The Case Against Standardized Achievement Tests

No phenomenon poses a greater threat to educational equity, and ultimately, to the quality of education in this country, than does the differential performance of minority and majority students on standardized achievement tests. In the wake of increasing public concern that schools are somehow not as rigorous as they used to be, standardized tests have been hailed as the “get tough” medicine we need to restore traditional standards of excellence in our classrooms.

As the use of such tests has dramatically increased over the past several years, so has the power which tests wield over the educational lives of students. In every state, standardized achievement tests are now routinely used to determine access to education from first grade through graduate school. What has occurred in our schools is, to use FairTest’s terms, nothing less than a “testing explosion.”

Testing Explosion

There is little question that this “testing explosion” has the power to effectively destroy whatever small measure of educational equity has been achieved in our schools over the past two decades. What is perhaps less obvious is the realization that standardized tests pose a profound threat not only to equity in our schools, but the quality of our children’s educational experience as well – a threat so great, in fact, that standardized testing should be abolished altogether.

The most outspoken and zealous advocates of testing in this country – representatives of the Educational Testing Service – argue that standardized achievement tests do not create inequities within the educational system. They merely reflect the inequities that already exist there. Minority students tend to perform less well on standardized tests because the quality of education they receive is often inferior to that of majority students. Consequently, they lack the requisite skills and knowledge to perform well on tests. From this perspective tests are not the problem; schools are. This view is well represented in a recent statement made by Donald Stewart, current president of thy College Board: “If the gap between minority and majority test scores is ever to be closed, improvements in the elementary and secondary education of minority students must be accelerated.”

It is this view which largely fuels the current “testing explosion.” Undergirding it are two widely accepted, but wholly erroneous assumptions about standardized achievement tests. One is that they are, in fact, a valid measure of excellence i.e., they assess conceptual understandings which are essential to learning and ultimately important to the betterment of society. 

The second is that standardized tests can be used as a mechanism for improving education, especially for poor and minority students. 

No Real Connection to Excellence 

The truth of the matter is that far from measuring excellence, standardized achievement tests tend to focus primarily on mechanical, lower-order skills and to reward students’ rapid recognition of factual information. Thus while the president of the College Board glibly asserts that the way to improve minority performance on standardized tests is to improve the quality of their educational experience, teachers are left to confront the reality that what will prepare students to perform well on tests is often in direct conflict with, or simply extraneous to, their real educational needs. In their position statement on standardized testing of young children, the National Association for the Education of Young Children (NAEYC) lays out this conflict in clear, unequivocal terms:

” …current research on reading instruction stresses a whole language/literacy approach thal integrates oral language, writing, reading, and spelling in a meaningful context, emphasizing comprehension. However, standardized tests of reading achievement still define reading exclusively as phonics and word recognition and measure isolated skill acquisition. Similarly, current theory of mathematics instruction stresses the child’s construction of number concepts through firsthand experiences, while achievement tests continue to define mathematics as knowledge of numerals. (Young  Children, March, 1988, p.42)

Resolving this conflict in the classroom forces good teachers to sustain a tenuous balance between providing instruction that truly fosters students’ conceptual understanding of a subject and drilling them on the kind of isolated skill exercises that earn points on standardized tests. The point here is not that students don’t sometimes need to work on isolated skills, especially when they’re first learning how to read and write. But work on isolated skills is only a means to a larger end – applying those skills in a meaningful context. Removed from context as they  are on standardized tests, such skills are meaningless. Held up as a measure of achievement, they become mistaken for the end, for what is most important, instead of what is only ancillary and ultimately trivial.

The truth of the matter is that there is little, if any, real connection between quality instruction and standardized and standardized test performance. Consider, for-example, the most successful of high school English classes, one in which students learned to write thoughtful, original essays about topics they genuinely cared about, in language that was clear and specific and drew in powerful ways upon students’ most profound personal and cultural experience. Assume also that the teacher taught students to carefully edit their work so that grammatical errors in their papers were occasional and minor.

