ul><li>EUATC Quality Standard </li></ul></ul><ul><ul><li>DIN 2345 </li></ul></ul><ul><ul><li>ASTM Standard for Language Translation </li></ul></ul><ul><ul><li>SAE J2450 </li></ul></ul><ul><ul><li>LISA QA Model </li></ul></ul><ul><ul><li>Academic translation theories and studies </li></ul></ul><ul><ul><li>Private sector methodologies </li></ul></ul>Translation Quality Initiatives
  31. 31. Quality Measurement: Our Proposal <ul><li>What Can Other Disciplines Teach Us? </li></ul><ul><ul><li>Use checklists to collect the data </li></ul></ul><ul><ul><ul><li>Identify types of errors, issues or problems </li></ul></ul></ul><ul><ul><ul><li>Determine relative importance of issues (may be different for different languages; e.g., spelling errors in English, French or Italian) </li></ul></ul></ul><ul><ul><li>Use sampling techniques to assess your quality level </li></ul></ul><ul><ul><ul><li>Determine percent thresholds for various levels of quality </li></ul></ul></ul><ul><ul><ul><li>Determine whether you have achieved your target quality or not </li></ul></ul></ul>
  32. 32. Criteria for Successful Quality Measurements <ul><ul><li>Translation quality measurements should be: </li></ul></ul><ul><ul><li>Repeatable (two assessments of the same sample yield similar results) </li></ul></ul><ul><ul><li>Reproducible (different evaluators should arrive at a similar assessment for the same piece of translation </li></ul></ul><ul><ul><li>Objective (void of subjective bias) </li></ul></ul>
  33. 33. Classification of Errors
  34. 34. Measurement through Circumstantial Evidence <ul><li>Errors are circumstantial evidence of quality </li></ul><ul><li>We believe that precise error measurement provides sufficient indication of good and bad translations </li></ul><ul><li>A good translation is a translation with very few errors or none at all </li></ul>
  35. 35. Definition of Errors <ul><li>Deal with errors only when they violate agreed upon protocols of engagement whether implicit or explicit </li></ul><ul><li>Examples of explicit and implicit criteria: </li></ul><ul><li>Non-compliance errors (e.g. not following instructions) </li></ul><ul><li>Violations of generally accepted language conventions </li></ul>
  36. 36. Summary: Error Categorization <ul><li>Select a (small) set of categories </li></ul><ul><ul><li>CTQ: Critical-To-Quality categories </li></ul></ul><ul><li>Provide clear definitions </li></ul><ul><li>Set tolerance limits </li></ul><ul><ul><li>Min / Max # of errors per X words </li></ul></ul><ul><li>Assign a weight </li></ul><ul><ul><li>Critical, Major, Minor </li></ul></ul>
  37. 37. Summary: Error Categorization <ul><li>Select a (small) set of categories </li></ul><ul><ul><li>CTQ: Critical-To-Quality categories </li></ul></ul><ul><li>Provide clear definitions </li></ul><ul><li>Assign a weight </li></ul><ul><ul><li>Critical, Major, Minor </li></ul></ul>
  38. 38. Real Life Examples <ul><li>Development of translation quality measurement at J.D. Edwards </li></ul><ul><li>Use of sampling techniques for quality assurance at Lionbridge </li></ul>
  39. 39. The J.D. Edwards’ QA Form Language Customization <ul><ul><li>Weighting the major categories </li></ul></ul>
  40. 40. The J.D. Edwards’ QA Form Language Customization <ul><ul><li>Weighting the items within the major categories </li></ul></ul>
  41. 41. The J.D. Edwards’ QA Form Language Customization <ul><ul><li>Weighting the items within the major categories (detail) </li></ul></ul>
  42. 42. How We Worked to Develop Our Spreadsheet <ul><li>Determine type of errors, issues or problems </li></ul><ul><li>Determine relative importance of issues (may be different for different languages; e.g., spelling errors in English, French or Italian) </li></ul><ul><li>Determine which are the responsibility of translation </li></ul><ul><li>Determine tolerance limits for various levels of quality </li></ul>
