Abstract

This paper offers a conceptual framework on test design from the perspective of social semiotics. Items are defined as arrangements of features intended to represent information, convey meaning, and capture information on the examinees’ knowledge or skills on a given content. The conceptual framework offers a typology of semiotic resources used to create items and discusses item representational complexity—the multiple ways in which the semiotic resources of an item are related to each other—and item semiotic alignment—the extent to which examinees share cultural experience encoded by items. Since the ability to make sense of items is shaped by the examinees’ level of familiarity with the social conventions underlying the ways in which information is represented, unnecessary representational complexity and limited semiotic alignment may increase extraneous item cognitive load and adversely impact the performance of examinees from certain populations. Semiotic test design allows specification of optimal pools of semiotic resources to be used in creating items with the intent to minimize representational complexity and maximize semiotic alignment for the maximum number of individuals in diverse populations of examinees. These pools of semiotic resources need to be specific to the content assessed, the characteristics of the populations of examinees, the languages involved, etc., and determined based on information produced by cross-cultural frequency analyses, cognitive interviews, focus groups, and expert panels.

Highlights

  • Current views of assessment as evidentiary reasoning emphasize the importance of systematic approaches for determining the numbers, formats, and features of items or tasks that are to be used in assessing a given domain of knowledge (Martinez, 1999; Pellegrino et al, 2001; National Research Council, 2006; Mislevy and Haertel, 2007)

  • The term, semiotic test design should not be confused with the term, test design, which is typically used in relation to the technical properties of tests (e.g., Wendler and Walker, 2006; van der Linden, 2016) and the ways in which content is covered through item and population sampling (Gonzalez and Rutkowski, 2010)

  • Semiotic test design should be confused with universal design and universal test design, which refer to the set of basic principles and practices intended to ensure that the needs of diverse students are taken into account during the entire process of test development and to maximize accessibility to all examinees

Read more

Summary

INTRODUCTION

Current views of assessment as evidentiary reasoning emphasize the importance of systematic approaches for determining the numbers, formats, and features of items or tasks that are to be used in assessing a given domain of knowledge (Martinez, 1999; Pellegrino et al, 2001; National Research Council, 2006; Mislevy and Haertel, 2007). We know that individuals from different cultural backgrounds may differ on the level of attention they pay to focal objects or contextual and background information (Nisbett, 2003; Chua et al, 2005); that the relative frequency of some features of item illustrations vary substantially across different assessment programs (Wang, 2012); and that the extent to which item illustrations influence student performance on science items in international comparisons varies across high- and low-ranking countries (Solano-Flores and Wang, 2015) Among many other, these findings speak to the need for a perspective of test design that allows systematic, detailed selection, and examination of the features of items. The ability of individuals to make meaning of semiotic resources depends on the extent to which they share that encoded cultural experience According to this reasoning, a test item can be viewed as an arrangement of multiple semiotic resources used in combination with the intent to represent information, convey meaning, and capture information on the examinee’s knowledge or skills on a given knowledge domain. This set of experiences is assumed to be unique to each examinee, multiple individuals can be regarded as a cultural group when they share many cultural experiences

Items as Samples of Encoded Cultural Experience
TYPES OF ITEM SEMIOTIC RESOURCES
Language Resources
Episodic Interactive
Metaphorical Devices
Representational Devices
User Interface Elements
Defining Semiotic Test Design
Cognitive Load
Item Representational Complexity
Item Semiotic Alignment
Design Parameter and Categories
Value or Category Selected
Item Design Parameters
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call