Even though this was our third full year of homeschooling, it was our first year of official reporting since my son was in first grade. At the end of the year, Georgia statutes require that I write up a summary of his academic studies and progress - and as I sat down to do that, I decided I'd like to give him a standardized test just to get an idea of how he's doing compared to other kids his age. Christian Liberty Press offers the California Achievement Test (1970 version) online for $25, so that's what we did this afternoon.
It provides immediate results, and I'm feeling very gratified about my decision to homeschool after seeing these:
|SUBJECT||Raw Score||Grade Equivalent||Percentile Rank||Stanine|
|Mathematics Concepts & Problems||43||3.4||90||8|
|Language Usage & Structure||18||5.1||95||8|
Explanation of the California Achievement Test ScoresSeveral different numbers are listed on your California Achievement Test results. Most of these are calculated based on a norm group (the original group of students tested on which the test is standardized). Results are listed by academic area. Below are basic definitions for these test scores.
Raw scores are the actual number of correct answers within a given test section, and are used to calculate the Grade Equivalent, Percentile, and Stanine results. They should not be used for any kind of comparison or statistical calculation since each test section varies in the difficulty and number of questions.
Grade Equivalent Scores
Grade equivalent scores represent the grade level (year and month separated by a decimal point) of a student's ability compared to the median score of students at the same academic level.
Scores that are above a student's grade level do not by themselves indicate proficiency to skip to the next grade level in the particular academic areas. These scores more likely show an understanding of only some of the concepts of that higher grade. Conversely, scores below a student's grade level may indicate a gap of understanding in those academic areas. Scores equal to a student's grade level are considered the minimum for grade level ability.
Percentile scores are equivalent to the percentage of students from the norm group who received lower scores than the student. For example, a student with a percentile score of 73 means that the student did better than approximately 73% of the students in the norm group. Percentile scores do not represent the number of questions answered correctly.
Percentiles are useful for comparing a student's performance over several sections of the test. However, because they are not measured on an equal scale of units, they are not suitable for most statistical calculations.
Derived from the term STAndard score from a NINE-unit scale, stanines are based on the mean of the norm group (who are given the score of 5) and a standard deviation of 2.0. Scores are determined by the amount of standard deviation from the norm group, in approximately one-half standard deviation increments.
The following table lists the meaning of each score:
|9 Highest Level||6 Slightly Above Average||3 Well Below Average|
|8 High Level||5 Average||2 Low Level|
|7 Well Above Average||4 Slightly Below Average||1 Lowest Level|
My son has no previous experience with standardized tests, and he doesn't get any computer time per se so I sat with him while he did the test, but I tried my best not to prompt him at all. I did have to explain that it's a timed test, so if he's not sure how to answer a question he should move on and come back to that one later; and also that if the answer he thinks is correct isn't one of the four offered, then he needs to select another answer. Overall he did very well - I did find some weak spots (namely, he doesn't know the value of currency even though we did touch on that this year), and I think he would have done a little better if he'd paid more attention to detail - but I'm happy about these results overall.
At the risk of sounding like an intellectual snob, both my husband and I are pretty intelligent people, and we both did well in school (and on standardized tests) so I knew the kids would likely be smart too - but it's still gratifying to have some hard evidence to back up my assumptions!