What about that highly successful class would prepare students to “correctly” con­clude on the American College Testing Program (ACT) that “pioneered” is preferred to “started up by;” “reach my destination” to “get there;” and “prove to be” to “come out to be?” What important conceptual understandings, what effective writing skills, would students from this class draw upon to decide on the ACT whether the underlined portion of “my thoughts were irresistibly sucked toward the moment when…” should remain or be replaced by pulled helplessly, uncontrollably drawn, or propelled mercilessly? (examples taken from Hoover, 1981).

One could well argue that a student who preferred “started up by” to “pioneered” or “get there” to “reach my destination” might in actuality be a more effective and sophisticated user of language than the student who chose the “correct” answers precisely because she chose the clearest, most easily understandable vocabulary in which to communicate ideas. The point, of course, is that the choice between ”get there” and “reach my destination” is a stylistic one, dependent upon what one is trying to communicate and to whom. Removed from a real communicative context, the choice is a meaningless and arbitrary one. It tells us nothing about a student’s level of language competence.

Or consider the following example from a standardized reading achievement test where the child is asked to determine the “right answer” in the following selection:

Father said: Once there was a land where boys and girls never grew up. They were always growing. What was Father telling?

The truth _____ A lie _____ A story _____

As Hoover, Politzer and Taylor point out, any of these could be “right” answer. “Metaphorically, it could be the ‘truth’ if the growth were mental and not physical. It could be a ‘lie’ in that the word ‘lie’ in blackspeech can also mean a joke or a story, and it could also be a ‘a story'” (Hoover, Politzer, and Taylor, 1987, p. 91)

Distinctions such as these are not arbiters of excellence. Nor do they provide us with valid information about a student’s ability to apply knowledge in a real context. A child’s performance on the word recognition portion of a standardized test doesn’t really tell us whether or not she can derive meaning from the books she reads in class or engage in discussion about them with her classmates, or apply the understandings she’s gained through her reading to a writing assignment or to her own life. A student who is asked to construct a well organized essay in 30 minutes on the Pre-Professional Skills Test in order to qualify for admission into a school of education might be an excellent writer and still perform poorly on this task because she thinks more deeply about issues than many other people and thus cannot as easily or quickly reduce her ideas to the highly simplistic, formulaic response necessary to produce a half-hour essay.

Yale psychologist Robert Sternberg makes the point that a major fallacy of standardized testing is the notion that “smart is fast” “Jumping into problems without adequate reflection,” says Stern­berg, “is likely to lead to false starts and erroneous conclusions. Yet timed tests of­ ten force a person to solve problems impulsively” (New York Times, no date).

Failure to Measure Important Abilities

Not only do standardized tests fail to provide an adequate measure of what they purport to assess – verbal and mathematical skills – they also ignore the much broader range of skills and abilities that enable one to function most effectively in a complex, pluralistic society – e.g., the ability to work collectively, to adjust to change, to function effectively in a variety of social and cultural contexts, to understand the perspectives of others, to persevere, to motivate, to solve problems in a real-life context, spatial abilities, kinetic abilities, bilingual and bi-dialectal abilities, analytical abilities, leadership abilities, oral and aural language abilities, social interaction skills, moral integrity, a sense of social commitment. As Harvard psychologist Howard Gardner points out, “there are hundreds and hundreds of ways to succeed, and many different abilities that will help you get there.”

It is an ironic, and for society, ultimately tragic coincidence that at the very time when advances in developmental psychology urge us to consider a broader and more complex conception of intelligence and ability, and of the multiplicity of talents and competences required to function effectively in the world, we have come to define excellence in our schools within the narrow parameters of what can be measured on standardized tests. When one considers the broad range of abilities necessary to do good work in the world next to the meager portion of these we’ve carved out to call excellence, it may well be as Joseph Renzulli has suggested, that when we use standardized tests to determine who has access to educational opportunity – to the accelerated class, the “best” high school, the college classroom – we may actually end up discriminating against the students with the greatest potential to make a contribution to society.

