Millennial Debate Standardized Testing Debate
A2: Culltural and Racial Bias in Testing Subject Matter
Download 1.17 Mb.
|
- Navigate this page:
- Testing only demonstrates the gap, it is not responsible for it
- A2: Teaching to the Test
- Yes, the instruction should match the testing
- Test design prevents over-teaching to the test
A2: Culltural and Racial Bias in Testing Subject MatterReview eliminates cultural and racial biasDean Goodman & Ronald Hambleton, University of Massachusetts @ Amherst, 2005, Defending Standardized Testing, page number at end of card Test publishers and state departments of education are relentless in their search for potentially biased test items by reviewing items for potential gender, racial/ethnic, cultural, religious, regional, or socioeconomic disparities in understanding or performance. Consider the steps taken routinely by many testing agencies and state departments of education to remove bias from educational assessments: 1. Item writers who are members of multicultural and multiracial groups are among those who are used to write the assessment material—directions, questions, scoring rubrics, etcetera. 2. Item sensitivity committees representing diverse minority groups are established to focus on aspects of educational assessment material that might be unfair to minority students, or may represent stereotyping of minority groups. 3. Item reviewers, prior to any field testing of assessment material, are instructed to identify aspects of test items that might be unfair to minority groups or represent stereotyping. 4. Statistical analyses are carried out on field-test data searching for assessment material that is potentially problematic for minority groups. 5. All test publishers and most state departments of education have a document that is used by item sensitivity committees and other reviewers to spot potentially problematic or biased assessment material. 6. At the final stages of test and assessment development, content committees are sensitive to the inclusion of material that is not assessing the content standards or may be biased against minority groups. . Normally item writers and reviewers would be asked to avoid or identify a variety of potential sources of item bias that might distinguish majority and minority groups of students: content that may not have the same meaning across groups, test items that contain vocabulary that may not have the same meaning in all groups, clues in items that might give an unfair advantage to students in one group over another, items that because of student prior knowledge may advantage one group over another, and so on. Often, item writers are asked to avoid 25 to 30 potential sources of item bias in their work, and just in case they slip up, item reviewers are given the same list of potential sources of item bias to see if they can spot these problems or any others in items that have been written. In addition to judgmental reviews, it is common to compile statistical data (e.g., using logistic regression or the Mantel-Haenszel procedure) to compare students in majority and minority groups (at least for males and females, and Blacks, Hispanics, and Whites). For several years, the Center for Educational Assessment at the University of Massachusetts has been conducting studies to identify potentially biased test items on seven of the state's educational assessments. Black-White, Hispanic-White, and Male-Female analyses are routinely carried out. In the year 2000, for example, potential item bias was studied in 696 items and only 24 items were identified for additional investigation. This is a rate of 3.5% of the test items with a combined potential bias of a small fraction of a point (on a 40 point test) if all of the potentially biased test items were actually biased against a single (2005-03-23). Defending Standardized Testing (Kindle Locations 2946-2953). Taylor and Francis. Kindle Edition. minority group. Items are flagged for further investigation if majority and minority groups matched on overall ability show a .10 or greater difference on a per point basis. (This means that there would need to be 10 of these potentially biased items in a test to result in an actual one point difference between the majority and minority group due to potentially biased test items.) Clearly the amount of bias that is appearing on educational assessments is likely to be small because of the efforts that are being made to spot and eliminate problems early. There is simply little or no evidence to claim that item bias is a serious problem today on state assessments. (2005-03-23). Defending Standardized Testing (Kindle Locations 2953-2957). Taylor and Francis. Kindle Edition. Testing only demonstrates the gap, it is not responsible for itLelac Almagor, September 2, 2014, Boston Review, The Good in Standardized Testing, http://bostonreview.net/us/lelac-almagor-finding-good-in-standardized-testing DOA: 10-25-15 Compared with state tests such as the DC CAS, early versions of these Common Core–aligned tests have often revealed bigger gaps in achievement between disadvantaged kids and their peers. But the measurement is not the problem. Testing doesn’t produce the staggering gaps in performance between privileged and unprivileged students; historical, generational, systemic inequality does. Testing only seeks to tell the truth about those gaps, and the truth is that the complex tasks of the Common Core are a better representation of what our students need to and ought to be able to do. I’m all for measuring that as accurately as we can. In recent years our schools have in fact made huge gains in helping our students tackle real complexity. I’d love to take genuine pride in our scores, knowing they reflect those strides toward rigor. A2: Teaching to the TestThat’s the point – the kids need to learn what is on the testNorman R. Augustine is chairman of the National Academies’ congressionally mandated review of U.S. competitiveness. He is a former chairman and chief executive of Lockheed Martin Corp, Bangor Daily News, August 3, 2013, Bangor Daily News, Here’s Why Schools Need Standardized Testing, http://bangordailynews.com/2013/08/03/education/heres-why-schools-need-standardized-testing/ DOA: 10-25-15 First, they contend that these exams detract from the larger goals of education by encouraging teachers to “teach the test.” In a certain sense, however, teaching the test is the whole point. Exams are instruments for measuring student proficiency. And, as I’ve learned during my career in the business world, measuring something is often the best way to maximize or improve it. Economist Dan Ariely of Duke University has said: “CEOs care about stock value because that’s how we measure them. If we want to change what they care about, we should change what we measure.” If an exam effectively gauges a student’s mastery of U.S. history or English grammar, then teaching the test is simply a matter of helping students develop that knowledge. Teachers who feel that a test ignores something essential should commit to fixing the test, not condemning the entire practice of testing.