  43. 43. Translation Quality Measurement Tool <ul><li>The Translation Quality Measurement tool helps to measure process quality </li></ul><ul><li>It is NOT an editing tool, but it serves to measure whether a process is effective </li></ul>
  44. 44. Use of the Tool <ul><li>Use the tool to measure the effectiveness of quality control process </li></ul><ul><li>Analyze the results obtained through the tool (control charts) </li></ul><ul><li>If the process is NOT in statistical control </li></ul><ul><ul><li>Discover special causes and deal with them appropriately </li></ul></ul><ul><ul><li>Remove them if they are negative </li></ul></ul><ul><ul><li>Incorporate them in process if they are positive </li></ul></ul><ul><li>Improve the process when it is in statistical control </li></ul>
  45. 45. A TQI Tool Prototype
  46. 46. ATA Implementation
  47. 47. ATA Implementation
  48. 48. SAE Implementation (Modified)
  49. 49. SAE Implementation (Modified)
  50. 50. TQI Log Error Category EP Remarks Bookmark Path File Grader Date Is Formal 2 irregular capitalization; should be is, not Is x2_is C:Documents and SettingsRS1643403DesktopQuality Measurement CoffeMakerTest.doc RS1643403 11/1/2003 aluminium Formal 1 British spelling; American English should be aluminum, not aluminium x3_aluminium C:Documents and SettingsRS1643403DesktopQuality Measurement CoffeMakerTest.doc RS1643403 11/1/2003 food Meaning 2 The container is not made to cook food, it is made to brew a beverage. x5_food C:Documents and SettingsRS1643403DesktopQuality Measurement CoffeMakerTest.doc RS1643403 11/1/2003 right for the gas cooker, the electric plate and the pyroceram Formal 1 A better phase might be: &quot;acceptable for use on gas and electric stoves.&quot; x6_right_for C:Documents and SettingsRS1643403DesktopQuality Measurement CoffeMakerTest.doc RS1643403 11/1/2003 pyroceram Meaning 4 The word &quot;pyroceram&quot; is unknown to most English speakers. x7_pyroceram C:Documents and SettingsRS1643403DesktopQuality Measurement CoffeMakerTest.doc RS1643403 11/1/2003 wash Meaning 2 In English, the word &quot;wash&quot; typically means water and soap. The instructions specify only using water, so a better word choice would be &quot;rinse.&quot; x8_wash C:Documents and SettingsRS1643403DesktopQuality Measurement CoffeMakerTest.doc RS1643403 11/1/2003 trow Meaning 2 The word &quot;trow&quot; is a misspelling of &quot;throw.&quot; x9_trow C:Documents and SettingsRS1643403DesktopQuality Measurement CoffeMakerTest.doc RS1643403 11/1/2003 total 14 N. of words 42 TQI 67%
  51. 51. Use of Checklists <ul><ul><ul><li>There are several quality assessment methodologies that rely on the use of checklists – among these the LISA methodology . </li></ul></ul></ul>
  52. 52. Use of Checklists <ul><ul><ul><li>There are several quality assessment methodologies that rely on the use of checklists – among these the LISA methodology . </li></ul></ul></ul><ul><ul><ul><li>We would like, however, to advocate the use not of “universal” checklists, but of checklist specifically tailored to each language. </li></ul></ul></ul><ul><ul><ul><ul><li>Checklists for evaluating translation companies </li></ul></ul></ul></ul><ul><ul><ul><ul><li>Checklists and tests for evaluating translators </li></ul></ul></ul></ul><ul><ul><ul><ul><li>Checklists for evaluating translations </li></ul></ul></ul></ul><ul><ul><ul><ul><ul><li>Limitations of universal checklists </li></ul></ul></ul></ul></ul><ul><ul><ul><ul><ul><li>Language specific checklists (example, different weight of spelling correctness for different languages) </li></ul></ul></ul></ul></ul>
