A Training Program Based on the Quality Standards of Achievement Tests to Improve the Competence of Teaching Staff in Constructing Tests at Middle East University in Jordan

The aim of this study is to construct a training program based on the quality standards of achievement tests and to verify their effectiveness in improving the competence of the teaching staff in constructing tests at Middle East University. To achieve the objectives of the study, the researcher constructed a training program and assessment tool, which included 58 items distributed on five dimensions. The study sample consisted of (70) faculty members, who were randomly selected from different faculties during the academic year 2018/2019. The results showed a statistically significant difference (α = 0.05) in the evaluation of achievement tests in favor of post measures, and revealed the effectiveness of the training program. The results showed no statistically significant differences (α=0.05) in the evaluation of achievement tests attributed to specialization. The study recommended the adoption of the training program to compensate the lack of educational qualifications in one aspect of the evaluation process, and to take advantage of the quality standard list in the evaluation of the achievement test.


Introduction
Achievement tests play an essential role in a student's life in all stages of education. Based on their results, the future of a student is determined and many decisions are used against him.
The tests are still the primary means in evaluation of the educational process, as an aid in learning about the students' achievements of the educational goals, comparing the students with each other, diagnosing the strengths and weaknesses they have, or the selection of individuals for different professions.
However, most students hate tests, and avoid trying to appear for the experience in any way.
The reason may be that tests, like any other means of evaluation, may be used to achieve non-evaluation goals, such as punishment.
The other reason is the constant association with scores, and the consequent potential negative results such as failure (Odeh, 2010, p. 71; Abu Jarad, 2011, p. 90).
Another criticism of the tests is that they do not encourage improvement, but rather memorization (Murad and Soliman, 2005).
There are some negative aspects of some forms of tests used in the evaluation process, including: confining the role of education in training and exercising students to perform certain behaviors only, focusing on results rather than processes, and relying more on passive learning (Nixon, 1990).
Focusing on the scope and characteristics of the content rather than on the statistical characteristics of the test items requires reviewing prevailing practices in the design and construction of achievement tests (Allam, 2011).
The process of preparing achievement tests is carried out in a series of steps: defining the purpose of the test, analyzing the content, writing the teaching objectives, preparing the table of specifications, formulating the items, making the test, giving and correcting it, and finally analyzing and interpreting the results (Odeh, 2010, p. 146-147).
It is important to construct the table of specifications when designing the achievement test as a framework for obtaining a representative sample of the learning outcomes (Allam, 2011;Al-Noor, 2014;Odeh, 2010).
Procedures for examining and verifying achievement tests' information are considered vital if there is a desire for achieving justice for students in an educational institution (Lambert & Lines, 2000, p.4).
Many studies have been conducted in the construction of achievement tests and analyzing, such as: (Sinan,

Study Problem and Questions
The researcher provided an evaluation study which included a sample of achievement tests in various facilities at the university, in order to evaluate and identify the degree of their compatibility with the quality of the achievement tests' construction standards.
It was noted that many faculty members lack the skills which are necessary for the preparation of achievement tests.
The ability to construct tests can be improved through various training programs. This study came to verify the effectiveness of a training program based on the quality standards of achievement tests to improve the competency of faculty members in constructing tests. So, the questions of the study are: 1. How effective a training program based on the quality standards of achievement tests can improve the competency of faculty teaching members in constructing tests at MEU? 2. Are there any statistically significant differences (α = 0.01) in the evaluation of the quality of achievement tests prepared by faculty teaching members who have completed the training program due to their specializations (sciences, and humanities)?


The present study acquires its theoretical importance from the importance of training of the faculty teaching staffs at the university in order to maintain an advanced level in constructing achievement tests. In addition, it makes sure that the faculty teaching members can state test items with acceptable psychometric properties.  The study provides information and data for faculty teaching members at the university to introduce them the quality standards of achievement tests, training them to use these standards, and connect their performances to contribute to the improvement and development of their performance.  The present study provides a training program that contributes to the improvement of the competencies of faculty teaching members in constructing achievement tests.  The study provides a tool that includes the most important criteria in constructing achievement tests, so that they can be used in evaluating achievement tests at the university.

Study Objectives
1. Construct a training program for faculty teaching members at the university according to the quality standards of the achievement test. 2. Measuring the effectiveness of the training program in improving the competencies of faculty teaching members in constructing the achievement test.

Methodology
The researcher used the quasi-experimental approach.

Study Population
The study population consisted of all (176) faculty teaching members at MEU for the academic year 2018/2019 according to the statistics of the Human Resources Department at the university.

The Study Sample
The sample of the study consisted of (70) faculty teaching members who were randomly selected and represented (39.2%) of the total society, of which (38) members in the scientific faculty, and (32) members in the humanities faculty.

