Word to the Middle School...

In regard to the Gates "practice" Test given to the students recently...

Is it ever appropriate to use the actual test forms (those used in the

current year or subsequent year) for test preparation?

No, providing students with test items or test answers in advance of the test

is highly unethical. Such activity puts the focus on getting particular test

questions right rather than on measuring student achievement in the subject

area represented by the questions. It is highly rare that a given test item is so

important that its content should be learned by all students or taught to

students directly. The questions on a test represent only a small sample from

all the questions that could be asked when measuring achievement in, say,

science or math. When the exact test is the focus of instruction, the test scores

lose their meaning, and they portray an achievement result that is dishonest.  Wink, wink!

Taken from http://www.education.uiowa.edu/itp/documents/ITPGuidanceDocument.pdf

Use of Practice Materials and "Test Prep"

In August 2005, Iowa Testing Programs distributed a document to all Iowa districts and schools to provide information about appropriate test use and test preparation activities. "Guidance for Developing District Policy and Rules on Test Use,Test Preparation, and Test Security for the Iowa Tests" also is available on this website. Questions and issues about use of practice materials or preparing students in test content should be addressed by referring to the relevant sections of this guidance document.

Each test in the ITBS and ITED batteries begins with one or more sample questions to give students some practice with the content and format of the test. Extensive experience with testing students indicates that no further practice is needed for most students to understand what to do. Furthermore, the sample items represent the only kind of "practice" that students in the norm group received.

There is disagreement among educators about what kind of practice is legitimate preparation for administering a standardized achievement battery. The controversy has centered on the question of what type of practice might give assistance to test takers in excess of that afforded the students in the norm group. That is, to what extent do certain pretest activities offer an advantage to test takers, causing their scores to represent higher achievement levels than they have actually attained?

The issue of legitimate practice is important because it relates directly to both the integrity and the interpretation of standardized test scores. At the heart of the matter is generalizability. The questions in each test are a sample of the many questions that could be used to test students in a particular content domain -- reading, language skills, math, etc. Of course, teachers are interested to some extent in how students in a class answer particular questions, but they are far more likely to be interested in what students' responses tell them collectively about the students' reading achievement or math skill levels. If students get advance practice on the actual test questions, or on very similar ones, the ability to generalize beyond those questions becomes restricted. In sum, the more the practice questions resemble the actual questions, the less the user can generalize about what the students are able to do or what they know.

The problem of generalizability can be illustrated easily by considering the weekly spelling test usually given to elementary-grade students. Suppose that students are given 20 words to learn to spell by the end of the week and that Jeremy gets 16 words right on the Friday test. In interpreting Jeremy's score, it would be foolish to generalize about his spelling ability because he was given the 20 words to learn in advance. However, if another set of 20 words were given to Jeremy on that Friday, words that he had not known would be tested, his score on that 20-word test would tell us something about Jeremy's spelling ability in general. As can be seen from this example, if students are given practice ahead of time on the spelling words on a standardized test, then it makes little sense to use their scores to generalize about how well those students can spell the words on a long, unseen list. Of course, the same principle applies to reading passages, math problems, or any other tasks that appear on a particular standardized test.

Commercially Developed Test Preparation Materials

There are various test preparation materials on the market that claim to ready students for standardized achievement testing. Thus far, however, there is no consistent research evidence to support the use of such materials. Most of these materials are designed to raise test scores without increasing student achievement in the long term. Such preparation actually serves to misrepresent what students know rather than to demonstrate real, substantial growth over the previous year. The authors and publisher of the Iowa tests do not endorse the use of such materials.

Taken from http://www.education.uiowa.edu/itp/itbs/itbs_use_prep.aspx