Improving reading ability and academic performance through working memory training
- Authors: Mukheibir, Adrienne Jayne
- Date: 2015
- Subjects: Reading -- Ability testing , Short-term memory , Attention-deficit hyperactivity disorder
- Language: English
- Type: Thesis , Masters , MEd
- Identifier: http://hdl.handle.net/10948/4840 , vital:20716
- Description: This dissertation is based on the results of a study that used a working memory (WM) adaptive computer programme known as Jungle Memory to determine whether training WM would have a positive impact on learners with reading difficulties, which would ultimately lead to an improvement in their overall academic performance. A pragmatic paradigm has been used, involving a mixed methodology that allowed for a postpositivistic as well as a constructive approach. The first section of the investigation involved the quantitative component, where specific variables (performance levels of the sample of learners) were determined. Thereafter, empirical measures were used to determine if these specific variables increased once the WM intervention programme had been completed. Concurrently, in the qualitative component, the sample group (namely the teachers working with these learners) provided qualitative information to verify if any changes in the participants were observed after the intervention. The participants included 30 boys who attend an ex Model C School in the Eastern Cape where I teach, all of whom had been identified as experiencing various reading difficulties and were receiving remedial assistance. The boys were placed into three groups according to specific criteria: Group A had all been diagnosed as having ADHD and were being treated pharmaceutically for the disorder; Group B all had significant ADHD symptoms but were not receiving any pharmaceutical assistance to alleviate these symptoms; and Group C displayed no significant ADHD symptoms. Five boys from each subgroup formed the Experimental group and the other five formed the Control group in each stratum. The participants' class-based teachers involved in the qualitative component of the study provided information-rich data regarding any notable progress made by each participant. Each participant took part in a battery of pre-assessment tests to determine their general level of academic performance. These assessments included the following well-known standardised tests in South Africa: the Young Group Reading Test, the Essi Reading Test, the Essi Spelling Test and the Vassi Mathematics Proficiency Test. The participants from all three Experimental Groups then took part in the eight week intervention programme, Jungle Memory, while the Control groups continued with their regular remedial programme. At the end of the eight weeks, all the participants were re-assessed using the same standardised tests.The findings from this study revealed that the participants from all three strata of the Experimental group showed significant improvements in the quantitative postassessment results and twelve of the fifteen also showed significant improvements in the comments made by the teachers in the qualitative component. The majority of the participants from the Control groups however, made very little progress in both components, and these small improvements were probably due to other variables such as maturity, incidental WM improvements through observing their peers, or familiarity with the tests.
- Full Text:
- Date Issued: 2015
Applicability of a health literacy test from the U.S. in a South African population
- Authors: Lecoko, Motlalepule Lebogang Elizabeth
- Date: 2001 , 2013-04-29
- Subjects: Literacy , Literacy -- South Africa , Literacy -- Ability testing , Reading -- Ability testing
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:3824 , http://hdl.handle.net/10962/d1005931 , Literacy , Literacy -- South Africa , Literacy -- Ability testing , Reading -- Ability testing
- Description: This thesis investigates the suitability and applicability of a health literacy test from the U.S. in a black, Xhosa-speaking, South African population. The concept of literacy is a controversial one which has been much debated, as it is not easy to classifY people as simply either literate or illiterate. As a result there are a number of definitions of literacy that vary with purpose and culture, but the most common one is that a person is literate if he/she can read and write. Estimating literacy from years of schooling is an inexpensive method but is also unreliable, since people generally read 3 to 5 grades below their stated educational level. This method affords little insight into the ability of patients to adequately function in a health care enviromnent, an ability which is referred to as functional health literacy. A number of health literacy tests such as the REALM (Rapid Estimate of Adult Literacy in Medicine) test have been developed to assess this skill. The REALM test is a word recognition test which places people into a relevant grade range estimate according to the number of words pronounced correctly. It appears to assume understanding of the word if the person is able to read that word correctly. In this project 125 black Xhosa-speaking respondents of varying educational levels who were literate in English were interviewed with the aid of an interpreter. Comprehensive demographic data were collected. Respondents were first asked to read all 66 words aloud during which time pronunciation was checked, and thereafter they were asked to explain each word. It was found that the ability to automatically decode and read the words did not necessarily guarantee comprehension of these words. Many of the words proved to be unfamiliar to the majority of the Xhosa respondents who were able to pronounce them correctly, but could not explain them. These tended to be phonetically transparent words which were therefore more accessible to the unfamiliar reader. This research has proven to be of great value in helping identify such words which should be substituted with simpler words for use in health information materials. A number of words could neither be pronounced nor understood by the population majority and, interestingly, a small group of words could not be pronounced but were satisfactorily explained by some respondents. The results showed an extremely poor correlation between the stated educational level and the REALM grade range estimate. This emphasizes the inappropriateness of years of formal schooling as an indicator of functional health literacy. The criteria were established for deciding cases in which the REALM test could be applied (or succeeds) and when it is inapplicable (or fails). It was found to be inapplicable in 41% of cases which clearly indicates that, in its current form, it is not a valid, reliable test to use in determining health literacy in this English second language population. It can, however, be used as a basis fur the development of a more appropriate test. Recommendations for future research direction are presented and an alternative structure for a health literacy test is suggested. , KMBT_363 , Adobe Acrobat 9.53 Paper Capture Plug-in
- Full Text:
- Date Issued: 2001