Personalized Problems

Personalized Problems
File:Match problems to skill levels.png
Contributors
Last modification	June 19, 2015
Source	{{{source}}}
Pattern formats	OPR Alexandrian
Usability
Learning domain	General
Stakeholders	Teachers Students
Confidence
Evaluation	PLoP 2015 writing workshop Talk:ASSISTments
Application	ASSISTments
Applied evaluation	ASSISTments

If students have difficulty answering a problem, then give students problems that fit their level of understanding.

Context

Teachers use ASSISTments to select the problems in an assignment, to specify the sequence and conditions for presenting questions to students, and to assign the homework or activity to their students.

Problem

To ensure that students are challenged but not overwhelmed by the questions they are asked to answer.

Solution

Assign to students math problems that they have enough skills to solve.

Forces

1. Teachers assign the same homework to all students.
2. The ASSISTments interface provides teachers with simple control mechanisms (e.g., sequencing of questions in a homework, selecting questions based on the correctness) to control how questions are presented to students.
3. Problem definitions (e.g., difficulty, presentation, wording, sequence) affect students’ learning experiences.

Consequences

Benefits

1. Students do not get bored or too frustrated by the level of question difficulty
2. Students are more likely to complete their homework
3. Students master the skills being taught in the assignment

Liabilities

1. Teachers and content experts will need to create content and define pedagogies for handling differences in student skills.
2. ASSISTments will need to be modified to adapt content based on a students’ skill level.

Example

A teacher would create problems for a homework with different difficulty levels appropriate for students who show low, medium or high performance. As students answer questions in their homework, ASSISTments will keep track of their progress to classify them as low performance (i.e., student makes mistakes ≥ 60% of the time), medium performance (i.e., student makes mistakes < 60% and ≥ 40% of the time) or high performance (i.e., student makes mistakes < 40% of the time). Depending on students’ performance level, ASSISTments will provide them with the corresponding question type so it is more likely for students to receive questions that are fit for their skill level.

Evidence

Literature

Learners experience cognitive overload when they are given a task that is too difficult for them to accomplish. Experiments showed higher learning outcomes, more task involvement, and less effort when tasks were adapted to learner's skill levels .

Data

Data showed that students got bored or disengaged when a problem was too easy or too difficult. [1].

Related patterns

Problems can be personalized much like Content Personalization (Danculovic et al. 2001), but will heavily depend on a model of student knowledge.