د/ايمان زغلول قاسم

استاذ تكنولوجيا التعليم المشارك بكلية التربية بالزلفي

book B99

7.2.6 Implementation of DT: Aggregation Level, Assessment Methods, and Feedback
Loops
Aggregation level. DT concerns the assessment of the educational needs of individual
students. Because of the nature of the instruments used in DT, data should not be aggregated
to levels higher than the individual level (Rupp et al., 2010). Furthermore, DT is not meant for
comparing students to one another, but for promoting the learning and developmental process
of individual students.
Assessment methods. In order to make inferences about the problem-solving process during
an assessment, the assessment tasks should be designed to make possible valid inferences
about how the student‘s task behaviour relates back to his or her thinking. This inferential
chain stems from the empirical knowledge available from information processing theories,
cognitive psychology, and learning trajectories (Daro, Mosher, & Corcoran, 2011; Leighton &
Chapter 7
164
Gierl, 2007a; Verhofstadt-Denève et al., 2003). Based on theoretical assumptions and
empirical research, the items in an assessment have certain characteristics that are assumed to
elicit a response behaviour related to the learner‘s developmental stage (Leighton & Gierl,
2007a).
The degree to which the assessment results are indicative of the developmental stage
of a student is crucial to the quality of the assessment methods used in DT. Although
including more items with the same characteristics in the assessment will increase the
certainty of the inferences about related misconceptions, it will also make the assessment
process less efficient (Rupp et al., 2010). For example, if the aim is to identify an arithmetic
misconception, and a student makes an associated error on one item, it is possible that this
error is caused by something else than that particular misconception. However, when the
student consistently shows the same error on several items with similar characteristics, the
inference about the misconception becomes stronger. Nevertheless, choosing details over
certainty, in terms of test accuracy, is not problematic with short feedback loops, because the
latter provides the opportunity to redirect the decisions made. In this case, the stakes in terms
of possible negative consequences for the learner are relatively small (Rupp et al., 2010).
Moreover, to cope with this trade-off between grain size and certainty about
inferences, assessment developers in DT often consider the design of (computerised) adaptive
tests, meaning that the selection of the next item depends on the student‘s response to the
previous item (Eggen, 2004). Adaptivity offers the possibility to make the assessment process
more efficient; items can be chosen based on their content and difficulty, for example, to
diagnose a student‘s strategy choice. Sometimes these types of assessments are referred to as
dynamic assessments, which are usually embedded in a computerised adaptive learning
environment. This means that when a student cannot solve a task, he or she will receive a
minimally intrusive hint. In this way, the materials are used for both assessment and learning,
by providing diagnostic information about a student‘s learning needs and item-based feedback
(Stevenson, Hickendorff, Resing, Heiser, & de Boeck, 2013).
Feedback loops. Although DT has the potential to be used for retroactive, proactive,
or interactive formative assessment, it is primarily used retroactively (Crisp, 2012; Stobart,
2008). In dynamic assessments, DT is used interactively; learning and assessment are
integrated. When DT focuses on the assessment of prior knowledge to plan instruction, it is
used proactively. Finally, when DT is used to identify, for example, misconceptions or buggy
problem-solving strategies, feedback is used for remediation, resulting in a retroactive
feedback loop. Short feedback loops in DT are preferred because the learner‘s thinking and
use of problem-solving strategies are highly likely to change over time. However, delayed
feedback could still be effective when the change in the learner‘s thinking and the
development of new strategies cover longer periods of time. Thus, the length of feedback
loops should match the student‘s learning curve for the subject matter that is the assessment‘s
objective. A mismatch between the two might result in negative consequences, hindering the
optimisation of the learning process.

الوقت من ذهب

اذكر الله


المصحف الالكتروني