د/ايمان زغلول قاسم

استاذ تكنولوجيا التعليم المشارك بكلية التربية بالزلفي

book B81

6.2. Theoretical Framework
The Standards for Educational and Psychological Testing (American Educational
Research Association [AERA], American Psychological Association [APA], & National
Council on Measurement in Education [NCME], 1999) contain only a few general standards
about how to report test scores. These standards are not directly usable for the (re)design
process of score reports. In recent years, several studies have attempted to provide guidelines
for improving score reporting practices (e.g., Goodman & Hambleton, 2004; Hattie, 2009;
Ryan, 2006; Zenisky & Hambleton, 2012), which can be very useful for direct application.
Zenisky and Hambleton (2012) have summarised the literature on score reporting in
terms of best practices. They have outlined the recommendations based on research in five
areas: 1) report development processes, 2) report design and format, 3) contents, 4) ancillary
materials, and 5) dissemination efforts. This study focused on the first three areas, although
the authors acknowledge the importance of the fourth and fifth areas as well.
From the literature about score reports in educational and multimedia research, various
design principles can be extracted regarding the content, design, and format (many of which
have been summarised by Zenisky & Hambleton, 2012). Furthermore, Kosslyn (2006) has
distinguished eight principles for the design of visual displays, inspired by various research
disciplines, such as psychology. For this study, the principles from the educational literature
were integrated into Kosslyn‘s (2006) eight principles:
1. Principle of Relevance. The graph designer should bear in mind the particular message
the graph needs to communicate. Moreover, score reports should be designed for a
specific test purpose (Zenisky & Hambleton, 2012). Ideally, there is a good balance
between presenting essential information and details. According to Mayer (2001),
irrelevant information should be avoided whenever possible, since it causes
unnecessarily high cognitive load and distraction. A score report should contain all
information necessary for interpretation, as intended by the designer. This includes a
description of the test purpose, intended uses for the test and test scores, and
confidence intervals with accompanying explanations (Goodman & Hambleton, 2004;
Ryan, 2003). Furthermore, various strategies can be used to support interpretation,
such as anchor points, market-basket displays, and benchmarking (Jaeger, 2003;
Mislevy, 1998). These methods will help the user understand the test results in terms
of both content-referenced and criterion-referenced interpretations (Zenisky &
Hambleton, 2012). Reporting subscores can help indicate strengths and weaknesses,
and improve the formative potential of the score report. However, from a
psychometric viewpoint, reporting at a fine-grained level (such as attribute subscores)
can raise reliability issues, depending on the properties of the test items (Monaghan,
2006). Ryan (2003) has suggested reporting results at the finest level possible, at an
acceptable level of accuracy. What is acceptable in terms of reliability depends largely
on the degree of reversibility of the intended decisions based on the test results (Nitko
& Brookhart, 2007). Furthermore, Goodman and Hambleton (2004) have suggested
that a certain degree of personalisation (e.g., inserting the pupil‘s name) can help the
user connect with the score report.

الوقت من ذهب

اذكر الله


المصحف الالكتروني