Destruction of Diversity

Perhaps no where is this discrimination better illustrated than in the devestating effect which standardized tests have had, and continue to have, on the number of minority students entering the field of education. In a 1986 study, University of Florida professor G. Pritchy Smith projected that the combined effect of the National Teacher Examination (required for initial teacher certification in at least 40 states) and the Pre-Professional Skills Test (required for admission to teacher training programs in a growing number of states, including Wisconsin) would nearly decimate the ranks of minority students entering the teaching profession in the next few years. The absence of minority teachers can only be expected to exacerbate the problems which many minority children experience in school. Minority teachers not only provide the role models which research and experience identify as crucial to academic achievement for minority students. They also bring understandings and perspectives to education which their white colleagues simply do not have access to. It is difficult to imagine how we will be able to improve the educational experience of students from diverse cultural communities without access to the perspectives of teachers who come from those communities.

In a recent article in the Harvard Educa­tional Review, Lisa Delpit gives voice to the growing sense of alienation which many minority educators feel from current educational debates about what constitutes effective instruction for minority students. She quotes a black principal discussing her experiences as a doctoral student at a well known West Coast university in which white professors and students talk authori­tively about issues concerning the education of black children:

If you try to suggest that that’s not quite the way it is, they get defensive, then you get defensive, then they’ll start reciting research.

I try to give them my experiences, to explain. They just look and nod. The more I try to explain, they just look and nod₁ just keep looking and nodding. They don’t really hear me.

Then when it’s time for class to be over, the professor tells me to come to his office to talk more. So I go. He asks for more examples of what I’m talking about, and he looks and nods while I give them. Then he says that that’s just my experiences. It doesn’t really apply to most Black people.

It becomes futile because they think they know everything about everybody. What you have to say about your life, your children doesn’t mean anything. They don’t really want to hear what you have to say. They wear blinders and earplugs. They only want to go on research they’ve read that other white people have written.

Delpit maintains that what lies at the heart of the alienation many minority educators feel from white dominated discussions of minority educational issues is fundamentally a question of power: “whose voice gets to be heard in the determination of what is best for poor children and children of color.” She argues that meeting the needs of poor and minority students in our schools requires authentic cross-cultural dialogue in which minority educators and scholars function as full and equal participants in the discussion. Corroborating Delpit’s perspectives is a small but growing body of research whose findings suggest that cultural factors, and in particular, the use of instructional methods which are congruent with students’ cultural background, play a significant role in the educational success of minority students.

As with advances in our thinking about the nature of intelligence and ability, the voices and experiences of minority teachers and scholars compel us toward a richer, more complex definition of what constitutes excellence, one that accounts for a greater variety of strengths and abilities and that encompasses a broader range of cultural perspectives. This is a conception of excellence that is deeply rooted in diversity.

Standardized achievement tests, and the simplistic notion of excellence they embody, sound the death knell of diversity in our schools. They silence a cross-cultural dialogue that has barely begun, not only in the field of education, but in every area of academic, professional and political life.  We live in a society which is becoming increasingly racially and culturally diverse. It is naive and ultimately suicidal to assume that we can solve the problems which confront such a society without access to both the perspective and the diverse range of skills and abilities represented in a multicultural population. We simply cannot afford a definition of excellence so narrow that it excludes access to the very resources we need to survive. 

Decline in Quality Education

Unarguably, Donald Stewart is correct in his assertion that minority students are often ill-served in our educational system. But given the fact that standardized tests bear little, if any, relationship to substantive learning, it makes no sense to assume that truly substantive improvements in the educational experience of minority students would necessarily have any effect whatsoever on their test performance. What is clear at this point is that in many schools something very like the opposite has occurred. As teachers have come under increasing pressure to raise students’ test scores, the quality of education these students receive has been significantly reduced. In too many classrooms, test content has come to dictate curriculum.

