Abstract
Assessment for Learning (AfL) is a process in measuring the learning outcome in students. Current practices in assessing the academic performance of students in most of the countries are still manual. It is based on the qualitative and quantitative feedbacks, obtained by expressed statement and marks, respectively. The issues associated with such assessment-practices are that it (a) lacks autonomy in students and the teachers to assess themselves for (1) better learning (ABeL) and (2) to learning (AtoL) with greater accuracy; (b) Self, peer and parents' involvements in the assessment process has often been underestimated, and (c) involved human bias while giving the qualitative and quantitative feedbacks. Given the background, this chapter attempts to showcase how various Artificial Intelligence (AI)-based solutions, such as Expert Control System (ECS)-based tutoring platform and Agent-based tutoring systems (AbS) can be used for the AfL, which in turn, improve ABeL and AtoL in students.
TopIntroduction
Assessment for learning (AfL) has been conceived as a series of formative assessments in contrast to summative assessment, which is conducted at one time (i.e., the final assessment) at the end of a teaching session. Michael Scriven coined the terms formative and summative assessments in 1967 to distinguish and explain the roles of evaluation processes during and after the academic session, respectively (Tyler, Gagne & Scriven, 1967). Benjamin Bloom and colleagues in 1971 suggested applying the same distinction to evaluate student learning, which in today’s education world is called as the ‘assessment’ (Bloom, Hastings, & Madaus, 1971; William, 2006). The role of summative assessment is just summing the achievement of students/learners after the end of the academic session (Bloom, Hastings, & Madaus, 1971; National Research Council [NRC], 2001; Sadler, 1989; Shavelson, 2006). Formative assessment enhances learning through giving active feedback to the students while the teaching still in process (Black & Wiliam, 1998a, 1998b; Black & Wiliam, 2003; Black & Wiliam, 2004; Black, Harrison, Lee, Marshal, & William, 2004; Sadler, 1989; Shavelson, 2006). Summative assessment is also referred to as Assessment of Learning (AoL), whereas formative assessment is termed as the AfL (Black &Wiliam, 2003; Broadfoot, 2008; Gipps & Stobart, 1997; Stiggins, 2002). It is important to mention here, that, Black and William’s (2004) research came up with three major findings on the overall assessment processes. These are as follows:
Key Terms in this Chapter
Assessment: It is defined as the process of evaluating the quality of learning among students. It is an integral part of classroom-based teaching.
Assessment for Learning (AfL): It involves teachers, peers, and parents using evidence about students’ knowledge, understanding and skills to reflect back on the lesson planning and delivery of the teachers throughout the teaching and learning process for clarifying students’ learning and understanding of subjects.
Summative Assessment: It is defined as the final evaluation of student’s learning at the end of an instructional unit by comparing against some standard or benchmark.
Agent-Based System: It is one of a class of computational models for simulating the actions and interactions among autonomous agents, which are independent of the trainers or creators. They can have dialogue among themselves, with the user through dialects.
Machine Learning: It is pivotal to artificial intelligence, which provides computer to learn the patterns in the data without being explicitly programmed. It is also termed as data mining, knowledge engineering, expert systems etc.
Formative Assessment: It is defined as a set of formal or informal diagnostic test or assessment procedures conducted during the learning process in order to modify the lesson-plan, teaching methods and learning activities in view of improving the student attainment.
Artificial Intelligence: It is defined as the theory and development of computer systems, which is able to perform tasks normally requiring intelligence involving visual perception, speech recognition, decision-making and translation between languages under cognitive domain.
Expert Control System: These systems emulate the decision making ability of a human expert for solving complex problems by reasoning about knowledge, represented by rules (rule-based reasoning systems) or cases (case-based reasoning systems) by using conventional procedural codes. It essentially has two parts – a knowledge base, which is composed of several ‘if-then’ rules of cases and an inference engine that deduces new facts (decisions) based on the old facts or rules or cases.