In the article by Gadd and Phipps (2012), they refer to the challenges faced by psychological and, specifically, neuropsychological assessment. Their study focused on a preliminary standardisation of the Wisconsin Card Sorting Test (a non-verbal measure) for Setswana-speaking university students. The US normative sample is described as participants (N = 899) aged 18 to 29 years who were screened beforehand to exclude individuals with a history of neurological, learning, emotional and attention difficulties. The South African sample consisted of university students (N = 93) from both genders, between the ages of 18 and 29, who were screened in terms of hearing and visual …show more content…
Which factors could have affected the test-retest coefficients?
(1) transfer effects such as practice and memory
(2) changes in the wellness construct over time
(3) error due to content sampling
(4) a lack of internal consistency Cronbach’s Alpha coefficients imply that the test is internally consistent
Test-retest reliability coefficient is a measure of how consistent the results of a test are over time. The retest correlation is sensitive to the nature of the sample used to estimate it.
Article Gradidge page 8
The test-retest method is only appropriate when the construct being measured remains relatively constant over time (Murphy & Davidshofer, 1994). The wellness construct is not one that we expect to remain stable over time, such as personality or IQ. Rather, we would expect individual levels of wellness to reflect changing lifestyle commitments and life circumstances.
In the same study, Gradidge and De Jager (2011) reported that no practically significant differences were found between the mean wellness of demographic groups (gender, language or academic year level). This is in line with wellness theory and previous research. Factor analysis also revealed the expected factor structure. Critically evaluate these findings by answering questions 5 and 6.
These findings reflect positively on