Clearing Up Testing Misconceptions

29 Apr 2015

As someone who deals daily with questions and concerns about testing homeschoolers, I have encountered a number of misconceptions and some erroneous information concerning testing of homeschoolers. I hope that this article will clear up some of these misunderstandings for the homeschool teacher!

First of all, it is the parent’s responsibility to follow the law in their state. Just because a testing company offers a particular test does not mean that it will qualify for meeting the requirements of our NC homeschooling law. One big misconception is that survey tests are acceptable in NC, but most survey tests do not test spelling and grammar, which are two areas required in NC. Most of the Survey Plus tests have added these subject areas, but it is your responsibility to check. The Woodcock-Johnson III and IV tests are both survey tests. These tests are administered professionally, but the information you receive, especially in math, is very limited because they are survey tests. For example, with the Math Calculation subtest, there are only two to three problems per grade level. A survey test gives you a sampling of the areas tested, and you have to be careful in using the results of a survey test for curriculum decisions. Math, especially, has too many building blocks, and you are not testing many of those skills. If your desire is to use your scores to plan your instructional program, more information is obtained from using a complete battery test. With any of these tests, the science and social studies tests are optional in NC.

As the teacher, you make the decision about what grade level testing you want to use with your student. It is important to keep in mind that if you chose to re-enter a public or private school, your grade level recommendation does not have to be accepted by the school. The principal will make that decision, and they often will do their own testing for placement. Generally, public educators are not very trusting when parents are allowed to do their own testing, as is allowed in NC.

It has been reported, erroneously, that some tests are aligned to the Common Core Curriculum. The Iowa tests, Form C, which have been recently released for homeschool use are one example. Form C of the Iowa Test is a parallel form to Form A, which has a number of outdated questions that are irrelevant to students in today’s world. (The most famous question from Form A that has frustrated parents is the question about the card catalog.) Both Form A and Form C use the exact same norms of 2005. In order to use the same norms, Form C has to have the exact same subtests, the same number of questions and the question types cannot be changed. Since the Common Core Curriculum was written in 2010, it is impossible for the publishers to develop a test aligned to the Common Core and use the older 2005 norms.

The California Achievement Test publishers have done the same thing with their TerraNova-First Edition and the TerraNova-Second Edition (also known as the CAT/6). Both of these tests also use their norms from 2005. The second edition is only an update of some question content and pictures, to make it more relevant to students today. It would be impossible for the TerraNova/CAT 6 to be aligned to the Common Core since it uses norms from 2005.

The CAT/5 is still allowed in NC, though many states will not accept tests that were not normed in the last ten years. I used this test when I was a teacher in the 70s, when this test first came out. There was a kindergarten test, but kindergarten was not part of the public school system in many states. It is important to keep this in mind because the first grade test from the CAT/5 is more similar to the kindergarten tests that are used today. How many public school parents have you heard say kindergarten is what first grade used to be when they were in school? Curriculum has definitely changed over the last forty years since this test was published. There is a much greater emphasis on the higher order thinking skills in the later curriculum guidelines because of the emphasis of preparing our students for the technological world they will be working and living in. This test meets state guidelines in NC, but the information you receive will be limited because of the age of this test. This test was renormed in 1991, so if you have a high schooler taking this test, that student is compared to students who are forty years old or older now.

The ACT college entrance exam/achievement test does count in NC for homeschool guidelines. The SAT college entrance exam is a cognitive abilities test, used to predict student success in college, so it does not count as an achievement test unless you take the subject area tests that meet NC guidelines.

When looking at your score report, keep in mind that none of these tests are pass/fail. This decision needs to be made by the teacher using information from the total school year. The purpose of these tests is to meet state requirements and/or to give you information about your child’s strengths and weaknesses so that you can gear your instruction to the meet the needs of your student. The most helpful score for you is the national percentile ranking of your student in each subject area. If your student scored at the eightieth percentile, then that tells you that your student scored as well as or better than 80% of the students who took the same test in the year that particular test was normed. When comparing your scores year-to-year, a year’s growth would be indicated with approximately the same percentile score. GE (Grade Equivalent) is the most confusing and misunderstood score on the report and has limited meaning. Keep in mind that your student is only being compared with other same grade students in the same grade normed sample. The students are not being compared to other grade level students. If your student is in third grade and scores a GE of 6.8 in math, it only means that your student scored as well as a student in the eighth month of sixth grade would score on that same third grade test. There are no sixth grade math skills on a third grade test, so your student is not being evaluated on skills other than the skills of a third grader. Some grade level material is included that is slightly above and below the level of the test.

What makes a test standardized is that the directions are exactly the same for all students taking the test. Time limits must be followed or your results will not be valid. If your child has a diagnosis of a specific learning challenge, then you may make modifications to the testing administration. A note should be made as to what modifications are made, and this should be included on the child’s score report. One modification that is never allowed is reading a reading test to a student. This would make the test a listening test, so the child has to do the best he can with the reading. Extended time is an allowed modification for a child who has a diagnosis. You are allowed to choose a lower grade level test that is more appropriate to the level of your student. Your scores compare your student to students who took the test with the time constraints, so this must be considered if modifications need to be made. If time restrictions are not followed with students that do not have a diagnosis, then your testing is not standardized and your results are invalid.