  53. 53. Development of Translation Quality Measurement at J.D. Edwards <ul><li>From the concept of checklists to a spreadsheet of measurements </li></ul><ul><li>Checklists are appropriate to control whether a certain action has been performed or not (e.g., spell check done or not – as opposed to a measurement of how many spelling mistakes were found) </li></ul><ul><li>Based on LISA model ( www.lisa.org ) </li></ul><ul><li>Flexibility (different settings for different languages) </li></ul>
  54. 54. Use of Quality Assurance Forms <ul><ul><li>The LISA Quality Assurance Form </li></ul></ul>
  55. 55. Purposes of sampling according to LISA <ul><li>To determine whether something has been done or not. </li></ul><ul><li>To accept / reject the batch of product at hand. </li></ul><ul><li>To determine if the process that produced the product at hand was within acceptable limits. </li></ul>
  56. 56. Guidelines for Sampling <ul><li>Select a sample </li></ul><ul><ul><li>Selection criteria (e.g. random, systematic) </li></ul></ul><ul><ul><li>Size considerations </li></ul></ul><ul><ul><li>Cost considerations </li></ul></ul><ul><li>Evaluate the sample </li></ul><ul><ul><li>Repeatable, reproducible, objective </li></ul></ul><ul><li>Investigate the outcome / causes </li></ul><ul><li>Correct / Improve </li></ul>
  57. 57. Statistical Methods <ul><ul><li>Defect Counts </li></ul></ul><ul><ul><ul><li>Statistics on Effort Per Defect </li></ul></ul></ul><ul><ul><li>Defect Density Prediction </li></ul></ul><ul><ul><li>Defect Pooling </li></ul></ul><ul><ul><li>Defect Seeding </li></ul></ul>
  58. 58. Defect Counts <ul><ul><li>Useful to obtain a quantitative measurement of how much QC work to do. </li></ul></ul><ul><ul><li>Ratio of new defects to defects solved. </li></ul></ul><ul><ul><li>Statistics on Effort Per Defect </li></ul></ul><ul><ul><ul><li>In order to estimate the scope of the defect correction work, it is necessary to have good data on the time necessary to fix the various types of defects </li></ul></ul></ul>
  59. 59. Defect Density Prediction <ul><ul><li>One way to judge whether the QC work on a translation project is complete is to measure its defect density (the number of defects per page, per 1,000 words or per screen). </li></ul></ul>
  60. 60. Defect Pooling <ul><ul><li>Defect pooling is a simple defect prediction technique that separates the defects found in a translation sample into two pools . </li></ul></ul><ul><ul><li>Depending on the number of defects found in either of the two pools (but not in both) it is then possible to estimate the defects that have not been found in the sample. </li></ul></ul><ul><ul><li>This number can then be used to estimate the number of defects in the entire project. </li></ul></ul>
  61. 61. Defect Seeding <ul><ul><ul><li>Defect seeding is a statistical technique in which a sample of a population is extracted and used to estimate the total population. </li></ul></ul></ul><ul><ul><ul><li>The technique works by deliberately inserting (“seeding”) defects in a complete translation that will be QCed. </li></ul></ul></ul><ul><ul><ul><li>The ratio of the seeded defects found compared to the total number of defects seeded provides a rough estimate of the total number of translation defects yet to be found. </li></ul></ul></ul><ul><ul><ul><li>A common problem with this type of technique is forgetting to remove the errors deliberately inserted . </li></ul></ul></ul>
  62. 62. Calibration and Error Seeding <ul><li>One of the things one can do to calibrate a translation quality measurement tool (or process) is to use error seeding : Not only to be able to estimate what percentage of errors is not discovered, but also in order to estimate how much variance there is in assessing the errors that do get discovered. </li></ul>
