Construction of A Graded Naming Test and Its Applicability in Taiwanese Participants

• Main Authors: Pei Chi Tsai, 蔡珮琦
• Other Authors: M. Y. Ho
• Format: Others
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/13189067033479794008

碩士 === 長庚大學 === 職能治療學系 === 100 === Naming difficulty is a common feature of patients with anomia or brain injuries; naming ability also tends to decline with normal aging. The assessment of naming in patients has mainly relied on the naming tests developed in other countries or a sub-test of an aphasia battery. The problems with cross cultural adaptation, and the fact that these tests are not demanding enough might have hampered their sensitivity to detect mild naming difficulty. The purpose of this study is to establish a graded naming test similar to that was developed by Warrington and McKenna (1983). It specifically aimed for people whose first language was either Mandarin or Taiwanese. In addition, the local sample were collected, and the clinical applicability of the test was validated. After a preliminary testing procedure, the Rasch analysis was performed on the data collected from volunteers (N = 283, aged between 16-75 years); and a 30-item version of the test was established based on the results of the Rasch analysis. The psychometric analyses showed the test has good internal consistency (.87), test-retest reliability (.96), and split-half reliability (.92). The naming performance of all volunteer also significantly correlated with a measure of the fluid intelligence and the scores for other verbal tests (all correlation coefficients were greater than .67). The naming ability appeared to decline with advancing age, but positively correlated with years of education; and the performance of males was significantly better than that of females. In addition, age, years of education, gender were used as predictor variables, and the score of the graded naming test as criterion variable for deriving multiple regression equations by the stepwise method. The effects of age, education and gender were all statistically significant; and could account for 53% variance of the naming score. The equation was also used to determine if there was a significant discrepancy between the estimated naming score and obtained score for each participant. The naming performance was compared across four age- and education-matched groups: normal volunteers (NV, n = 30), normal clinical group (NC, n = 28), mild cognitive impaired group (MCI, n = 26), and mild dementia group (MD, n = 24). The results showed that naming abilities of the MCI and MD groups were impaired. In addition, by analyzing the error type of health volunteers and clinical participants, the results of the present study are in support of the notion that semantic memory appears to be compromised by ageing or cerebral degeneration.