The last misconception on my list is that we don’t need to worry about being comfortable with our homeschooling law in NC. As someone who deals with testing across the United States, I am seeing considerable tightening of homeschooling laws in many states. It takes only one legislator in NC to get the ball rolling towards tightening our present homeschooling law. We need to continue to be involved advocates for our homeschooling rights in NC. We must not allow ourselves to become too comfortable, or we may lose some of the benefits of our present homeschooling situation in NC.

GREENHOUSE is NCHE's flagship publication. 

GREENHOUSE magazine is published fall and spring plus an annual graduate issue in May. GREENHOUSE is mailed to NCHE members.

Subscribe

Comments

Stacey's picture

So, since we are required by law to do Standardized testing yearly - which would you suggest we use? Please email me at gandsmaddrey@embarqmail.com

Diane allen's picture

You are free to choose any of the tests that qualify as nationally normed and standardized achievement tests. There was an excellent article by Debbie Thompson a year ago (which is linked in the archives of this site) which listed all of the commonly used tests and their particulars so that parents could make an informed decision. You can also consult the testing information on the NCHE main webpage. You know your children and your homeschool better than anyone and should choose the test which works for you.

Serena Manning's picture

This article is a bit misleading about the Woodcock Johnson test. Even the author administers the WJ herself. The Woodcock Johnson ABSOLUTELY meets state testing requirements. It is a highly valid and reliable test. The Basic test is called "standard battery" by the publishers. It covers reading, writing, and math (including spelling and grammar). I do not find the information from the math scores to be limited at all. Math Calculations and Applied Problems (word problems), as well as math fluency, are on the standard battery. Most moms tell us that they after our consultation, they know what instructional choices to make for the next year. Serena Manning, Lighthouse Educational Services www.lighthouseedu.net info@lighthouseedu.net

Pat Brewer's picture

If you read the article, you will see that I never said the WJ was not accepted in North Carolina. NC does accept this test and I have no problem with that. I administer the WJ test to students myself. I said it is a survey test. When you can test a subject area in 10 min., then you are doing a brief sampling of skills in that subject. That is what a survey test is. The reading tests are pretty reliable in determining the student's instructional level in reading. I believe that you need to take the math scores with a grain of salt. The math calculation and math problem solving tests only give you about 2 problems per grade level, so you are only giving a very brief sampling of math skills and, in my opinion, are not testing enough of the math building blocks to determine curriculum placement. Math is a building block subject and the student will run into problems if you leave holes. Some students score high on the test because they have been taught some higher level problems early. The test will not show whether they have mastered the math skills on their grade level. The test has been around for decades, so publishers are aware of which kinds of higher level math problems to throw into their books. I saw this first hand with 2 different math curriculums that I used for my homeschooling. My concern about the WJ math test has come from personal experience as a tester, where parents have come to me after being given the advice to jump their student to the GE score based on WJ testing. Each one of these students was struggling in math and before they took this advice, they were very successful. The jump resulted in serious gaps in their math instruction. These parents had to back up and regroup to get their students back on a good math track. The WJ III has 22 subtests for Forms A and B. The Brief Form has 8 subtests that are parallel to 8 of the subtests in the longer forms. Its original purpose was to briefly test using the standard battery and see if there were areas that needed to be tested more in depth to determine learning issues. (More in-depth testing is the purpose of the other 16 tests.) It is a nationally standardized achievement test, so meets our state requirements. We are fortunate to have choices in NC. This test meets the needs of many homeschooling families and is a popular choice. I just like to see that folks have the information, if they want it, to make informed choices.

Serena Manning's picture

This is why it is important for a trained consultant to explain those scores to parents. If parents were advised to jump their student to the GE based on the WJ testing, that is a weakness of the testing consultant, not the test itself. I feel the math percentiles are very accurate on the WJ. Fortunately, the WJ requires that professional evaluation of the scores, where other test scores are sent to parents in the mail. My consultation might sound like this: "The Grade Equivalency is showing what an average 9th grade student can EASILY do in math. This does not mean that your son has MASTERED math up through 9th grade concepts. Look at the percentiles and you will see that your child does math better than 85% of students in 6th grade. I noticed he struggles a little with decimals, so you should review those this summer. Your son will easily transition into 7th grade math next year." The flow of your second paragraph just comes across a bit confusing and people will think the WJ is not accepted in NC: "Just because a testing company offers a particular test does not mean that it will qualify for meeting the requirements of our NC homeschooling law. One big misconception is that survey tests are acceptable in NC, but most survey tests do not test spelling and grammar, which are two areas required in NC. Most of the Survey Plus tests have added these subject areas, but it is your responsibility to check. The Woodcock-Johnson III and IV tests are both survey tests."

Diane Allen's picture

Thank you for clarifying your point Pat. As you may see from the many questions and comments on the NCHE Facebook page (where most discussion takes place) the wording in your original paragraph did lead many to the wrong interpretation of your intended message.

The bigger message it seems to me is to beware of incorrectly applying any standardized test scores. I agree that the math score has always needed to be read with a grain of salt and a bit of insight into the daily math processes as well as the material particular curriculum covers. This is not to say the WJ Math subtests are not valuable though since the test administrator can make observations on processes which would be lost in another testing situation.

On the whole, as you point out, it is incorrect application to skip grades based on GE scores on any standardized test. www.dalleallenassociates.com dianeallen@dallenassociates.com