Usually districts and state school boards make the decision about whether or note adaptive testing will be used. Adaptive testing is always technology based. Today we will dissect the concept for you...break it down into its parts then build it up into a unified picture for you at the end.

What's the difference between old paper and pencil testing and current adaptive testing? Standard tests have a question with an A B C D choice option. Usually a bubble sheet. Student has one chance of getting the right answer. Students can often use deductive reasoning to hone in on the most likely choice. The test is always the same for every person who uses it. If they wrote the test correctly the standard deviation is usually 80% on these traditional, pencil and paper, bubble sheet standardized assessments.

Adaptive testing is MUCH different beast! The answers are not as cut and dried as picking A B C D. (Screen share sample). Adaptive testing is very interactive, multifaceted and much more sophisticated than anything students have experienced in the past with pencil and paper standardized testing. During an adaptive test students will actually draw, and figure and write on the test itself. Paragraphs are written to answer the question and the computer does the scoring without any human interaction whatsoever. These adaptive tests required what is called a constructed response. Students look at a problem and try to solve it in steps. With first generation adaptive tests what set them apart from the traditional tests is that the test adapted to the student's response. The more questions answered correctly the more difficult the questions became. These first generation tests looked very similar to the traditional pencil and paper bubble sheet tests.

It is amazing to think that these tests can actually differentiate the assessment by student ability. Traditional tests could never be that discriminating. Adaptive tests give students an opportunity to more accurately demonstrate their real learning.

The reaction of students moving from traditional to adaptive tests? Students are now highly aware of where they stand and the performance level to which they need to be reaching. Sometimes this is good and sometimes this is not so good. Kids can learn to "game" adaptive tests especially the older they get. If I don't do well the first time they are tested they can really impress my teacher with the growth they demonstrate when they take it the second time.

With younger K-2 students when adaptive tests were first started the kids gave it an heroic try and really wanted to please their teacher. These tests were stressful for them. The more advanced students were not accustomed to being the student who has to take the most amount of time to finish the test. These students were accustomed to being the first one done. With adaptive tests putting more and more difficult test questions in front of the student who gets the last answer correct, the length and the difficulty of the test can get pretty demanding. Matt remembers being in the computer testing lab when kids were testing on first generation adaptive tests in Sevier School District. He witnessed kids finishing up their tests and the teacher showed them their score. Teachers gave kids lots of positive reinforcement in order to help kids feel good about the hard test and the results they achieved.

Preparation of students for taking these tests is important. They need coaching in the dynamics of how to manipulate and navigate the format of these tests. It takes a little time to wrap one's head around the format of the test. Do you drag an icon? Do you click an icon?

Selena piloted State Standardized testing some time ago. Originally the students (especially the younger one) had to be taught how to drive a mouse. Over the years kids have become more savvy and the learning curve is now not so steep. Many classroom textbooks are now online as are the labs associated with them so the learning that is done in the classroom sets the kids up for success in manipulating the tests. Interface for the students on the actual test may not be as much of an issue any more.

We like the idea that the tests are becoming more than just A B C D. These adaptive tests are more of a virtual experience for kids. They can assess more than what kids are learning in the classroom. They can assess things they learn at home from mom and dad. A child who can do simple multiplication in kindergarten did not learn that in kindergarten; he learned it at home from his CPA dad. A student can sometimes even reason through the question incrementally tackling more difficult steps by simply thinking and reasoning through things. These students will often score higher on an adaptive test than they score in the classroom grade level.

Adaptive tests have no time restrictions which in and of itself makes adaptive tests a different beast than when kids face timed tests---especially SAT and ACT tests used for entrance exams to college. Is it a drawback for students to have the majority of their testing experience focused on adaptive tests when there are such high-stakes attached to the more traditional pencil and paper tests like the ACT and SAT?

It is highly unlikely that SAT and ACT tests will ever go adaptive. The high-stakes attached to these tests make this testing environment stringently controlled for security purposes.