  63. 63. Suggested process: calibration of a (generic) translation quality measurement tool <ul><li>Have the sample translations (a suitable number of them) scored &quot;by hand&quot; by expert translators, so as to obtain a suitable range of evaluated samples, from very good to very bad. </li></ul><ul><li>Importance of tightly defining the pool of reviewers </li></ul><ul><li>Importance of instructions for reviewers </li></ul><ul><li>Have other expert translators score the same tests, but using the tool </li></ul><ul><li>On the basis of the results of the previous two steps, adjust the weights, types of errors, etc. in the tool until you are satisfied it is going to help in assessing translation quality - that is, until you are confident that trained evaluators are going to obtain with the tool consistent and reliable scores </li></ul><ul><li>In doing this remember to remove from the kind of errors that can be assessed those that are controversial, i.e., those that lead to differences of opinion whether they are errors or not </li></ul><ul><li>Finally adjust the tool so that it produces the range of error scores that is useful for your organization (e.g., if you want &quot;0&quot; or 100% as your perfect score) </li></ul>
  64. 64. Translation Quality Index (TQI) <ul><li>The TQI is a number—obtained by the rigorous application of a QA process—that indicates the quality of a given translated text </li></ul>
  65. 65. The concept of a “Translation Quality Index” <ul><li>Translation Quality Index (TQI) </li></ul><ul><li>A number—obtained by the rigorous application of a QA form—that is indicative of the quality of a given translation </li></ul>
  66. 66. Delusions of Accuracy <ul><li>“ Averages can be calculated to nineteen places of decimal with astonishing ease. </li></ul><ul><li>When the job is done, it looks very accurate. </li></ul><ul><li>It is an easy and fatal step to think that the accuracy of our arithmetic is equivalent to the accuracy of our knowledge about the problem in hand.” </li></ul><ul><li>M.J. Moroney, Facts from Figures </li></ul>
  67. 67. Index / Indices <ul><li>Depending on one’s purpose, there may be more than a single TQI. </li></ul><ul><li>E.g., a TQI may be developed for external purposes (to standardize the work obtained from outsourcing). </li></ul><ul><li>Another TQI may be primarily for internal purposes (to measure the quality of a given special process). </li></ul>
  68. 68. An Example of a “Translation Quality Index” (1) <ul><li>LISA QA Model ver. 1.0 (1995) </li></ul><ul><li>3,000 words (12 pages @ 250 words) </li></ul><ul><li>30 error points </li></ul><ul><li>30 error pts / 3,000 words = 1.0% </li></ul><ul><li>10,000 error pts out of 1 million words </li></ul><ul><li>DPMO = 99.0% = TQI </li></ul>
  69. 69. An Example of a “Translation Quality Index” (2) <ul><li>Microsoft Quality Standards for Print ver. 1.0 (1998) </li></ul><ul><li>10,000 words (40 pages @ 250 words) </li></ul><ul><li>0 major errors </li></ul><ul><li>15 minor errors </li></ul><ul><li>15 errors / 10,000 words = 0.15% </li></ul><ul><li>1,500 errors out of 1 million words </li></ul><ul><li>DPMO = 99.85% = TQI </li></ul>
  70. 70. An Example of a “Translation Quality Index” (3) <ul><li>2,000 words (8 pages @ 250 words) </li></ul><ul><li>1 critical error </li></ul><ul><li>2 major errors </li></ul><ul><li>3 minor errors </li></ul><ul><li>6 errors / 2,000 words = 0.3% </li></ul><ul><li>3,000 errors out of 1 million words </li></ul><ul><li>DPMO = 99.7% = TQI </li></ul>
  71. 71. Let’s Calculate Two TQIs Implicit TQI = 93.2% Implicit TQI = 99.0% 250 words (estimate) 17 error points 17 error pts / 250 words = 0.068 3,000 words (12 pages @ 250 words) 30 error points 30 error pts / 3,000 words = 0.01 ATA Framework for Standard Error Marking LISA QA Model ver. 1.0 (1995)
