The development of a set of guidelines for the revision of psychological tests and the use of revised psychological tests
- Authors: Cronje, Johan Herman
- Date: 2020
- Subjects: Psychological tests
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/46350 , vital:39567
- Description: The psychological testindustry has produced a wide variety of psychological tests that are used by professionals to facilitate measurement and decision-making. Tests are updated and revised periodically in order to remain current, valid and reliable in what is a competitive psychometric industry. Despite the prevalence of test revisions, especially in recent years, a number of authors have commented on the lack of comprehensive guidelines for test revision. Guidelines shouldcover aspects such as what the different types of revision are, when to embark on a revision, whatprocessto followand how test users should use revised tests. Test revision differs from test construction in a number of ways. There are external factors that affect the regularity with which a test should be revised. Test revision also involves more role players than test construction, including the opinions of those test users who may be resistant to any change in the previous test edition. Finally, revised tests sometimes have to contend with requirements from the test publisher who purchased the test or distribution rights from the developer. Test revision is expensive and time consuming, which leaves little scope for experimentation or trial-and-error. The availability of expertise, as well as the human and financial resources required to complete test revisionscan make such projects unaffordable, especiallyforprofessionals indeveloping countries, such as South Africa.It may be more feasible for such professionalsto collaborate with international revision projects. By doing so they cangain experience in test revision, contribute indigenous information that could shapethe revision ofan international test, increase opportunities to engage with international users, and potentially source international funding for research in their own country. The current studydevelopeda comprehensive and practical set of 30 guidelines to assist those involved in test revision. These guidelines were peer-reviewed and refined. Finally, the guidelines were field-tested using a case study of a recently revised ivdevelopmentaltest, the Griffiths III. Professionals from South Africa, including the present researcher,formed part of the international team for the extensive revision of the Griffiths III, which makes this test an ideal case study from both the perspectives of the developed test revision guidelines as well as collaboration of professionals from a developing country in an international test revision. The knowledge gainedfrom the development of guidelines and international collaboration in test revision is reflected on.
- Full Text:
- Date Issued: 2020
- Authors: Cronje, Johan Herman
- Date: 2020
- Subjects: Psychological tests
- Language: English
- Type: Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10948/46350 , vital:39567
- Description: The psychological testindustry has produced a wide variety of psychological tests that are used by professionals to facilitate measurement and decision-making. Tests are updated and revised periodically in order to remain current, valid and reliable in what is a competitive psychometric industry. Despite the prevalence of test revisions, especially in recent years, a number of authors have commented on the lack of comprehensive guidelines for test revision. Guidelines shouldcover aspects such as what the different types of revision are, when to embark on a revision, whatprocessto followand how test users should use revised tests. Test revision differs from test construction in a number of ways. There are external factors that affect the regularity with which a test should be revised. Test revision also involves more role players than test construction, including the opinions of those test users who may be resistant to any change in the previous test edition. Finally, revised tests sometimes have to contend with requirements from the test publisher who purchased the test or distribution rights from the developer. Test revision is expensive and time consuming, which leaves little scope for experimentation or trial-and-error. The availability of expertise, as well as the human and financial resources required to complete test revisionscan make such projects unaffordable, especiallyforprofessionals indeveloping countries, such as South Africa.It may be more feasible for such professionalsto collaborate with international revision projects. By doing so they cangain experience in test revision, contribute indigenous information that could shapethe revision ofan international test, increase opportunities to engage with international users, and potentially source international funding for research in their own country. The current studydevelopeda comprehensive and practical set of 30 guidelines to assist those involved in test revision. These guidelines were peer-reviewed and refined. Finally, the guidelines were field-tested using a case study of a recently revised ivdevelopmentaltest, the Griffiths III. Professionals from South Africa, including the present researcher,formed part of the international team for the extensive revision of the Griffiths III, which makes this test an ideal case study from both the perspectives of the developed test revision guidelines as well as collaboration of professionals from a developing country in an international test revision. The knowledge gainedfrom the development of guidelines and international collaboration in test revision is reflected on.
- Full Text:
- Date Issued: 2020
The Bender Gestalt Test: an investigation into problems concerning administration and scoring and its application to low-educated adults
- Authors: Dyall, Kate
- Date: 1996
- Subjects: Bender-Gestalt Test , Psychological tests , Brain-damaged children -- Psychological testing
- Language: English
- Type: Thesis , Masters , MA
- Identifier: vital:2968 , http://hdl.handle.net/10962/d1002477 , Bender-Gestalt Test , Psychological tests , Brain-damaged children -- Psychological testing
- Description: The study investigates .the use of the Bender Gestalt Test (BGT) amongst low-educated adults. Three versions of the BGT are used in this study; the original 'copy' version as well as the 'immediate' and 'delayed'recall versions. This is done so as to expand the ability of the BGT to identify neurological impairment and to differentiate between this and functional impairment. A literature review explores the problems of standardization in the administration, scoring and application of all three versions of the test Suggestions are made to correct the problems identified and a novel system of scoring the recall versions are proposed, which allows for the comparison of results of the three versions of the test and which is based on Lacks's (1984) and Weiss's (1970) systems. Administration procedures were also developed to suit the context of the study. The copy, immediate and delayed versions of the BGT were administered to a group of 184 low-educated adults. Statistical analyses revealed significant education effects for the sample tested with regards to both test scores and performance time. The finding of an education effect for performance time is discussed at length, as some literature regards excessive time as a neurological indicator. An anomaly for the group with no education was found to exist, with the scores of these subjects not Significantly different from those with 4-6 years of education. Possible reasons for this were explored. In addition, the findings of this research revealed a plateau effect with those having less than 6 years of education scoring substantially lower than those with 7 years and more. The scores of adults with 7 and more years of education level out with no significant differences between educational levels. This appears to suggest that education effects rather than the developmental maturity level proposed by Koppitz, are involved. In addition, the scores of low-educated adults on the expanded Bender Gestalt Test were significantly lower than those of children with similar educational levels, in other studies. These findings and possible explanations are discussed. The study concludes by suggesting new research areas and emphasizing the urgent need for separate normative data on the expanded BGT for low-educated adults, and the establishment of appropriate 'cut-off' points.