In some cases, abilities and skills not measured on achievement tests have simply been removed from the curriculum all together. According to FairTest, Gerald Bracy, former Director of Research, Evaluation and Testing in the Virginia Department of Education, found, for example, that some teachers in the system didn’t bother teaching students how to add and subtract fractions because the state’s minimum competency tests only included items on multiplication and division of fractions. Similarly, Deborah Meier, a public school principal in Manhattan, reports that when the New York City test for word meaning eliminated items on synonyms and antonyms, these were also eliminated from the curriculum.

In some states, matching curriculum with the content of standardization achievement tests has become a system-wide mandate. FairTest reports that school systems in at least 13 states are attempting to “align” their curriculum (with standardized tests) so that students do not spend hours studying materials upon which they will never be tested, regardless of the value or benefits which could be derived from that effort. Pressure on teachers and administrators to standardize curriculum in order to raise test scores can be intense. Weiss (1987) reports, for instance, that in 1985, the superintendent of schools in St. Louis decided to fire 60 teachers and principals because their students\failed to improve their scores sufficiently on standardized multiple choice tests.

The fact is that far from serving as a vehicle for educational reform, testing has had a negative effect on the quality of education in this country. Clearly, standardized tests neither measure excellence nor foster it in our schools. What they do, instead, is to provide a seemingly objective basis upon which to allocate educational resources. In the midst of discussion about equity in testing, it is easy to lose sight of the fact that the purpose of standardized testing is, after all, to make discriminations, to grant access to educational opportunities to some students while at the same time denying it to others. To that end, test items are deliberately selected so as to maximize differences between high and low scorers. The design and purpose of standardized tests thus insure that only some people will do well while oth­ers will necessarily perform poorly.

Bias and Decline in Equal Opportunity

In this sense, it doesn’t really matter whether or not standardized tests actually measure excellence, so long as society is satisfied that educational resources are being allocated to those who most “deserve” them. There can be little doubt that if a large percentage of white middle class students performed poorly on standardized tests, these test results would be viewed as invalid and their use in making education decisions termed discriminatory. The differential performance of minority students on standardized tests, however, occasions no such concern, not even in the face of an extensive body of research on cultural bias in testing which has been collected over the past twenty years. Most of this research has focused on various forms of linguistic bias in testing. Essentially, researchers in this area make the same points today that they did twenty years ago. The tests have multiplied, test items have been revised and updated, but the bias remains unchanged. 

According to a study by Hutchinson, for instance, fully a third of the items  in typical reading achievement tests are dialect prejudiced. For example, they require minority students to make distinctions between words, removed from context, which are often homonyms (sound alike) in their everyday oral speech patterns, such as had/hat and right/rat for black dialect speakers or this/these and tag/tack for Spanish dominant speakers (Examples taken from, Hoover, Politzer and Taylor, 1987). Despite the considerable body of research which indicates that phonological distinctions like these do not necessarily have any effect on reading comprehension when the words appear in meaningful context, such distinctions are used on standardized tests to help measure minority children’s level of reading achievement.

Items which require test takers to recognize or add inflectional endings, which are dialect distinctions for many minority students, appear frequently on the language usage portion of standardized tests. The following example was taken from the Law School Admissions Test (LSAT) (the student is asked to identify any errors):

Afrikaans is the language (A) of the ruling party in South Africa and (B) of Afrikaners (C) whose vote (D) maintain the status quo. (E) No error (taken from Weiss, 1987)

Whether or not deliberate bias was the intent here, it would be hard to imagine a speakers manipulate a second dialect under normal conditions, they are likely to revert to features of their first dialect when under unusual stress. Thus the person most likely to get this item wrong would be a bidialectal Black student experiencing the dual stress of taking a test crucial to her future while at the same time being forced to attend to grammar in a sentence about apartheid in South Africa.

