In D-KEFS PT, eight common and uncommon proverbs are presented to the examinee in two formats: Free Inquiry and Multiple Choice.
Given that there were no other executive tests previously translated or developed for the population of the current study (to be used for convergent validation of the adapted tests), we selected BADS and translated the test instructions using the same forward strategy conducted for the D-KEFS instructions (Hambleton & Patsula, 1999).
The adapted materials for eight D-KEFS tests described above, along with the computerized Persian D-KEFS CWIT (Daryadar, 2014) were administered.
Given this fact and the lengthy time of administration in Phase 2, besides using the Computerized CWIT and the break, we made two of the less psychometrically established D-KEFS tests optional (WCT and PT; these tests were also not included in our assessment of TBI patients, Ghawami et al.
In the validity analysis, intercorrelations of measures within individual D-KEFS tests and correlations between measures of different D-KEFS tests were computed; also, correlations of D-KEFS tests with the BADS scores were calculated.
Descriptive statistics for selected D-KEFS and BADS variables, including means and standard deviations, are displayed in Table 1.
The internal consistency and test-retest reliability values for the primary scores of the D-KEFS tests are presented in Table 2.
As a response to the need for adapting well-established western neuropsychological tests for assessments of TBI patients in Iran, particularly in the south-eastern region, the current study was aimed mainly to develop and preliminarily validate a Persian version of the D-KEFS tests for assessments of TBI patients in Zahedan, Iran.
In the reliability analysis, except for CWIT, the same internal consistency and test-retest methodologies as in the original standardization study of D-KEFS (Delis et al.
Schmidt (2003) criticized the reported reliabilities in the D-KEFS manual, pointing out that only 17% of the values were above a .
On the other hand, consistent with the findings of Delis et al (2001c), correlations between executive scores from various D-KEFS tests were generally weak.
The weak correlations between the D-KEFS tests in the current study are not solely attributable to reliability, given the higher reliabilities for most of the executive measures.