WEATHER & EARTH SCIENCE RESOURCES |
|
OTHER ACADEMIC COURSES, STUDENT RESEARCH, OTHER STUFF
|
|
TEACHING TOOLS & OTHER STUFF
 |
Higher Standards for Standardized Tests
Erica Bland and Sarah Westheimer
NSII Final Lab, semester 2
Chris Myers
5.4.01
Abstract
Our topic surrounded institutionalized racism as it is demonstrated in the composition of standardized tests. These tests have been under scrutiny recently for leaning towards certain sexes, classes and racial groups. We decided to take on this idea by surveying Miami students selected by their ethnic background and comparing their feelings of whether the tests accurately reflected their abilities. Erica felt this topic is largely ignored at Miami despite the number of theory classes offered at Miami, and Sarah felt that standardized tests probably inherently reflect the interests of the majority simply because of the nature of a test that measures skill. It seems that there has been growing, but sill too little, criticism for the lack of questioning where we get our standards, and if we need to change them to reflect all educational, racial, and economic backgrounds.
Purpose
We felt that research in this area may bring to light some of the flaws in the current system of standardized testing. It would help the members of the group gain a more profound understanding of instiutionalized racism as presented in class, but with a unique lens. By presenting our results to the class, we can help other Miami students gain a better understanding of some of the issues faced by institutions of higher learning today. It is also especially relevent at Miami, as the adminstration seems to place an extremely high priority on increasing the diversity of the student body. This can be seen with the "I Am Miami" campaign.
Problem
Our problem was that most people did not recognize these test bias. It has been so engrained into society that you must do well on these tests to be accepted to the institution of your choice that people become obsessed with getting good scores. This SAT was designed at one point and time to test the skills and ability of the persons who took it but we feel that time is long gone. Our ultimate problem is that the biases of these tests are preventing a lot of students with potential, but not good tests scores, or the right type of intelligence (as portrayed by this test) from getting into four-year colleges and universities. Standardized tests are making it more difficult for the world to have a promising future.
Hypothesis
We felt that females will feel the tests to be more biased then males. We also assumed that African Americans would feel the test was more biased then Caucasians based on past test bias designed against African Americans in the past. We felt that the non-white and non-black group would have results in the median of the results from the surveys taken by the black students and the surveys taken by the white students.
How did you decide on this project?
Our general topic was chosen in the partially the same manner for both group members: we looked at the time of the semester when we had the least due and found the week with the readings about "measuring heads" to be the ideal time. Of course, that was not the only element. It also seemed to be an interesting topic… one that would not only be engaging, but also involved controversy and a good amount of debate.
We could see the benefits of narrowing the topic down to standardized testing, especially at the college level. It’s a topic that is important to us as college students. It is important to Erica as someone who wants to go to law school. It is easily surveyable among the population of Miami, as we can assume that every student at Miami took a standardized test to gain entry.
The way we came to the specific topic was more sinuous… it was difficult at first to narrow the topic of "Measuring Heads" to something that can be measured at a school as, well, white as Miami. As we discussed the topic with Chris, we went through different topic ideas. They included the history of the IQ test, race equality, SAT scores and gangs. For the reasons mentioned above we chose SAT scores.
From the recently narrowed topic of SAT scores we began to develop the idea more in terms of what we specifically wanted to know about them (the scores) and we came up with:
From there we came up with:
How did you decide on your questions?
Erica decided on the questions. At this point we narrowed the study to research on people’s personal feelings about their own achievement and if the tests reflected that achievement accurately. Erica felt that with her background in coming from an under-privileged minority background and having a chance at a higher education she had been wondering for a while whether others felt the same way she did about standardized tests. She chose questions that would both quell this curiosity and answer the issues that had been brought up in the hypothesis. We ended up handing out two different editions of the survey in order to get a more robust idea of peoples’ opinions. Some controversy surrounds how much the way questions are asked in a survey affects the way people answer the questions, so we felt that mixing the questions up a bit may help us to avoid that problem.