Even when dialect items are not embed­ded within such provocative and racially insensitive content, they still place bidialectal and bilingual speakers at a major disadvantage compared to native standard English speakers, who do not have to waste cognitive energy monitoring interference between two language systems. Even if the stated intent of dialect-prejudiced items were to determine whether or not minority test-takers had mastered standard English conventions (which it is not), such items would not provide us with  any valid information about how these students actually used language in a real-life context as opposed to the highly stressful conditions involved in a testing situation.

Distinctions which involve dialect or language differences are not just trivial, as with most standardized test items, but highly discriminatory as well. Such items don’t just discriminate against individual minority students however. They discriminate against excellence. They turn what is an asset in a multicultural society – the ability to speak more than one dialect or language and thus communicate across a variety of social and cultural contexts – into a deficit, on the very tests which supposedly measure who is likely to make a meaningful contribution to society and thus deserves access to our best educational resources.

Another area of potential linguistic bias which has been extensively examined by researchers is the use of questions which assume cultural values and/or experiences which may not be shared by members of minority communities. On the vocabulary subtest of a standardized reading achievement test, for example, students are directed to choose the best synonym for inequality from among the following items: absence, foreign, difference, similarity, poor. Hoover, Politzer, and Taylor comment “of  the responses, all except ‘absence’ and ‘similarity’ could be ‘correct’ in  cultures in which students  are aware that difference, poverty, and foreignness are associated with inequality” (Hoover, Politzer, and Taylor, 1987). For children whose experience, or whose family’s experience, with the police or other officials has been a negative one, similar ambiguity exists as to which is the one “correct” answer in the following item, also taken from a standardized reading achievement test:

If a person does something against the law, he is an:

_____ambassador
_____official
_____offender
_____officer

Similarly Weiss (1987) found that questions which appear on recent forms of the Scholastic Aptitude Test necessitated familiarity with such class-tied pursuits as polo, golfing, tennis, minuets, pirouettes, property taxes, melodeons, and horseback riding. Again even leaving.aside the issue of cultural bias, what does knowledge of these activities have to do with scholastic aptitude?

Although there have been few studies done on the subject, there is some research which indicates that minority children do much better on test items whose content relates to familiar cultural experience.

A.P. Schmidt (1986) found, for example, that on a reading comprehension passage about life-style changes in Mexican­ American families, Mexican American students scored significantly higher than they did on reading comprehension passages whose content was less related to their lives. Similarly, Darlene Williams (1979) found that the I.Q. scores of Black students were raised when test items included pictures of Black people and of events related to Black culture.

Doubtless, many people would strongly object to the proposal that standardized tests be revised so as to include a sizable body of content specifically related to minority cultural experiences on the grounds that this would place middle class white students at a disadvantage. After all, why should they be expected to know anything about minority experience?

Yet aren’t minority students placed at an even greater disadvantage when standardized tests reflect little or nothing about their cultural experience, while for middle class white students almost everything on the test is familiar cultural terrain? There is one common ground on tests, to be sure. But middle class white students have to smuggle far less often than minority students to make meaning out of test items which take for granted experiences they’ve never had and which have absolutely nothing to do with ability. Is i ability or cultural experience that is being measured, for example, on the following item from the Scholastic Aptitude Test:

RUNNER: MARATHON

1. envoy : embassy
2. martyr : massacre
3. oarsman: regatta
4. horse : stable

In this example, it is marathons and re­gattas which a student must be familiar with to prove her fitness for college. At the other end of the educational ladder, it’s piano lessons, airplane trips, zoo excursions, musical recitals, museums, daddies who read story books, farm animals, historical sites and friendly policemen.

