Credit-granting institutions such as colleges, school boards, and Manitoba's learning centres tend to favor standardized assessments over diagnostic and competency-based assessments (see Figure 3). The term "standardized" means that the tests are always expected to be administered and scored the same way. The majority of standardized tests are "norm-referenced." This means that the test compares a student's scores to the scores of a representative or norm group.
The CAAT is ranked as the most frequently used standardized assessment across delivery agencies. The CAAT measures an adult's current functional level in math, reading, measures an adult's current functional level in math, reading, and language. It consists of a battery of nine sub-tests: reading vocabulary, reading comprehension, spelling, number operations, problem solving, mechanical reasoning, language, science, and study skills. There are four levels of CAAT, with each level corresponding to years of education and featuring a different combination of sub tests. Level A, for instance, is for adults who have completed 1 to 3 years of formal education, while Level D is for adults who have completed 11 to 12 years of formal education. CAAT, which was normed on adults, provides norm-referenced and content-referenced scores.2
A total of 83 (22 percent) respondents reported that CAAT is one of the tools used for is one of the tools used for initial assessment. Of the 83 respondents, 54 stated that CAAT was the was the most frequently used tool for initial assessment. The findings clearly indicate that the 54 respondents do not always agree about CAAT's strengths and weaknesses.
Many respondents appreciated the fact that CAAT provides grade equivalency and stanine provides grade equivalency and stanine scores. The grade equivalency scores were used to determine placement in classes, while the stanine scores were used to determine areas of strengths and weaknesses across a range of domains.
2 It should be noted that the standard definition of a reliable test is one that yields consistent test scores over time and different test situations, so many respondents were introducing a slightly modified definition of reliable.