The surveys follow:
Survey 1.
1.Are you a majority or minority student?
2.Are you male of female?
3.What is the economic status of your parents?
4.Did you take an ACT, SAT, LSAT, GMAT, MCAT or any other type of standardized test for higher education?
5. Did you score high or low on this test?
6. Do you think that this test was bias against you?
7. Do you feel that this test accurately portrayed your ability/skill?
8. Do you feel that your ethnic background played a role in your score?
9.Do you feel that your gender played a role in your score?
10.If these tests are proved to be biased do you think that this is fair?
Survey 2:
1. Gender: F M
2. Age:
3. Do you consider yourself an ethnic minority? Y N
4.What is the economic status of your parents?
lower class, middle class, upper middle class, or articulate other...
5.Did you take an ACT, SAT, LSAT, GMAT, MCAT or any other type of standardized test for higher education? Y N
(If no, please go to number 11. If yes, continue with number 6.)
6. Did you score high or low on this test (High being above the average national
score)? high low
7. Do you feel that this test accurately portrayed your ability/skill? Y N
(If yes, go to 11. If no, please answer numbers 9 and 10.)
9. Do you feel that your ethnic background played a role in your score?
10. Do you feel that your gender played a role in your score?
11. If you have additional thoughts on this subject, please write them out on the back of this paper.
We asked the people who were surveyed if they were male or female to find out if women thought the test was more biased then men. From the research we gathered from the class readings the answer to this question of whether the women thought the test was biased more against then men was yes, because women are socialized to be better at things other than standardized testing… especially math and science skill tests.
We asked if the surveyed were majority or minority students to get a clear picture of how many people were surveyed thought their ethnic background played a role in how they scored on the test. We asked socio-economic status of their parents. This was felt to be very important because it hinted at what kind of education their parents could afford. Exceptions to this certianly occur, but one can assume that a parent who can afford the best education their money can buy for their children, they will give them that. Exceptions include scholarships to high-end schools and high quality public schools. And finally we asked if each individual thought the test they took whether it was the ACT, SAT, GMAT, LSAT, of MCAT, accurately portrayed their skill and ability. This is the most important question in that it .
What did you accomplish?
We discovered a plethora of things about the bias of standardized tests. In our research, we discovered several things before we even began handing out surveys or counting results. We discovered that standardized tests had signifigant race bias. We discovered they were also male and female, upper and lower class biased. We discovered that these tests don’t accurately portray skill and ability. The only thing that they DO measure is your skill of test taking. We discovered that on any given day the test scores could be different based on your anxiety level, stress level, and even personal events that may happen in your life beyond your control. Many mind altering things happen to us everyday and we cannot determine why the score may have been high or low for any one person.
We discovered that these test are helping to perpetuate a viscous cycle of dependant families by stopping higher education. These tests block a fair number of talented students every year from entering into colleges and universities of higher learning. We discovered that the University of California’s chairperson has submitted dropping the test from its list of requirements. He is not alone in this action. The University of Texas, University of Massachusetts at Amherst and the college of Wooster in Ohio are all following along this progressive path. We feel this is because the test no longer holds its validity.
"280 of the nation’s 2,083 four-year colleges and universities make the SAT optional for some or all of their applicants; a handful of prestigious colleges including Franklin, Marshall and Mount Holyoke, have joined their ranks since the early 90’s and say they aren’t admitting idiots as a result"(Cloud 64). Miami University has a program called the Scholastic Enhancement Program, which has proven in its existence that students can be successful despite low scores. The SEP program looks at the potential of the student, the classes that they have taken, and the difficulty of the classes. They also look at extra curricular involvements as well as submitted essays to get a more rounded view or perspective of the student for admission to the university.