The advantages which middle class white students have on standardized tests extend beyond the linguistic features and the content of test items, however. There is good reason to believe that the test tak­ing situation itself is experienced very differently by majority and minority students. Test taking is a skill which middle class students are very good at because they receive extensive practice in answering “test questions” from the time they first learn to speak. Numerous studies of language socialization in white middle class communities indicate that the largest percentage of questions addressed to preschoolers by mothers and other primary

Caregivers consist of simply structured questions to which the questioner already has the answer – e.g., “how many eyes do you have?”; “what color is this dolly’.s dress?”; “how many fingers is mommy holding up?” The purpose of such questions is not for the questioner to gain information, but for the child to display information, for which she is typically rewarded with extensive non-verbal and verbal praise. Shirley Brice Heath notes that middle class caregivers are quite conscious of using such questions not only to teach children communicative behaviors that they will need to use in school in order to display competence in the classroom and on tests, but also to direct children’s attention to what it is they should learn.

When reading stories to preschool children, middle class parents often intersperse their reading with questions which focus the child’s attention on noting and recalling specific details of the text (e.g., ” now, how many balls is the little boy holding?” “What is the bird doing?”) Perhaps the most important “lesson” which preschoolers learn as a result of such interactions is how one is expected to com- municate about, and respond to, text.

In a highly detailed study of their daughter Rachel’s preschool language development, Scollon and Scollon (1981) amassed a large body of data which indicates that by the age of two, Rachel already knew that reading was a participatory literacy event in which one person assumed the role of “exhibitor and question­ er” while the other assumed the role of “spectator and respondent” Even though she could not yet literally “read,” Rachel could “pretend read,” changing her intonation patterns to imitate reading prosody. And when playing the role of respondent to questions about what she had “read,” she understood the rules governing proper display of knowledge in school and on tests: i.e., “(the ‘respondent’) should not provide a full knowledge of the text, but rather give responses to specific questions about the text addressed to her by the exhibitor/questioner.” These are, of course, the same rules which govern display of knowledge on a standardized test.

Advantages of the Culture of Power

The point here is that white middle class children enter school well rehearsed for performing well on standardized tests. Not only have they received considerable practice and experienced a great deal of prior success in answering test questions. They also tend to share in common a number of taken-for-granted assumptions about the nature of communication which serve them well in a testing situation:

1. Adults and children have a communicative relationship in which the adult is the “spectator” and the child is the “exhibitionist.” Thus it is perfectly appropriate, and indeed, expected, that a child will exhibit her abilities, or “show off’ for an adult.
2. Adults often ask children questions to which they already know the answer. Both the adult and the child understand that the purpose of such questions is for the child to demonstrate or “prove” that she knows the answer, rather than for the adult to gain new information or to seek the child’s point of view on a topic.
3. Text (written material which the child reads or listens to) is an appropriate context for being” asked test-type questions. Thus the child learns to direct her attention to those aspects of text which “fit with” her expectations about test questions – i.e., such questions tend to focus on highly specific aspects of the text, rather than on more global, thematic features; they deal with one aspect of the text at a time, they often reward the child’s ability and/or willingness to state, or demonstrate awareness of, what is obvious.
4. One is valued (i.e., viewed as competent and knowledgeable) for one’s ability to display information; to do this better than someone else is a mark of distinction and achievement.

Research indicates that many working­ class and minority children come to school with very different values and assumptions about what constitutes meaningful communication. While numerous language studies indicate that the “test question is the type most frequently addressed to middle class preschoolers, Heath (1983) found that in the working class Black community where she spent 11 years studying language socialization, children were almost never asked questions to which the adult or older child already knew the answer. According to Heath’s data, the most prevalent type of question addressed to preschoolers in this community was the  “analogy question,” calling for an open-ended response drawn from the child’s experience (e.g., “what do you think you are?” to a child crawling under the furniture). Other frequently asked questions were “story’ starters” (e.g., “did you see Maggie’s dog yesterday?”) and accusations (e.g., “what’s that all over your face?”) Children were also asked questions to which only they knew the answer (e.g., “What do you want?j, but very seldom test type questions, the assumption obviously being, why would you ask someone something you already know the answer to?

