Measuring language learning using standardized score scales and adaptive assessment engines

Patent metadata

Field	Value
Publication number	US-10497281-B2
Application number	US-201916396440-A
Country	US
Kind code	B2
Filing date	Apr 26, 2019
Priority date	Nov 8, 2016
Publication date	Dec 3, 2019
Grant date	Dec 3, 2019

Abstract

The present invention allows a language learning platform to calculate a user's overall language skill. This may be performed as a standalone test or before, during and/or after a language course. The language learning platform may perform a Pre-CAT process, a CAT process, a stage 2—fixed form process and a scoring and reporting process. The Pre-CAT process is designed to quickly get an initial estimate of the user's overall language skill. The CAT processes improves on the accuracy of the Pre-CAT process, preferably until the conditional standard error of measurement for the user's overall language skill is less than a predetermined number. The stage 2—fixed form process may use the user's overall language skill to select speaking, listening and writing assessment items of an appropriate difficulty for the user. The language learning platform may score the responses of the user, present the measured user's overall language skill to the user and save the user's overall language skill in a database for use in future tests.

First claim

The invention claimed is:

1. A system for adaptively assessing a language learning of a user configured to:
receive a user identification from the user using a user interface displayed on a client device;
search in a first database for a history of the user associated with the user identification;
estimate an overall language skill of the user based at least on the history in the first database of the user;
retrieve a random item pattern comprising a sequence of question types, wherein at least one question type is a multiple choice type of question and at least one question type is a fill in the blank type of question;
sequentially for each item in the random item pattern:
select a first next question type in the random item pattern,
select a first assessment item from a second database of a plurality of calibrated assessment items, wherein the first assessment item matches the first next question type in the random item pattern and matches the overall language skill of the user,
modify the user interface to present the first assessment item to the user,
generate a first score for a first response from the user to the first assessment item, and
update the overall language skill of the user and a conditional standard error of measurement for the overall language skill of the user based on the first score;

select one or more fixed forms for speaking, listening and writing based on the overall language skill of the user;
modify the user interface to present the one or more fixed forms for speaking, listening and writing to the user;
generate a third score for a third response to the one or more fixed forms for speaking, listening and writing;
update the overall language skill of the user and the conditional standard error of measurement of the overall language skill of the user based on the third score for the third response to the one or more fixed forms for speaking, listening and writing;
modify the user interface to notify the user of the overall language skill of the user;
record the overall language skill of the user to the first database; and
determine a future language lesson of appropriate difficulty for the user based on the recorded overall language skill of the user.

2. The system of claim 1 , further configured to:
before the estimate the overall language skill of the user, retrieve a difficulty level of the product being used by the user.

3. The system of claim 1 , wherein the overall language skill is expressed on a logarithmic scale.

4. The system of claim 1 , wherein the second database comprises the plurality of calibrated assessment items, wherein each assessment item in the plurality of calibrated assessment items is associated with a known difficulty level of the assessment item and a question type of the assessment item.

5. The system of claim 1 , wherein the first database comprises a record of every assessment item presented to the user.

6. The system of claim 5 , wherein the system does not present an assessment item in the plurality of calibrated assessment items to the user that was previously presented to the user.

7. The system of claim 1 , further configured to:
before the selection of one or more fixed forms for speaking, listening and writing, repeat the steps, at least once, of:
select a second next question type corresponding to a second language skill of the user,
select a second assessment item from the second database of calibrated assessment items, wherein the second assessment item matches the selected second next question type and matches the second language skill of the user,
modify the user interface to present the second assessment item to the user,
generate a second score for a second response from the user to the second assessment item,
update the overall language skill of the user and the conditional standard error of measurement for the overall language skill of the user based on the second score, and
compare the conditional standard error of measurement for the overall language skill of the user to a predetermined threshold,

until the conditional standard error of measurement for the overall language skill of the user is less than the predetermined threshold.

8. A system for adaptively assessing a language learning of a user configured to:
retrieve a random item pattern comprising a sequence of question types;
repeat the steps, at least once, of:
sequentially select a first next question type in the random item pattern,
select a first assessment item from a database of a plurality of calibrated assessment items, wherein the first assessment item matches the first next question type in the random item pattern and matches the overall language skill of the user,
modify the user interface to present the first assessment item to the user,
generate a first score for a response from the user to the first assessment item, and
update an overall language skill of the user and a conditional standard error of measurement for the overall language skill of the user based on the first score,

until the conditional standard error of measurement for the overall language skill of the user is less than a predetermined threshold; and
determine a future language lesson of appropriate difficulty for the user based on the recorded overall language skill of the user.

9. The system of claim 8 , wherein the overall language skill is expressed on a logarithmic scale.

10. The system of claim 8 , wherein the comprises the plurality of calibrated assessment items, wherein each assessment item in the plurality of calibrated assessment items is associated with a known difficulty level of the assessment item and a question type of the assessment item.

11. The system of claim 8 , wherein the database comprises a record of every assessment item presented to the user.

12. The system of claim 11 , wherein the system does not present an assessment item in the plurality of calibrated assessment items to the user that was previously presented to the user.

13. The system of claim 8 , further configured to:
before the selection of one or more fixed forms for speaking, listening and writing, repeat the steps, at least once, of:
select a second next question type corresponding to a second language skill of the user,
select a second assessment item from the second database of calibrated assessment items, wherein the second assessment item matches the selected second next question type and matches the second language skill of the user,
modify the user interface to present the second assessment item to the user,
generate a second score for a second response from the user to the second assessment item,
update the overall language skill of the user and the conditional standard error of measurement for the overall language skill of the user based on the second score, and
compare the conditional standard error of measurement for the overall language skill of the user to a predetermined threshold,

until the conditional standard error of measurement for the overall language skill of the user is less than the predetermined threshold.

14. The system of claim 8 , further configured to cause the processor of the hardware server to:
repeat the steps, at least once, of:
select a second next question type corresponding to a second language skill of the user,
select a second assessment item from the database of calibrated assessment items, wherein the second assessment item matches the selected second next question type and matches the second language skill of the user,
modify the user interface to present the second assessment item to the user,
generate a second score for a second

Assignees

• Pearson Education Inc

Inventors

• Booth David
• Seddon Paul

Classifications

• G06Q50/20

  Education · CPC title

• G09B19/06 Primary

  Foreign languages (with audible presentation of material to be studied G09B5/04) · CPC title

• G06Q10/0639

  Performance analysis of employees; Performance analysis of enterprise or organisation operations · CPC title

• G09B7/04

  characterised by modifying the teaching program in response to a wrong answer, e.g. repeating the question or supplying a further explanation · CPC title

• G09B7/08

  characterised by modifying the teaching program in response to a wrong answer, e.g. repeating the question or supplying further information · CPC title