Students who suffer from test anxiety or process more slowly are at a disadvantage on a traditional timed standardized test vs the freedom they experience on an adaptive test.

Who are the pioneering states and what's the current state of Utah? Several years ago Idaho, Washington and Oregon were cutting edge. We're not sure what the Eastern states have done. Selena plays the devil's advocate. Some real pluses for adaptive testing....allowing more time and allowing students to think in more sophisticated ways.

The real issue surrounding adaptive testing currently is what is done with the results? How do we use that data? That's when things get really dicey. These tests were originally designed to measure INDIVIDUAL student progress. They were not originally designed to measure the effectiveness of a system; however, under current accountability pressures, teachers are facing the reality that these tests are being used to "score" whether or not individual teachers or schools are effective.

The trend is to now factor in the student growth component. It isn't just about how proficient a student in a particular subject area but more about how much growth did that student get out of the learning opportunities they received in school? This puts more pressure on teachers to serve the upper performing students. The student growth component will probably become a big piece of what we do when we start scoring school effectiveness in the future.

If students score poorly how would schools help? Tiered instruction, Intervention strategies and additional instructional opportunities are the first steps. Things such as tutoring, extra time with the teacher. RTI (Response to Intervention) plans. If a kid is really skilled then what happens? Can kids be moved to higher grades? Basically these tests give teachers the knowledge that they need to extend learning for these kids and even which specific areas need to be targeted. It is expected that there be some opportunities beyond grade level instruction. Matt remembers in high school the tension created when the gifted students were not accommodated. This is still a classroom reality today but at least it is a goal and an expectation that schools are expected to meet.

Novices like Andrew and Dean have some questions. What does grading look like? Is it percentage based? These tests, especially in the lower grades, are rarely used for grading purposes. In the secondary realm the grading will be based on the growth between assessments. Kids are graded based on growth. These adaptive tests aren't really conducive for grading...it's more of a measurement to gauge student progress in learning rather than an end result that can be concretely defined.

Are there some limitations or cautions you'd caution others against? YES! One of the biggest issues with adaptive testing and student growth component is the desire to rate teacher effectiveness based on student growth. This is coming at educators like a freight train. Caution! In a pure world where a student is highly motivated, deeply engaged in the test, and nothing else is going on in their life that may be an OK thing but when you have unmotivated kids under a lot of life stressors the test is not all based on what the teacher does in the classroom...and in many ways multiple teachers may be responsible for learning an individual student learns. There is also the home factor. Students need to take some responsibility for their own learning.

Dean took the adaptive test Matt sent us as a link. He thought it was very interesting to be able to manipulate the testing world. He wonders if this is a good thing. He's says he's so visual....when words are in front of him he doesn't test as well as when a picture is placed in front of him. How consistent or reliable can tests be when learning is measured in rather novel ways...like pictures?

Matt is wondering about the types of questions; specifically, where there is one question with multiple parts. Selena mentions that this type of question reflects if a student has complete or partial learning of a concept. The specific end result is not the point. Adaptive testing allows for that partial learning.

Does the test take into account how close the kid is? The student will see multiple questions on a single concept. Rarely will a student face one question that's make or break. The test will adapt until it can pinpoint. The test is probing for the level of understanding.

The more standardized tests like ACTs....if all this adaptive testing is being done is that a problem for kids being adequately prepared for the standardized testing. Test taking strategies. The two tests are completely different animals. By doing adaptive testing we're not preparing kids to cheat their way through to college? Are kids limited? Selena talked about her students taking a bubble sheet test today and the kids had never taken a test like this that was not computer based. Filling out a bubble sheet requires perfect, thoroughly colored circles not a mouse click.

Utah has chosen not to test K-2 students. Learning at these levels needs to be more of performance observation. It is too expensive for a state to test those extra levels. Students this young

Special needs accommodations? Computers are a better way of testing students who struggle. Accommodations for testing can be written into student's Individual Educational Plans. Sections of the test can be read to students and/or can be delivered editorially.