Not only were questions addressed to preschoolers in this community much more open-ended and elicitive of children’s unique perspectives and creative elaborations on a topic than were questions asked of pre-schoolers in middle class communities, reading was also an activity

perceived of much differently in this community. Reading tended to be a social event, in which listeners, young and old, were free to throw in comments or to elaborate on some connection to their personal experience, rather than a context for testing children’s comprehension or teaching them appropriate school behaviors. People in this community were admired for their ability to tell a good story, draw insightful analogies, or present an interesting and unique point of view, rather than for their ability to display information or show off knowledge for its own sake.

Research in other communities corroborates Heath’ findings. Many working class and minority children grow up in language communities where little value is placed upon asking children to display information for its own sake and where “stating the obvious,” “saying what everyone knows” is a communicative behavior which is not encouraged because it is perceived as having no meaningful communicative purposes. Thus for many working-class and minority students, the testing situation is not only an unfamiliar one, but one which violates some of their most deeply held assumptions and values about the nature and purpose of communication.

Communicative style – including such features as how one interprets the meaning of another utterance, what aspects of a message “need to be attended to, how one frames questions and structures answers, notions about what is worth talking about — is learned very early in life and is deeply tied to one’s personal and cultural identity. To suggest that members of minority communities change the way in which they interact with their children so that they will be better prepared to take tests in school is really to ask that they surrender a significant portion of their cultural identity. And to what end? Of what use is facility in answering test questions except to answer test questions? Everything about standardized testing – its assumptions of middle class experience, its emphasis on information removed from any meaningful context, its focus on individual achievement and competitiveness, the narrow range of abilities which it measures – is deeply at variance with the values and strengths of many minority communities. Those who argue that it is possible to make standardized tests less discriminatory by removing their cultural bias seriously underestimate the enormity of their task. What is a “culture­ fair” test in a multicultural society? And who is it that could design such a test? The truth is that any knowledge worth having is inextricably linked to culture and to context and thus isn’t reducible to measurement on a standardized test.

In the final analysis, the most fundamental question to be answered about standardized testing is not why minority students tend to perform less well than majority students, or even what can be done about it, but, rather, what is wrong with a society which allocates its educational resources on the basis of tests which not only fail to measure excellence, but which discriminate against the vast majority of its minority population?

References

Delpit, Lisa (1988). “The Silenced Dialogue: Power and Pedagogy.” The Harvard Education Review, Vol. 58, No. 3.

Heath, Shirley Brice, (1983). Ways With Words: Language, Life, and Work in Communities and Classrooms. New York: Cam. bridge University Press.

Hoover, Mary  R., Politzer, Robert L., & Taylor, Orlando (1987). “Bias  in Reading Tests for Black Language Speakers: A Sociolinguistic Perspective.” The Negro Educational Review, Volume 38, Nos. 2.3_

Hoover, Mary R. (1981). “Bias in Composition Tests with Suggestions for a Culturally Appropr!ate Assessment Technique.” In M. F. Whiteman (Ed.), Writing: The Nature, Development, and Teaching of Written Communication Hillsdale: Lawrence Erlbaum Associates.

Scollon, Ron & Scollon, Suz.a:nne, B.K. (1981). Narrative, Literacy and Face in In­terethnic Communication. Roy 0. Freedle (Ed.) Norwoord, NJ: Ablex Publishing. 

Williams, Darlene (1979). “Black En­glish and the Stanford-Binet of Intelligence,” Stanford, CA: Stanford University School of Education,Ph.D. Thesis, 1979.

Schmidt, A. P. (1986). “Unexpected Differential Item Performance of Hispanic Examinees on the SAT-Verbal, Forms 3FSAOS and 3GSAOS,” Unpublished statistical report of the Educational Testing. Service.

Smith, G. Pritchy, (1986). “Unresolved Issues and New Developments in Teacher Competency Testing.” Urban Educator, Fall, 1986.

Weiss, John G. (1987). “It’s Time to Examine the Examiners.” The Negro Educa­tional Review, Volume 38, Nos. 2-3.