Types of errors that are prominent in these standardized tests are caused by factors that do not change. "Known as a systematic error, it is the result of characteristics other than those the test intended to measure. Gender bias in testing is often the result of such systematic error" (www.ed.gov/databases/ ERIC_Digests/ed328610.html). From this information and the above listed statistics we accomplished a well-rounded point of view on the issue at hand. We accomplished a great deal of information that will ultimately help mold our project. We gained information that was useful and both congruent and contradictory to our results. We accomplished the honest opinions of people towards standardized tests, and we found out that our project was one of great importance. Reform is currently being considered carefully despite the rumors of these tests staying around forever.
Relevance, if any. Why is this research interesting?
Our relevance remains the same and has been published as a national interest. We believe that this test no longer hold s relevance to today’s college society. We feel that it is keeping a lot of potential CEO’s and community leaders from achieving their goals in life because they cannot "look smart" from these tests scores. People are forced to get jobs, rather than careers, because they are not "college material." Not all but some people who had the potential become dependant, and then the cycle begins to start and it never stops. The anxiety of these tests is passed on to children because their parents did not do well and do not look fondly upon taking the tests. The statistics of these tests show a fluctuated racial gap that never really declined. This affects hopes and aspirations of doing better on the tests then their parents did. Today black young people who have the same educational background as white young people tend to score 120 points less then white kids. If they are going to the same school and getting the same education, why does this happen? The most likely explanation is that the tests are biased against certain socio-economic backgrounds.
Materials and Methods
Our design was original in that we asked people to do a self-evaluation. Other studies on test bias include scores and backgrounds, but we asked people to talk about how they felt that they did. In this way, no one person or group of people is judging the intelligence of others, it is entirely self-monitored. Opinions can never be wrong. This is one strength of our process.
As mentioned earlier in the lab, we tried to get a more well-rounded idea of what people felt by using two different formats of the survey. That way, the wording of the questions and the order of the questions had a diminished effect on the answers and results. We were originaly planning to look at the way people of different ages scored on the different tests, but found that would be too time consuming for this specific assignment. Instead, we doubled up on the number of undergraduates that we surveyed, which probably helped our sample be much more accurate. Also, while looking at people of different ages would help us get a more profound understanding of the past of standardized tests, it also added in an extra element that we felt may make the lab farther removed from scientific process.
To complement our own research, we studied what other people had to say about bias in standardized testing.
Materials
Materials were simply paper, pens, our minds, computers and legwork. We used Excel for the statistics.
Data
Females: 21% thought the tests were gender biased.
Males: 13% thought the tests were gender biased.
Black: 44% thought the tests were race biased
White: 17% thought the tests were race biased
Other minorities: 45% thought the tests were race biased.
Timeline of Execution:
We started surveying right after spring break, and that went till three weeks before school ended. Our presentation was on the 10th week. We presented to the class on the very last class… Friday, April 27.
Discussion and Conclusions
We got the results that got because we surveyed Miami University students who are almost all upper-class white students. Most of the Minority students that attend Miami University are also upper-class so we really didn’t get a well rounded selection of people to survey. Maybe if we were on a more diverse campus not only diverse in ethnic background but also in socio-economic status then our results would be a little more close to national averages that have been reported. Our results were perfect for the Miami population though because we surveyed all different types of students to create the widest diversity margin we could.
We also got that only twenty one percent of females thought that the test was biases against them because of their gender. This might actually be correct in their minds because little girls are now being socialized into once before all male dominated subjects such as math and science. This year sixty percent over forty percent females to males entered into law school which is a record high. I think that this may be good because if women were put into important political positions making decisions that ultimately affected the use of these test scores they would probably create more fair test. With no intentional bias I’m sure. Thirteen percent of males thought that the test was bias because of their gender. This was because they thought that the verbal part was easier for females. There was a spot for the surveyed to comment and this was one of the responses. I believe this may be true because women are socialized to be more social then males and you have to know how to express yourself verbally to be more social.