- Full Text:
- Date Issued: 1996
- Authors: Dyall, Kate
- Date: 1996
- Subjects: Bender-Gestalt Test , Psychological tests , Brain-damaged children -- Psychological testing
- Language: English
- Type: Thesis , Masters , MA
- Identifier: vital:2968 , http://hdl.handle.net/10962/d1002477 , Bender-Gestalt Test , Psychological tests , Brain-damaged children -- Psychological testing
- Description: The study investigates .the use of the Bender Gestalt Test (BGT) amongst low-educated adults. Three versions of the BGT are used in this study; the original 'copy' version as well as the 'immediate' and 'delayed'recall versions. This is done so as to expand the ability of the BGT to identify neurological impairment and to differentiate between this and functional impairment. A literature review explores the problems of standardization in the administration, scoring and application of all three versions of the test Suggestions are made to correct the problems identified and a novel system of scoring the recall versions are proposed, which allows for the comparison of results of the three versions of the test and which is based on Lacks's (1984) and Weiss's (1970) systems. Administration procedures were also developed to suit the context of the study. The copy, immediate and delayed versions of the BGT were administered to a group of 184 low-educated adults. Statistical analyses revealed significant education effects for the sample tested with regards to both test scores and performance time. The finding of an education effect for performance time is discussed at length, as some literature regards excessive time as a neurological indicator. An anomaly for the group with no education was found to exist, with the scores of these subjects not Significantly different from those with 4-6 years of education. Possible reasons for this were explored. In addition, the findings of this research revealed a plateau effect with those having less than 6 years of education scoring substantially lower than those with 7 years and more. The scores of adults with 7 and more years of education level out with no significant differences between educational levels. This appears to suggest that education effects rather than the developmental maturity level proposed by Koppitz, are involved. In addition, the scores of low-educated adults on the expanded Bender Gestalt Test were significantly lower than those of children with similar educational levels, in other studies. These findings and possible explanations are discussed. The study concludes by suggesting new research areas and emphasizing the urgent need for separate normative data on the expanded BGT for low-educated adults, and the establishment of appropriate 'cut-off' points.
- Full Text:
- Date Issued: 1996
Validation of a predictor battery for engineering technicians
- Authors: Taylor, Jonathan Maclaren
- Date: 1980
- Subjects: Occupational aptitude tests , Engineering -- Vocational guidance , Vocational guidance , Prediction (Psychology) , Psychological tests
- Language: English
- Type: Thesis , Masters , MA
- Identifier: vital:3110 , http://hdl.handle.net/10962/d1004548 , Occupational aptitude tests , Engineering -- Vocational guidance , Vocational guidance , Prediction (Psychology) , Psychological tests
- Description: From summary: This study describes a procedure for predicting course success for certain first term engineering technicians. The aim of the study is to reduce the high attrition rate of trainee engineering technicians through the early identification of candidates who are likely to fail their first term of study. This identification is done by testing all applicants to the courses on a battery of psychological tests, and from this information estimating the applicants' first term course results. It is suggested that the student counsellors attached to the various Technikons integrate the suggested procedure into a flexible vocational guidance service for engineering technicians. It should be borne in mind that no validation study can predict future success with a hundred percent accuracy, and that the sample used in this study may be specific to the Witwatersrand Technikon.
- Full Text:
- Date Issued: 1980
- Authors: Taylor, Jonathan Maclaren
- Date: 1980
- Subjects: Occupational aptitude tests , Engineering -- Vocational guidance , Vocational guidance , Prediction (Psychology) , Psychological tests
- Language: English
- Type: Thesis , Masters , MA
- Identifier: vital:3110 , http://hdl.handle.net/10962/d1004548 , Occupational aptitude tests , Engineering -- Vocational guidance , Vocational guidance , Prediction (Psychology) , Psychological tests
- Description: From summary: This study describes a procedure for predicting course success for certain first term engineering technicians. The aim of the study is to reduce the high attrition rate of trainee engineering technicians through the early identification of candidates who are likely to fail their first term of study. This identification is done by testing all applicants to the courses on a battery of psychological tests, and from this information estimating the applicants' first term course results. It is suggested that the student counsellors attached to the various Technikons integrate the suggested procedure into a flexible vocational guidance service for engineering technicians. It should be borne in mind that no validation study can predict future success with a hundred percent accuracy, and that the sample used in this study may be specific to the Witwatersrand Technikon.
- Full Text:
- Date Issued: 1980
- «
- ‹
- 1
- ›
- »