Presentation is loading. Please wait.

Presentation is loading. Please wait.

DEPT, FERRARI AND MENDELOVITS: HOW TO ANALYZE AND EXPLOIT FIELD TEST RESULTS WITH A VIEW TO MAXIMIZING CROSS-LANGUAGE COMPARABILITY OF MAIN SURVEY DATA.

Similar presentations


Presentation on theme: "DEPT, FERRARI AND MENDELOVITS: HOW TO ANALYZE AND EXPLOIT FIELD TEST RESULTS WITH A VIEW TO MAXIMIZING CROSS-LANGUAGE COMPARABILITY OF MAIN SURVEY DATA."— Presentation transcript:

1 DEPT, FERRARI AND MENDELOVITS: HOW TO ANALYZE AND EXPLOIT FIELD TEST RESULTS WITH A VIEW TO MAXIMIZING CROSS-LANGUAGE COMPARABILITY OF MAIN SURVEY DATA COLLECTION INSTRUMENTS Washington D.C., March 23, 2012

2 State of the Art  Proliferation of multilingual comparative studies  Survey design includes a pilot, or Field Test (FT) carried out on a smaller scale  items adapted in multiple languages before FT  key moment for linguistic quality control (LQC) : right before FT – translation verification  Comprehensive documentation of adaptation, adjudication and validation processes

3 Between Field Test and Main Survey  In survey designs that include FT and MS, analysis of FT results = a wealth of information  Can be used to inform item selection  But also to perform a more focused linguistic and formal verification before MS  Open communication channels between item writers, national experts, verification coordinators

4 The PISA paradigm  Inception in 2000, currently 5 th survey cycle  Double translation, double source design  32 national versions (2000) -> 85 n. v. (2012)  from pencil and paper to computer-delivered assessments and background questionnaires  compiling data on adaptation history of each item in each language

5 Analysis of FT Results  At item level: item stats (itanals) Item discrimination Item fit Ability ordering Point biserial correlation (MCQ)  Differential item analysis gender country language

6 Multiple choice item: not dodgy Options A,B,C,D Key Answer Should be positive for key answer Should be negative for distractor Mean ability and standard deviation for the group of students who selected responses A, B, C or D Higher than 0.2 Item fit

7 Multiple choice item: Dodgy Less than 0.2 Low discrimination between high and low achiever Value significantly higher than 1 (item discrimination between high and low achievers is less than expected)

8 Action  Dodgy item reports sent to countries to cApStAn  reflect on the data, examine national version; explain why such results may have occurred.  As a result, FT to MS corrections proposed by: Item writers / test developers countries / verifiers

9 Dodgy item

10 MS version management  Base national MS version prepared for countries (using final FT version) segment status indicates type of action locked segments if no FT > MS changes  Country review followed by focused verification  Difference reports (before/after) generated automatically  Reports examined by referee  Final check on key corrections

11 CHALLENGES  Convincing reviewers / verifiers that if it isn’t broken, don’t fix it  Document each change with its justification  Check whether changes have not introduced new errors or inconsistencies  Make more systematic use of dodgy item reports, including for background questionnaires  Embed these processes in the platforms and IT adaptation management systems

12 ANY QUESTIONS? THANK YOU steve.dept@capstan.be


Download ppt "DEPT, FERRARI AND MENDELOVITS: HOW TO ANALYZE AND EXPLOIT FIELD TEST RESULTS WITH A VIEW TO MAXIMIZING CROSS-LANGUAGE COMPARABILITY OF MAIN SURVEY DATA."

Similar presentations


Ads by Google