Educational Tests and Assessment Resource Guide

Achievement tests are used in diverse contexts to measure the degree to which examinees can demonstrate acquisition of knowledge or skills deemed to be important. The contexts range from teacher-made testing in elementary and secondary school settings to high-stakes testing for college admission, licensure to practice a profession, or certification. The design of achievement tests varies depending on whether the inference intended to be drawn regarding examinees' performance is the absolute or relative level of mastery of specific knowledge and skills.

he term aptitude test has been traditionally employed to refer to tests designed to assess the level of development attained by an individual on relatively homogenous and clearly defined segments of ability, such as spatial visualization, numerical aptitude, or perceptual speed. Aptitude tests measure the effects of learning under the relatively uncontrolled and unknown conditions of daily living. In this sense, they differ from achievement tests that measure the effects of a relatively standardized set of experiences encountered in an educational program. The two types of tests differ in use as well. Achievement tests generally represent a terminal evaluation of an individual’s status on the completion of training.

The Code of Fair Testing Practices in Education (Code) is a guide for professionals in fulfilling their obligation to provide and use tests that are fair to all test takers regardless of age, gender, disability, race, ethnicity, national origin, religion, sexual orientation, linguistic background, or other personal characteristics. Fairness is a primary consideration in all aspects of testing.

Criterion-referenced testing is a method for examining a person’s performance with respect to a standard or criterion. It is commonly contrasted with norm-referenced testing, in which a person’s performance is compared with that of other persons who make up a norm group.

The cultural test bias hypothesis is the contention that racial and ethnic group differences in mental test scores are the result of inherent flaws in the tests themselves. These flaws bias, or cause systematic error, in a manner that causes ethnic minorities to earn low scores. Mean differences in scores among groups are then interpreted as artifacts of the test and not as reflecting any real differences in mental abilities or skills.

Intelligence can be defined as the ability purposively to adapt to, shape, and select real-world environments. However, most investigators of intelligence want to go beyond simple dictionary-like definitions to a deeper understanding of the construct.

Intelligent testing is a philosophy or model of assessment widely espoused and best represented in the writings of Kaufman et al. (Kaufman, 1979, 1994; Kaufman & Kaufman, 1977; Kaufman & Lichtenberger, 1999; Reynolds & Clark, 1982; Reynolds & Kaufman, 1986). The intent of the intelligent testing model is to bring together empirical data, psychometrics, clinical acumen, psychological theory, and careful reasoning to build an assessment of an individual leading to the derivation of an intervention to improve the life circumstances of the subject.

Differences in human behavior are overwhelmingly cultural not biological as anthropologists have understood since at least 1910. Known biological controls of subtle, sophisticated behavior (within the normal range) are trivial. Behavior patterns rarely match the visible biological variations or putative race categories

Measurement is the assignment of numbers to observed behaviors or actions. The observation may be made by humans or by machines. The behaviors may be simple and discrete such as movement of a child’s head, or may be complex such as a conversation between a therapist and patient. The measurement of simple or complex behaviors is based on a rule for assigning numbers; the degree of inference is important. Inference is the amount of interpretation used by a human observer in assigning the number associated with the behavior. The rule structure of the assignment may vary from one measurement to another, and the defining characteristic is in the quality of the number.

Since the Binet Scale (a rudimentary test that arranged questions and tasks according to degree of difficulty) was imported from France and tried on “feebleminded” children at Vineland Training School in 1908, Americans have made intelligence tests their own. Testing constituencies and controversies have proliferated.

Norm-referenced tests (NRT) refer to a broad array of standardized tests, the results of which are interpreted by comparing the performance of examinees with that of a specified population of individuals, or norm group. In a broader sense, NRT often refers to a type of test frequently defined by contrasting it with criterion-referenced tests (CRT), where test results are referenced to a particular content domain and provide information about the skills an examinee has acquired, not the rank of an examinee in a norm group (Anastasi, 1982).

Standards and guidelines in assessment cover three related areas: the people who use tests, the tests the people use, and the process of assessment within which tests are used. Both nationally and internationally, all three of these issues have been addressed in various ways.