Study Tool
A training program was developed based on the quality standards of the achievement tests (Odeh, 2010;Rawashdeh et al., 2000;Coop, 2006).
The training program consisted of (5) training sessions, (2) hours long, two sessions per week. The trainees were divided into groups, each group consisted of (4-6) members. The topics of the sessions were as follow: The researcher has prepared a tool to assess the quality of achievement tests which consisted of (58) standards that were derived from a review of theoretical literature and previous studies related to the international standards to construct achievement tests. The assessment tool consisted of (5) main dimensions as follows:  Dimension (1)  Each of the above standards is given a score between (0-2).  Score (2): If the given standard is completely performed.  Score (1): If the given standard is performed to lower degree.  Score (0): If the given standard is not performed at all.
The validity of the evaluation tool was verified by consulting (12) specialized examiners working in the field of educational measurement and evaluation.
The reliability of the tool was evaluated in its final form by applying it to (18) achievement tests.
The value of the coefficient of reliability of internal consistency; Cronbach alpha was (0.87). The values of reliability coefficients of the dimensions were between (0.83-0.89).

Data Collection
Pre-test was performed by analyzing the study sample exams. Then the training program was implemented, and a questionnaire was distributed to assess the quality of the test sheet.
Post-test was performed by collecting the final exams prepared by the study sample and then analyzing them.

Results and Discussions
To answer the first question: How effective a training program based on the quality standards of achievement tests can improve the competency of faculty teaching members in constructing tests at MEU?
The results were prepared by the faculty members who completed the training program and compared with the results of the pre-performance.
The mean and standard deviations of the standards for each dimension of achievement tests were calculated. Table 1 shows the results of the analysis: It is noted from the results of table 1 that there are apparent differences between the values of the means in the pre-and post-measures.
The total mean of the pre-performance is (0.73) with a standard deviation of (0.16), while the value of the mean of the post-performance is (1.89) and standard deviation (0.20).
To determine the significance differences between the values of the means, the researcher used a paired sample t-test, and the effect size was calculated using the following equation: Eta Square (η 2 ) = t 2 / (t 2 + df) Where: η 2 :Eta Square. t 2 : square t-value. df: degrees of freedom.
To determine the effect size (the ratio of the total variance in the dependent variable, which can be attributed to the independent variable). The following values were adopted (Kiess, 1989, p. 448):  The value of the Eta-Square is less than (0.20); the effect size is weak.  The value of the Eta Square between (0.20-0.80); the effect size is medium.  The value of the Eta-Square is greater than (0.80); the effect size is high.
The effectiveness of the training program was calculated using Modifies (Black Gain Ratio) as follows (Al-Jarallah, 2013, p. 87): Where: Y: mean of post-performance scores. X: mean of preperformance scores. d: Final score of the assessment.
The range is determined to measure effectiveness based on the percentage set by Black for the effectiveness which is placed between (1-2). Black suggests the boundary of this ratio is (1.2%), so that the effectiveness of the program can be considered acceptable. Table 2 shows the results of the analysis.
It is noted from the results of table 2 that there is a statistically significant difference (α=0.01) attributed to the post-performance; that is, the sample performance improved after training.
All values of Eta-square (η2) were large, indicating that the value of the effect of the experimental variable (training program) was large. The adjusted gain ranges between (1.33-1.63), and all values above (1.20%).
The adjusted total performance gain was (1.48), which is large. This indicates the effectiveness of the training program.
It can be said that the proposed training program improves the skills of faculty staffs in constructing achievement tests, as it meets the needs of faculty staffs in all aspects of constructing achievement tests.
The training program focused on the cognitive and practical aspects of the stages of constructing the achievement test. It offered diverse activities and teamwork, as well as the trainees' desire to be benefited to the fullest extent from the development of achievement tests in accordance with the quality standards.
To answer the second question: Are there any statistically significant differences (α = 0.01) in the evaluation of the quality of achievement tests prepared by faculty teaching members who have completed the training program due to their specializations (sciences, and humanities)?
Independent sample t-test analysis was used. Table 3 shows the results of the analysis.  It is noted from the results of table 3 that there are no statistically significant differences (α=0.01) in assessing the quality of achievement tests attributed to specialization, where all values of "t" were not statistically significant (α=0.01).
This result can be explained by the nature of cognitive competencies related to constructing achievement tests, which are acquired through learning, training and experience.
The results are consistent with the study of (Saif, 2008; Algdah, Miqdadi & Abu Asfer, 2008; Al Miloz & Sherbini, 2015). Studies have shown that there are no statistically significant differences attributed to specialization.

Recommendations
1. The study suggested adding preparation programs for faculty members before and during service to the list of quality standards of achievement tests because of its positive impact in improving the quality of their professional performance. 2. Adopting the training program to compensate the lack of educational qualification in one aspect of the evaluation process. 3. Utilizing the list of quality standards of achievement test in evaluating tests prepared at the university. 4. Making studies to develop training programs in all aspects of the measurement and evaluation at the university.