- Title
- Examinee invariant condition of test score equating : the case of school certificate examination in computer studies for examinees across rural and urban areas of selected secondary schools in Malawi
- Creator
- Phiri, Gerson Mutala
- Subject
- Educational tests and measurements -- Malawi Computer studies -- Malawi -- Examinations Malawi School Certificate of Education
- Date
- 2018
- Type
- Thesis
- Type
- Doctoral
- Type
- PhD
- Identifier
- http://hdl.handle.net/10353/8507
- Identifier
- vital:32853
- Description
- This study examined the invariant condition of equating test scores of examinees for the Malawi School Certificate of Education (MSCE) Computer Studies test across rural and urban areas of Zomba District in Malawi. Equating of test scores to create comparable and interchangeable scores in assessment is increasingly becoming popular to most examining bodies. However, the Malawi National Examinations Board (MANEB) and many other examining bodies in Africa and Europe do not embrace test score equating in their assessment systems. The consequences of not equating test scores have not been clearly documented. Furthermore, there are no proper equating designs for some agencies to employ because they administer tests annually to different examinee' populations and they disclose all items after each administration. Therefore, the study sought to establish whether the invariant condition of equating test scores of examinees holds for the Malawi School Certificate of Education (MSCE) Computer Studies test across rural and urban areas of Zomba District in Malawi. Data were collected through two Computer Studies tests which were administered to a sample of rural and urban examinees. The data were analysed using means, standard deviations, T-tests and Statistical Package for the Social Sciences (SPSS). The study revealed that score distributions and test difficulties were similar across test forms signifying that test score equating is necessary. The classification of examinees into rural and urban geographical areas across the test forms were different before score equating, but similar after score equating. The equating of scores for the two sub-populations of rural and urban examinees displayed similarities in their Standard Error of Equating (SEE) values but they were not the same. The equated scores for rural examinees displayed more equating errors than the equated scores of urban examinees, indicating a better performance in Computer Studies by urban examinees than the performance of rural examinees. The study further revealed that equating of test scores was necessary because it was almost impossible to construct multiple forms of a test that were completely parallel. Even though test developers use the same test specifications to develop test items and make every effort to develop items in one test form as similar as possible to the items in another test form, the study has shown that there is no guarantee that the difficulty levels of the items would be the same. Among the many recommendations made, one major one is that the Malawi National Examinations Board (MANEB) should embrace the practice of equating test scores to improve fairness of decisions which are made on examination results.
- Format
- 247 leaves
- Format
- Publisher
- University of Fort Hare
- Publisher
- Faculty of Education
- Language
- English
- Rights
- University of Fort Hare
- Hits: 1115
- Visitors: 1091
- Downloads: 110
Thumbnail | File | Description | Size | Format | |||
---|---|---|---|---|---|---|---|
View Details | SOURCE1 | Mutala PhD Thesis .pdf | 6 MB | Adobe Acrobat PDF | View Details |