SOFTWARE MAPPING ASSESSMENT TOOL DOCUMENTING BEHAVIORAL CONTENT IN COMPUTER INTERACTION: EXAMPLES OF MAPPED PROBLEMS WITH KID PIX PROGRAM
THE NEED
There is a growing consensus that traditional methods, such as standardized testing, criterion-referenced tests, and teacher-constructed tests fail to measure important learning outcomes (Shepard, 1989; Anderson & Bachor, 1998; Shavelson & Ruiz-Primo, 2000; Koker, 2001; Davies, 2001). Such tests provide little to indicate either the level at which a student understands or the quality of individual thinking (Nickerson, 1989; Slack, 1993; Raychaudhuri, 1998; Lee, 2002). They emphasize homogenized recall of memorized factual knowledge and procedures rather than unique, and highly differentiated reflection. Because external criteria, they typically emphasize standards which can be applied to typical students. Changing the way we assess will inevitable change how teachers teach and how students learn. Today new ways of thinking about learning call for new ways for monitoring learning.
Reform in school assessment builds from the vision that assessment can become the bridge for instructional activity, accountability, and teacher development. Romberg (1995) stated that if assessment results are used by the learners or teachers, then the assessment tools must be available in the classroom on a regular basis, weaving together instruction and assessment (p. 29). The content of tests influences teaching and learning processes. Teachers often "teach to the test" rather than emphasizing underlying concepts.
Skills are thought in the manner measured on tests rather than how they are used in everyday contexts. When tests require the recall of memorized information, students develop memorization strategies that tend to de-contextualize their knowledge, promoting compliant cognition (McCaslin & Good, 1992; Pettig, 2000; Dolan & Hall, 2001). In order to become capable, learners need experience in solving real problems and understanding complex tasks (Duffy, 1997; Linn et al., 2000). Shepard (1989) stated that assessments need to approximate real-life tasks and to reflect multiple perspectives and diversity-versus-singularity of problem solutions.
Another problem with traditional testing is that it tends to emphasize evaluation, or classification, as a primary goal (Hart, 1994; Wilson, 1995; Ayala et al., 2002; Yin et al., 2004). Since a primary goal of education is to promote students' thoughtfulness, the basic of concept of testing needs to change, not just the structure of the tests (Brown, 1989; Koretz, 1998; Sizer, 2001). Also, McLellan (1993) pointed out that assessment needs to be dynamic, and reflect every-emerging samples of the learner's progress. As a consequently, traditional testing strategies are often counter-productive for the solving of real-world problems (Collins, 1990; Yin & Shavelson, 2004).
Choi & Hannafin (1995) and Reese (2003) stressed that in order to be useful in promoting higher thinking skills, testing needs to shift from domain-referenced evaluations to student-center assessments. Student-centered assessment emphasizes the ability to diagnose and manage cognitive growth rather than to evaluate student achievement. They said that since assessment in situated learning environments emphasizes cognitive and learning processes, improvement of learning strategies, and higher-order thinking skills, assessment alternatives typically require varied evidence (Pettig, 2000). As a programmatic change is occurring, there is a need to align student assessment practices with curricular aims, instructional practice and performance standards (Black &Wiliam, 1998; Their & Daviss, 2001).
The development of problem solving involves students' efforts to overcome obstacles and attain goals (Stecher, 1998). It involves the orchestration of a large number of other processes toward this end (Siegel & Thier, 2002). How well students encode to form mental modes are among the key determinants of their success on many problems. As Siegler (1991) stated that their success also depends critically on the ability to integrate general and specific knowledge, and on their selection of the right process in the right situation. Choi and Hannafin (1995) pointed out that constructivists will have to develop ways of expressing what is to be accomplished that do not constrain learning outcomes as they feel specific objectives would. Without some idea regarding student outcomes, evaluation would be an empty exercise (Pisha & Coyne, 2001; Sizer, 2001; Reese, 2003).