Three variables which affect students' achievement test scores and which can be manipulated prior to test administration are discussed: testwiseness, practice tests, and test practice. Findings were presented from literature reviews. Regarding testwiseness, (1) it can be described, measured, and taught, therefore, testwiseness exists; (2) testwiseness is only mildly related to general intelligence as measured by group intelligence tests and is probably a network of specific skills, not a general cognitive ability; (3) testwiseness increases with maturity and is unrelated to sex; (4) although testwiseness skills can be taught to students of all ages, using a variety of techniques, the affects of such instruction do not last long; (5) the differential effects of testwiseness instruction for students at various points along the testwiseness continuum is unknown and may vary with the type of testwiseness skill under consideration; and, (6) for some groups of students, testwiseness instruction improves the reliability and predictive power of certain tests. No research was found on practice tests' effects on student performance or on the reliability or predictive power of the associated test. Regarding test practice, it was found that taking one IQ test once for practice will improve IQ scores on a subsequent test given no more than two months later, and practice effects may last as long as ten months if the posttest is the same as or a parallel form to the pretest. (Author/GK)