  72. 72. Control Charts <ul><li>Concept of “statistical control” </li></ul>
  73. 73. Process Flow Diagram
  74. 74. Example of Process for Accepting or Rejecting a Translation Process <ul><li>1) Determine and describe what your process actually is (NOT what you think it is or what the process should be) </li></ul><ul><li>2) Measure the quality you have now </li></ul><ul><li>3) Determine if you have special cases, and if so, eliminate them (what the special cases are can be seen through the use of control charts) </li></ul><ul><li>4) Once the process is in statistical control (i.e., any quality variance is not due to special cases) </li></ul><ul><li>5) Change the process to improve quality </li></ul><ul><li>6) Measure the new level of quality to determine the effectiveness of the changes to the process </li></ul>
  75. 75. Very Important <ul><li>Improvements made to the overall process should result in improvements to the product (the translation) </li></ul><ul><li>Measurements of the product quality should indicate if there have been actual improvements to the process </li></ul><ul><li>Therefore, means to measure product quality must be in place </li></ul>
  76. 76. How to Apply Statistical Methods for Quality Improvement <ul><li>Define error categories and tolerances </li></ul><ul><li>Create a QA form </li></ul><ul><li>Obtain a TQI index </li></ul><ul><li>Use the TQI index to improve the translation process </li></ul>
  77. 77. How to Set Up a Quality Measurement System – Stage 1, Preparation <ul><li>Collect examples of good and bad translations </li></ul><ul><li>Analyze the examples to separate controversial issues from agreed upon errors </li></ul><ul><li>Decide what to measure (error categorization) </li></ul><ul><li>Define what to measure in as many details as necessary (error definition) </li></ul>
  78. 78. How to Set Up a Quality Measurement System – Stage 2, Calibration <ul><li>Assign a weight to various types of errors </li></ul><ul><li>Determine critical errors (if necessary) </li></ul><ul><li>Repeat 3, 4, 5, and 6 until the system works in an objective, repeatable, and reproducible way </li></ul>
  79. 79. Quality Assurance Forms and Tools <ul><li>Create a QA form (or a tool) to help graders give objective scores </li></ul>
  80. 80. How to Set Up a Quality Measurement System – Stage 3, Sampling <ul><li>Sampling </li></ul><ul><ul><li>Selection criteria (e.g. random, systematic) </li></ul></ul><ul><ul><li>Size considerations (the greater the sample, the more accurate the results) </li></ul></ul><ul><ul><li>Select confidence intervals, margins of error </li></ul></ul><ul><ul><li>Cost considerations (find the point of diminishing returns) </li></ul></ul>
  81. 81. How to Set Up a Quality Measurement System – Stage 4, Measurement <ul><li>Measurement </li></ul><ul><ul><li>Evaluation must be repeatable, reproducible, objective </li></ul></ul><ul><ul><li>Use of independent auditors </li></ul></ul><ul><ul><li>Calculation of a Translation Quality Index (TQI) </li></ul></ul>
  82. 82. How to Set Up a Quality Measurement System – Stage 5, Statistical Analysis <ul><li>Investigate the Outcome </li></ul><ul><ul><li>At this stage there shouldn’t be any special causes (use of control charts) </li></ul></ul>
  83. 83. How to Set Up a Quality Measurement System – Stage 6, Process Improvement <ul><li>Take corrective actions (process improvement) </li></ul><ul><li>Compare the TQI values before and after a process change to check for actual process improvement </li></ul>
  84. 84. How to Set Up a Quality Measurement System – Summary <ul><li>Preparation </li></ul><ul><li>Calibration </li></ul><ul><li>Sampling </li></ul><ul><li>Measurement </li></ul><ul><li>Statistical Analysis </li></ul><ul><li>Process Improvement </li></ul>
