There are several tests that are used by counselors to predict the success of individuals. The tests range from projective personality testing, individual intellectual assessment, achievement tests and many others. Achievement tests which deal with the assessment of the skills, knowledge and accomplishments in an area of interest levels is the most used kind of test by counselors. Achievement testing focuses on what the client knows or what the client can do in the present time. Achievement tests are such as California Achievement Test, Metropolitan Achievement Test, WIAT, WAIS and many others (Bradford, 1998).
These tests are meant to show the current trends of performance of individuals. When counselors use the information from the assessment report, then they can be able to predict if an individual will be successful or not. This kind of assessment may not be effective as the predictive validity of this instrument has to be known. One can use an approach with the knowledge that it is effective, not knowing that the quality of services he/she is offering using that service is low. It therefore creates a barrier to other effective methods that can be used (Carlson, 2007).
It is important to determine the predictive validity of a test so that the counselor ensures a fair judgment or the education admission officer, ensures that the selected candidates for admission are the right ones/a fair selection. Apart from fair judgment, correct guidance is given by the counselors and chances of misleading the public are reduced. This research aims at finding out the predictive validity of achievement tests. It will evaluate the use of achievement tests in childhood and adulthood, and from this give a conclusion.
Information is to be retrieved from scholarly journals with information about achievement tests and their predictive validity. Predictive Validity of Young Children's Achievement Tests. Achievement tests these days begin with the children and not as before when it was meant for upper grades. They have gone further to primary level. The tests as have been mentioned such as Metropolitan Achievement Test and California Achievement Test, are now given to children in grades one and two (Fair Test).
According to this Fair Test article, achievement test fail to give what is required of them. They do not give the real picture of the skills that students have since the tests do not test beyond test taking ability (Fair Test). This achievement tests are often used to evaluate the students ability in terms of what they have learnt, where the students need to put more effort and the quality of teaching the children have obtained . Parents are always advised to make use of the achievement tests reports to determine their children's performance in school.
It is because the tests are meant to show what the child does in school, what progress the child is showing, and even what weaknesses that child has. Achievement tests are also used to determine if a child qualifies for an admission. The big question arises when the predictive validity of these tests is doubted by many (Fair Test). Problems of Evaluating Children Using Achievement Tests According to the National Center for Fair and Open Testing, achievement testing in children does not yield what it is meant to.
This is because of what the test relies on that makes the evaluation of a child's skills, knowledge and others, be inadequate. The National Center for Fair and Open Testing therefore recommends that the achievement tests should not be used for making important decisions about a child's education (Fair Test). There are several presented reasons to this. These are; that the achievement tests do not measure the exact elements of importance needed to be measured. An example is the standard reading achievement test.
This test requires children to read what they are provided with and therefore just reveals the ability of the child to spell words, but not the actual reading ability (Fair Test). The tests undermine higher order thinking which is expected to develop in young children as they grow. This is because some of the achievement tests are of multiple choice formats. The multiple choices have one right answer, or in some case two, or no correct answers at all. It undermines the freedom to think of the correct answer and present the answer in a child's own opinion, which is important in skill development (Fair Test).
Another reason why this assessment method does not prove to be effective is the period of time taken when the children are taking the tests. Achievement tests take along period when doing that might frustrate a child and make his/her performance is poor, which is not a clear indication of the child's abilities, knowledge and skills. Several other factors affect the predictive validity of achievement tests. These are the child's health status, such as lack of sleep, nourishment (Fair Test).
Some others which affect the performance in the test are, the child's trust in the person giving out the test, a good night sleep and so many others. The National Center for Fair and Open Testing describes the achievement testing as biased considering the gender, culture, socio economic status and language (Fair Test). In a school, there are always different children from different cultures, families with different incomes, children with different languages and of course different sex. Achievement testing scores are often used to grade the children as either poor performing, as low or good, among others.
Children who know English as a second language, who come from poor families and who come from the minority, are always not favored by this kind of prediction of achievement as there are other factors in their lives affecting their performance. Another problem is with the aim of education. Children are taken to school to learn, for this learning to take place, the children must be assessed as part of the classroom program. This assessment should be done by the teachers as they are the ones who know the children and their behavior.
Achievement tests are developed by companies who know little about the local communities and the local curricula (Fair Test). These factors disqualify the achievement test as an evaluating instrument for determining the success of a child or for giving guidance. Predictive Validity of Graduate Achievement Tests In higher education institutions, these achievement tests are used to evaluate student's performances and to predict their academic achievements. In determining the predictive validity of achievement tests, Geiser compared achievement test to SAT tests (2008).
The tests are done to determine the grades of the students and those who should go for the next testing. For example the SAT-II Achievement tests recommended by most universities (Blacks In Higher Education, 1998). According to the journal of blacks in higher education, the SAT-II tests are racial biased. The blacks in higher education institutions are disadvantaged. The black students in colleges are said to do better that white students in certain SAT-II subject tests. These were tests on foreign languages that were foreign to the white students too.
Results of the black students revealed a better performance than the white students yet in other subjects, the whites dominated. Affirmative action fought its battle and ensured that standard tests were used for admission of students into higher learning institutions. This standard assessment method was not thought to have any disadvantages on the people being fought for to join the universities, which are the minority. It is quite evident that even though it is considered a standard measure, which has no bias in performance, the minority are still disadvantaged (Blacks In Higher Education, 1998).