Seventy eight percent of females thought that the test was not bias against them, this may be attributed to the sudden breakthroughs of women in the business and political world but they still do exist. I do think that this one may have shocked us the most because there are so many classes offered on campus that explain the unfair differences between men and women and they make them more obvious for the people who may not be aware of it. I know that I have personally taken quite a few of these classes, which might explain my surprise toward the females who responded this way. Eighty seven percent of men said that they thought the test was not gender biased. From the beginning of time everything that exists today was primarily designed for men and therefore they would not be in the position of thinking that this test was biased to their advantage.
44% of African American Students thought that the test was biased towards them based on their ethnic background. This was primarily because of the statistics that are placed on display every year saying that they do worse then Caucasian students. It has been proven that even if an African American family and a Caucasian a family make the same amount of money provide the same education and have the same education the African American students still score 120 points less. These are just some suggestions but we would suggest they be strong suggestions as to why they think the test was racially biased. 17% of Caucasian students felt that the test was biased against them based on their ethnic background.
We think that these results are just a result of listening to the statistics year after year, and also being socialized to be better at one thing rather than another based on your gender and having better educational opportunities based on economic status.
We think that this project fits in to what others have done based on the socialization aspect of females. It fits into Leesa Gresham’s project on the evolution of female sexuality. We definitely think that, that has everything to do with the way a young girl is brought up. If she’s brought up to think that sex is bad then she will think that sex is bad, but if she is not then she won’t think so. We definitely think that their project is one worth while because if parents realize the socialization of their daughters has been wrong from their statistics then they will change their parenting skills and bring them up so that the problems that were posed in the past are no longer prevalent.
We would like to have seen more comments written on the surveys but most people are reluctant to even take the survey let alone write extra comments on the back. We would like to know what type of education most of the people we surveyed had… whether it was a college prep education or a public school education so that we could have made a direct correlation between the two of them.
The most exciting thing about his research project was that there was an eight page article in Time magazine on the issue. We didn’t think that anybody would want to listen to anything that we had to say because they thought that it would not be a very popular subject and that this test was perfectly legit. Erica writes, I was glad to see that many people sided with me because I did not do so hot on my test scores but because the Scholastic Enhancement Program I was admitted and now I am excelling in my academics, I just don’t do tests well but that’s alright because I’m learning. If it weren’t for that program I could be more wasted potential like I was mentioning earlier.
If we were to administer this survey again for further it would be interesting to add in more elements. We would like to see the differences between students in the inner-city versus suburban and urban students. We would like to compare different time periods of testing, and peoples’ opinions of the accuracy of those tests across the years. We would like to explore further different racial groups such as Asian, Indian, native American, and mixed race. This may provide much more insight into what peoples’ preconceptions are, and what more individualized test biases are. We might want to look into the type of high school that each person surveyed attended.
As far as amending the survey questions, we might want to specify what exactly constituted each economic class, and what scores exactly were low and what scores were high. If we asked the age question at all, we should follow up with a question about if they felt their age when taking the tests made any difference.
Our only suggestion would be not to make any assumptions as to what to expect. Keep an open mind and enjoy and put effort into your research because it may make an impact in the end.
To further explore race issues at Miami, check out the following:
http://www.IamMiami.com
http://www.muohio.edu/sep
http://www.muohio.edu/admissions
Sources:
Time Magazine March 12,2001 "Should SAT’s Matter" by John Cloud
NS Reader WCP 222 "The Nature" Cummins, Meyers Spring 2001
Eric Digests January 1,1990 Source: ERIC Clearinghouse on Tests Measurement and Evaluation Washington D.C., American Institutes for research Washington D.C.
MMSS Methods for Identifying Biased Test Items volume 4 By Gregory Camilli, and Lorrie A. Shepard 1994 Sage Publications Inc.
Next Article
Previous Article
Return to the Topic Menu
IMPORTANT: For each Response, make sure the title of the response is different than previous titles shown above!
WEATHER & EARTH SCIENCE RESOURCES |
|
OTHER ACADEMIC COURSES, STUDENT RESEARCH, OTHER STUFF
|
|
TEACHING TOOLS & OTHER STUFF
|
Listen to a "Voice Navigation" Intro!
(Quicktime or