Gregory Cizek, professor of educational measurement and evaluation, 2005, Gregory J. Cizek teaches courses in applied psychometrics, statistics, program evaluation and research methods. Prior to joining the faculty, he managed national licensure and certification testing programs for American College Testing, served as a test development specialist for a statewide assessment program, and taught elementary school for five years in Michigan. Before coming to UNC, he was a professor of educational research and measurement at the University of Toledo and, from 1997-99, he was elected to and served as vice-president of a local board of education in Ohio, Defending Standardized Testing, Kindle edition, page number at end of card Another version involves narrowing teaching to include only those objectives covered by the high-stakes test. Many testing professionals (and others) would also agree that exclusion of other, valuable outcomes and experiences from the curriculum is undesirable. Finally, it is possible to align instruction with the curriculum guide, content standards, and so forth (depending on the terminology used to describe the valuable student outcomes in a particular locale). And, it is obviously desirable that any high-stakes test be closely aligned with the curriculum or content standards it purports to assess. Thus, it would neither be a coincidence—nor inappropriate—if the well-aligned instruction and testing bore a strong resemblance to each other. This is sometimes mistakenly referred to as teaching to the test where the more accurate (and supportable) practice should probably be distinguished by use of a different descriptor, such as teaching to the standards or similar. (2005-03-23). Defending Standardized Testing (Kindle Locations 1328-1336). Taylor and Francis. Kindle Edition. Test design prevents over-teaching to the testDean Goodman & Ronald Hambleton, University of Massachusetts @ Amherst, 2005, Defending Standardized Testing, page number at end of card To address concerns about "teaching to the test," state assessment programs typically administer different forms of a test within and across each testing cycle. These forms often share a sufficient number of items to ensure that the forms can be placed on a common scoring metric through a statistical equating process (see Cook & Eignor, 1991; Kolen, 1988; and Kolen & Brennan, 1995; for discussions of ways different test forms can be equated). The test forms also contain unique sets of items that enable reliable information to be collected on a wide range of skills and concepts taught throughout the course of study. Ironically, when curriculum and assessment are in alignment, "teaching to the test" is exactly what teachers should be doing because the test or assessment will contain a sampling of questions from the curriculum, and so the only way to effectively prepare students to perform well on the tests is to teach the curriculum to which the tests are matched. (2005-03-23). Defending Standardized Testing (Kindle Locations 2772-2775). Taylor and Francis. Kindle Edition. Directory: uploads uploads -> 587 Return function, r i(X) r i(0) r i(1) r i(2) r i(3) 1 0 2 4 6 Thermal Station, I 2 0 1 5 6 3 0 3 5 6 10 uploads -> Department of science and humanities department of mathematics part-a questions and answers uploads -> The Case Against the nypd’s Quota-Driven ‘Broken Windows’ Policing What Is It? uploads -> For want of a nail uploads -> Ownership Models There are two different types of ownership models uploads -> Police Militarization uploads -> City of rising star regular meting thursday, july 14, 2016 uploads -> Township of winfield uploads -> Stop Militarizing Law Enforcement Act of 2014 uploads -> The Black Panther Party’s Ten-Point Program What We Want Now! Download 1.17 Mb. Share with your friends:
|