From this analysis, it is important to note that SAT-II tests are not effective in predicting student achievement. This if compared to the achievement tests that consider mastery of curriculum content is a poor predictor of the student performance (Geiser, 2008). Geiser notes that in college admissions, achievement tests are better than SAT tests that have so many impacts on the poor and the minority as admission criteria. Achievement tests have to consider the class rank, high school grades and other academic achievement grades.
This makes it a better criterion for selecting students for admission (Geiser, 2008). Another problem with the achievement test is also shown by Wang et al, (2007) who conducted a research on “Predictive Validity of an English Language Arts Performance Assessment”. The research involved looking at the disadvantaged situation of a student and relating it to the performance in the achievement test. The study was also meant to determine the validity of the achievement test considering other factors to be constant.
The results lead to a conclusion that achievement tests are indeed a true picture of an individuals skills, knowledge and could be used to measure performance but was biased (Wang et al, 2007). This is because the disadvantaged students were affected and so could not perform better due to the various disadvantages. Examples found out are such English language proficiency, immigrant status and home language. One different finding though was that family SES and ethnicity were not factors that affected the scores of the ELA PA, which was one of the achievement tests (Wang, et al, 2007).
The group of researchers concluded that the achievement test were effective in measuring the ability of the students but were insensitive to certain demographic variables as English proficiency, immigrant status and home language (Wang, et al, 2007). Results on Other Achievement Tests Chulalongkorn University Language Institute is a Higher learning institution that values the importance of students taking English proficiency tests before they join the institution for their studies. This achievement test is referred to as CU-TEP.
A study was conducted on CU-TEP to determine to what extent this test could predict the students' achievement on Business English Graduate Course. The study was also to determine if this achievement test (CU-TEP), could predict academic achievement of the students in other courses (Opanon-Amata, 2003). The basis of analysis was correlating the CU-TEP test with the academic achievement of each student, obtained from the records of academic achievement, and with the achievements on the Business English Graduate Course. Students selected for the study were graduates undertaking accounting master's degree course program (Opanon-Amata, 2003).
Predictive validity of CU-TEP test study revealed that there was a correlation between the achievement test and the Business English Graduate Course achievement tests, but had no correlation to the students’ academic achievement tests. This kind of information is a clear show that CU-TEP can be used to predict the achievement of students in Business English Graduate Course, but considering the non correlation to the academic achievement, it shows that this achievement test cannot be used to predict the achievement of students in non related academic performance (Opanon-Amata, 2003). Discussion
How can validity be established? According to Oosterhof, validity of a test is established by correlation (1999). The writer notes that in order to determine the validity of what a test measures, a correlation between the student's performances on the test is computed against the scores on other variables. The basis of this is that, high or significant correlation of the performance of the students to the variable, indicate that the test measures a related thing to the variable. Low correlation indicates that the test does not measure anything related to the variable (Oosterhof, 1999).
Validity of a test to a variable can therefore be easily determined. Predictive validity can also be determined in this manner. If a high correlation is obtained between an achievement test and the expected achievement, then the test is valid and can be used to predict the achievement of students in that line. For example, the case of CU-TEP test used to predict the achievement of Business English Graduate Course students. If in any case there is no correlation, or little correlation, then the test is not valid for determination of student achievement in that line of education or course (Oosterhof, 1999).
Another example of predictive validity of a test is shown by House and Jonson who note in their report that the advance psychology tests (GRE), which is an achievement test, had a significant correlation to the grade performance in most of the psychology courses (2002), therefore valid. Factors Affecting Validity Even though Oosterhof shows that determination of validity is through correlation (2008), there are some factors that affect students that affect performance too. These factors as had been mentioned are discussed below. Having English as a second language, yet doing the same test with the native speakers.
This will not predict the real ability of the student as his/her understanding has been reduced by the language barrier. If the same test was given in an understandable language, the student would score higher. Other factors are such as socio economic status, period that the test takes and the health status of the student. This applies mostly to the children. Conclusion. Children's education and achievement have so many factors that affect them and so achievement testing cannot be a valid method for predicting their achievement.
An example is the situation a child might be in. If the child is stressed, he/she cannot be able to do the test and a good performance be expected from that. They are vulnerable and so should be taken care of even in the classrooms. There education should not be like the adults' education who knows how to deal with stress situations. Additionally, it is very rare to find a malnourished adult in a school/college. This is evidence enough that the children should not have the same system of predicting achievement.
This does not mean that achievement tests are suitable for predicting student/adult achievements. From the information gathered, it is only valid in cases where the measured variable is related to the test, but it has weaknesses like not considering the demographic variables. Recommendations The system of prediction is not effective in some instances and improvements should be made considering the language proficiency, period of doing the test and other important demographic factors found to affect the predictive validity.
These tests should be set considering the demographic variables that affect some students to make achievement tests standard, and an effective test that can be used to predict students’ achievement. Children still in lower grades should not do the test as an achievement prediction test. If it should be so, then changes should be made considering the various factors that affect the children, for example lack of trust in administers of the achievement tests.