| 
          
          The assessments 
          included 22 tasks that involved students in silent reading to obtain 
          information, answer questions and make decisions. 
           
          Sixteen of the tasks were identical for year 4 and year 8 students. 
          The remaining six tasks were given only to year 8 students. 
           
          Four tasks are trend tasks (fully described with data for both 2000 
          and 2004), eight are released tasks (fully described with data for 2004 
          only) and 10 are link tasks (to be used again in 2008, so only partially 
          described here). The tasks are presented in that order, with tasks for 
          year 8 students only following tasks for students at both levels. 
           
          When results for year 4 and year 8 students in 2004 are compared, it 
          is clear that year 8 students demonstrated consistently higher levels 
          of reading comprehension than year 4 students. Averaged across 147 components 
          of 16 tasks, 21 percent more year 8 than year 4 students succeeded with 
          the components. Year 8 students scored higher on all except one  
        component. The margin 
          was greatest for two tasks involving scanning for information under 
          a time constraint (Toyworld and Link Task 11), for which the margins 
          were 28 and 39 percent. As was the case in the 2000 assessments, many 
          of the students did not appear to be very efficient at scanning for 
          information. 
           
          Four years ago, we reported similar trends in performance on comprehension 
          tasks to those found on oral reading tasks: very substantial gains between 
          1996 and 2000 at year 4 level and small gains at year 8 level. This 
          time, the trends for oral reading (Chapter 3) and comprehension (this 
          chapter) do not match. Averaged across 27 components of three trend 
          tasks in this chapter, year 4 students performed at the same level in 
          2004 as in 2000. For year 8 students, with 31 components of four trend 
          tasks included, on average three percent fewer students succeeded with 
          task components in 2004 than in 2000 – a small decline. 
           
          It is not clear why the trends in Chapter 3 (continued improvement at 
          both levels) and this chapter (no change at year 4 and a small decline 
          at year 8) do not match. One possibility is that schools have been giving 
          greater focus to improving oral reading than to improving comprehension. 
            
           
                  |