The Psychological Corporation was the world’s oldest and largest commercial test publisher. It was founded in New York City in 1921 by three noted professors from Teachers College of Columbia University: James M. Cattell, Edward L. Thorndike, and Robert S. Woodworth. Over its 65-year history, the corporation’s primary mission has been the application of principles of psychology and measurement to the solution of educational, clinical, industrial, and social problems.

The term “psychometrics” refers in psychology to the science of assessment - more particularly the assessment of psychological characteristics of people. Sometimes, the term is used narrowly in connection with the development and use of paper-and-pencil tests of, for example, intelligence and personality, although other forms of individual measurement/assessment are embraced by the term.

Broadly speaking, reliability refers to the dependability of measurement (see s. 1, measurement and statistics). More specifically, the index of reliability indicates the accuracy of measurement, which is given by the correlation between the observed scores on a measurement instrument and the hypothetical true scores on the construct. T

School psychology is one of the four original specialty areas of professional psychology recognized by the American Psychological Association—specifically, organizational psychology, clinical psychology, counseling psychology, and school psychology. While there is a common science of psychology undergirding all four of these professional areas, school psychology focuses on the application of social and organizational, educational and learning, child and adolescent, normal and abnormal, and biological and ecological psychology to the school and schooling process.

The standard deviation is a measure of the dispersion of sample or population scores around the mean score. It is the most important and most widely used measure of dispersion for quantitative variables when the distribution is symmetric

Introduction Children in US schools are amongst the most heavily tested in the world. It has been estimated that 41 million US students take 127 million tests a year – more than three a year for every student. Tests are used to assess students' suitability for higher education, to allow them to progress from one grade to the next in school, and by administrators to judge how school districts, individual schools, and teachers are performing.

Tests are biased when irrelevant or systematic factors skew performance or scores for all or some of the test takers. Bias can result from numerous factors including the social context, question or item context, and setting of test administration.

Validity concerns the extent to which a measurement (see s. 1, measurement and statistics) instrument measures what it is intended to measure. It is not a dichotomous (valid versus invalid) or quantitative (more versus less valid) notion, but rather a context-specific one. Thus the same instrument may show evidence of validity in one situation, but not in another. There are many different types of validity.

The Bilingual Verbal Ability Tests (BVAT; Munoz-Sandoval, Cummins, Alvarado, & Ruef, 2000) are designed to provide a measure of overall verbal ability and a unique combination of cognitive and academic language abilities for bilingual individuals.

The Brigance Inventories are a comprehensive set of individually administered criterion-referenced tests. There are several inventories, each covering a specific age range, that are used for assessment, diagnosis, record keeping, and instructional planning. The following paragraphs briefly describe the intended use of each of the inventories.

The draw-a-person (DAP) is an assessment technique used with both children and adults for a variety of purposes. Harris (1963) provided a set of instructions, a scoring system, and norms for using the technique as a measure of children’s intelligence. The test has also been widely used as a projective personality assessment technique following a suggestion by Machover (1949). Although specific instructions vary, the examinee is typically asked to draw a picture of a person.

Informal reading inventory (IRI) is a generic term that refers to some type of nonstandardized technique used to assess aspects of reading performance. According to Smith and Johnson (1980), the most informal application of the diagnostic method involves having the child read selections of material silently and orally, asking comprehension questions about what has been read, and making note of the quality of reading, particularly word identification errors. Typically, the reading selections cover a range of grade or difficulty levels, from the preprimer through the eighth or ninth reader levels, as found in basal reading series. The comprehension questions based on the reading passages are usually of three types: factual or literal recall, inferential thinking, and vocabulary knowledge.

The Kaufman Test of Educational Achievement–Second Edition (KTEA-II; Kaufman and Kaufman) is an individually administered battery that consists of an assessment of the key academic skills in reading, math, written language, and oral language. The KTEA-II Comprehensive Form allows for error analysis along with detailed prescriptive information for remediation planning. The KTEA-II was redesigned and expanded, and covers IDEIA, Reading First, and NCTM achievement areas.

The KeyMath-Revised (Connolly, 1988) is an untimed, individually administered diagnostic test that measures mathematical functioning. The KeyMath consists of 13 subtests examining areas such as knowledge of numbers and fractions, ability to perform operations such as addition and subtraction, and real life skills involving money and time. The subtests are classified into three major divisions: content, operations, and applications.

