A new computer-based solution promises to streamline K-3 oral reading assessments, giving teachers valuable new data on student competencies while freeing up classroom time.   Not-for-profit NWEA recently unveiled MAP Reading Fluency, an adaptive, automatically scored 20-minute assessment of oral reading fluency, comprehension and foundational reading skills that can be administered in groups.   Early users give it high marks. “It’s huge in terms of giving teachers valuable and important data in a short amount of time,” said Daisy Rojas, principal of the Gardner Academy in San Jose, Calif.   The automated system responds to a long-standing complaint among teachers that reading assessments administered one-on-one take up too much time and yield too little information. “You end up with one piece of information, reading rate and the number of correct words read per minute. And you spend a tremendous amount of time to get there,” said Jennifer Knestrick, NWEA
senior product manager.   To automate the labor-intensive assessment process, the new product leverages the speech recognition and scoring technology EduSpeak, licensed from SRI International. The technology has been calibrated to recognize the idiosyncratic speech patterns of readers ages 4 to 8, including word and line skips, substitutions and long pauses.   NWEA has customized the technology further through extensive user testing. Developers asked teachers to score student reading one-on-one, and then fine-tuned the automated system to pick up on the nuances of younger readers’ patterns until the scoring outcomes were closely matched.   With an automated solution, teachers can now assess reading ability for an entire class simultaneously, rather than having to score students individually. They also may get a deeper understanding of student abilities.   Deeper Data
Unlike conventional assessments, which are fairly static, the computer-based system automatically adjusts the difficulty level of the reading sample up or down in response to students’ demonstrated abilities. “They will do an online assessment, much like a game. They sit at a computer or a Chromebook or an iPad for 20 minutes and they read out loud, and they get different levels of difficult depending on how they are doing,” Knestrick said.   This adaptive component delivers a new data point, in the form of an instructional reading level. As Knestrick describes it, a sentence like “the cat sat under the tree” is qualitatively different from “while we were out walking, we saw the strangest thing, a cat on top of a tree.”   “That increase in text complexity is very important for the teacher to know,” she said. “There is wide variation in the classroom and the teacher needs to know where the child is on that scale of text difficulty.”   The system then goes a step further, augmenting the basic fluency assessment with comprehension questions: What was the main idea? Why did the character do this? The software also looks to assess foundational skills. For those who can’t read a sentence, it will assess their grasp of word recognition and letter sounds.   With the new tool the teacher can glean outcomes at a glance. Color-coded outputs indicate proficiency on a rising scale of red-yellow-green-blue, and the system can automatically group students of like proficiency for easy reference.   Finally, the entire testing session is automatically recorded, giving the teacher the ability to revisit and review student performance.   “That is the thing teachers have latched onto more than anything, just being able to listen to the child reading independently any time, any place,” Knestrick said. “They can hear exactly where the child is struggling, they can hear how the child approaches an unfamiliar word. Those are valuable insights.”   More than just a replacement for conventional assessments, the NWEA system seeks to bring a new level of understanding to the table.   “Our goal is not just that the technology should replace the one-to-one human scoring, but that it should give a deeper look into the child’s reading,” Knestrick said. “We want to enable teachers to tailor their instruction to the specific children in their classrooms.”