The purpose of this paper is to analyze basic assumptions of a newly emerging standards-based and assessment-driven backward curriculum design model and to study effects of the instruction based on backward design on achievement. Wiggins and McTighe, ...
The purpose of this paper is to analyze basic assumptions of a newly emerging standards-based and assessment-driven backward curriculum design model and to study effects of the instruction based on backward design on achievement. Wiggins and McTighe, advocates of backward curriculum design model, argue that it benefits practitioners who are held accountable for identifying desired outcomes from national, state, and local content standards. The backward curriculum design and assessment model relies on a purposeful task analysis method that begins with breaking down big ideas of the unit of study into small pieces of content and simplified skills. In effect, this curriculum model as backward can be traced to Tyler's behavioral objectives model and to Bruner's early version of discipline-centered curriculum. The Underlying concept is structural in that the system of three development stages prescribes curriculum decisions at the classroom level.
For this experiment, 80 junior middle school students were selected and they were assigned at random to two experimental conditions, in instruction based on backward curriculum design or in instruction based on traditional textbook-centered design. Of the participants middle graders(20%) were excluded from the experiment so, upper 32(40%) and low 32(40%) were participated.
The following major conclusions were drawn on the basis of data analysis:
First, there were statistically significant differences between the group taken the instruction based on backward curriculum design and the other group taken the instruction based on traditional textbook-centered design in the advance of academic achievement.
Second, the interaction effect of the instruction based on backward curriculum design and academic achievement was not significant statistically.