The Peabody Individual Achievement Test–Revised (PIAT-R; Markwardt, 1989) is an individually-administered measure of academic achievement designed for children and adults, ages 5 to 22. The PIAT-R assesses six academic content areas with the following subtests: General Information, Reading Recognition, Reading Comprehension, Mathematics, Spelling, and Written Expression. The subtests are combined to yield a Total Reading score, Total Test score, and a Written Language Composite score. Administration time is approximately 60 minutes.

The Test for Auditory Comprehension of Language–Third Edition (TACL-3; Carrow-Woolfolk, 1999) is an individually administered measure of receptive spoken language that assesses a subject’s ability to understand the following categories of English language forms: Vocabulary, Grammatical Morphemes, and Elaborated Phrases and Sentences.

The Test of Early Reading Ability–Third Edition (TERA-3; Reid, Hresko, & Hammill, 2002) measures the actual reading ability of young children. The test consists of three subtests: Alphabet (29 items) measures knowledge of the alphabet and its uses; Conventions (21 items) measures knowledge of the conventions of print; and Meaning (30 items) measures the construction of meaning from print

The Test of Nonverbal Intelligence–Fourth edition (TONI-4; Brown, Sherbenou, & Johnsen) was created as a language-free measure of intelligence, aptitude, abstract reasoning, and problem solving for persons ages 6:0 through 89:11. It does not require the subject to read, write, speak, or listen. The test is entirely nonverbal and does not involve much movement more than a point, nod, or other symbolic gesture indicating a response decision. It is particularly useful for individuals who are deaf, language disordered (both expressive and receptive), non-English speaking, or culturally different, and those with conditions resulting from mental retardation, developmental disabilities, autism, cerebral palsy, stroke, traumatic brain injury, or other neurological impairment.

The Test of Reading Comprehension–Fourth Edition (TORC-4; Brown, Hammill, & Wiederholt) comprises eight subtests grouped under either the General Reading Comprehension Core, which yields a Reading Comprehension Quotient, or the Diagnostic Supplements. The General Reading Comprehension Core includes General Vocabulary (the understanding of sets of vocabulary items that are all related to the same general concept),

The Test of Word Reading Efficiency (TOWRE; Torgesen, Wagner, & Rashotte, 1999) is an individually administered timed test of word-reading efficiency in the English language for children and adults ages 6:0 to 24:11. This test is a measure of an individual’s ability to pronounce printed words fluently and accurately. The test focuses on two main abilities: the ability to sound out words with quickness and accuracy and the ability to recognize familiar words as whole units or sight words. The goal of the TOWRE is to quantify an individual’s level of skill with regard to word identification. Research shows that this is critical for ultimate reading success.

The Wechsler Individual Achievement Test–Third Edition (WIAT-III; Wechsler) is a measure of academic achievement skills and problem-solving abilities for ages 4 through 85. Administration time varies depending on the age of the examinee and the number of subtests administered, but the test usually takes 45 minutes to complete for children in kindergarten or prekindergarten, 90 minutes for grades 1 through 6, and approximately 90 to 120 minutes for grades 7 through 16 (4-year college).

The Wide Range Achievement Test–Fourth Edition (WRAT-4; Wilkinson is an instrument used to measure the development of basic academic skills in the areas of reading, spelling, and arithmetic for ages 5 through 75. There are two alternate test forms, the Blue and Tan version, providing three subtests that may be given in any order: (1) reading (recognizing and naming letters and pronouncing printed words), (2) spelling (writing names, letters, and words from dictation), and (3) arithmetic (counting, reading number symbols, computing oral problems).

The Woodcock Diagnostic Reading Battery (WDRB; Woodcock, 1997) is a set of carefully engineered (Woodcock, 1992) tests for clinical measurement of reading achievement and important abilities related to reading. Test development, item calibration, scaling, cluster composition, and interpretation were accomplished through the Rasch single-parameter logistic test model (Rasch, 1960; Wright & Stone, 1979), and stepwise multiple regression analysis.

The WJ III measures a great many aspects of academic achievement with a wide variety of relatively brief tests. Many of these achievement tests can be used with the WJ III Tests of Cognitive Abilities to assess a student’s abilities on many specific Cattell-Horn-Carroll (CHC) Gf-Gc “cognitive factors.” Examiners are permitted to select the tests they need to assess abilities in which they are interested for a particular student.