  85. 85. Practical Recommendations <ul><li>Importance of </li></ul><ul><li>Glossaries (for terminology) </li></ul><ul><li>Style Guides (for syntax) </li></ul><ul><li>Translation Instructions (for special cases) </li></ul><ul><li>Protocols of Engagement (regulating the treatment of errors/defects and defining the acceptance/rejection criteria) </li></ul><ul><li>Translation Guide for Customers (including a detailed customer checklist to specify what is important and what is not) </li></ul>
  86. 86. Conclusions <ul><li>Desirability of common standards (see GAAP - Generally Accepted Accounting Principles) </li></ul><ul><ul><li>It is not possible to directly compare different quality initiatives </li></ul></ul><ul><ul><li>A common standard would still permit assigning different weights to different categories but in a much more transparent and comparable way </li></ul></ul>
  87. 87. Translation Quality Scale <ul><li>Quality Continuum </li></ul>
  88. 88. Translation Quality Scale <ul><li>Quality Grades </li></ul>TQI A E D C B 90 60 70 80 50 100
  89. 89. Select Bibliography • Brue, G. : Six Sigma for Managers, New York, McGraw Hill , 2000 • Deming, W. Edwards: Out of the Crisis, Cambridge (Mass), MIT Press, 2000 • Eckersley, H.: “Systems for Evaluating Translation Quality”, in Multilingual Computing & Technology, #47 Volume 13 Issue 3, April/May 2002 • Grove, A.: High Output Management, 2nd ed., New York, Vintage Press, 1995 • Hönig, H. : “Positions, Power and Practice: Functionalist Approaches and Translation Quality Assessment”, in Schäffner, C. (ed.) Translation and Quality. Clevendon, Multilingual Matters, 1998 • Language International: “Engineering Language Quality – A word with quality-standards consultant John Gagliardi”, in Language International Vol. 12 No. 3, June 2000 • Lauscher S.: “Concepts of Translation Quality and Quality Assessment”, in Proceedings of the 39th Annual Conference of the American Translators Association, 1998 • Ling Koo, S., and Kinds, H.: “A Quality-Assurance Model for Large Projects”, in Sprung, R. (ed.) Translating into Success. Cutting-edge strategies for going multilingual in a global age. Amsterdam/Philadelphia, John Benjamins Publishing Company, 2000 • LISA: “Microsoft Quality Standards”, in Case Studies and Client Requirements, 1998 • McConnell, S.: Software Project Survival Guide, Redmond, Microsoft Press, 1998 • Moroney, M.J.: “Facts from Figures”, Harmondsworth, Penguins, 1951, 1956(3rd) , • Reiss, Katharina: Translation Criticism - The Potential & Limitations. Categories and Criteria for Translation Quality Assessment. Translated by Erroll F. Rhodes. St. Jerome Publishing 2000 • Schäffner, C. (ed.): Translation and Quality, Clevendon, Multilingual Matters, 1998 • Shewhart, W. : Statistical Method from the Viewpoint of Quality Control. 1939. New York: Dover Publications. Reprint, 1986. (Originally published: Washington, D.C.: Graduate School of the Department of Agriculture, 1939.) • Spurr W., and Bonini C. : Statistical Analysis for Business Decisions, Homewood, IL: Richard D. Irwin, Inc., 1967 • Sturz, W.: “DIN 2345 Hits the Language Industry” in Language International Vol. 10 No. 5, May 1998 • Vogel, S.; Nießen, S.; Hermann, N.: “Automatic Extrapolation of Human Assessment of Translation Quality” http://www‑i6.informatik.rwth‑aachen.de/PostScript/InterneArbeiten/Paper_2000/Vogel_Evaluation_LREC_2000-corrected.ps.gz , 2000 • Woyde, R.: “Introduction to the SAE J2450 Translation Quality Metric”, in Language International Vol. 13 No. 2, April 2001
  1. A particular slide catching your eye?

    Clipping is a handy way to collect important slides you want to go back to later.

×