Samakouri M, Bouhos G, Kadoglou M, Giantzelidou A, Tsolaki K, Livaditis M
Post-graduale Program in Social Psychiatry, Alexandroupolis, Psychiatric Department, Medical School, University of Thrace, Alexandroupolis, Thrace, Greece.
Psychiatriki. 2012 Jul-Sep;23(3):212-20.
Self-rating Anxiety Scale (SAS), introduced by Zung, has been widely used in research and in clinical practice for the detection of anxiety. The present study aims at standardizing the Greek version of SAS. SAS consists of 20 items rated on a 1-4 likert type scale. The total SAS score may vary from 20 (no anxiety at all) to 80 (severe anxiety). Two hundred and fifty four participants (114 male and 140 female), psychiatric patients, physically ill and general population individuals, aged 45.40±11.35 years, completed the following: (a) a demographic characteristics' questionnaire, (b) the SAS Greek version, (c) the Spielberg's Modified Greek State-Trait Anxiety Scale (STAI-Gr.-X) and (d) the Zung Depression Rating Scale (ZDRS). Seventy six participants answered the SAS twice within a 12th-day median period of time. The following parameters were calculated: (a) internal consistency of the SAS in terms of Cronbach's α co-efficient, (b) its test-retest reliability in terms of the Intraclass Correlation Coefficient (ICC) and (c) its concurrent and convergent validities through its score's Spearman's rho correlations with both the state and trait subscales of STAI-Gr X and the ZDRS. In addition, in order to evaluate SAS' discriminant validity, the scale's scores of the three groups of participants (psychiatric patients, physically ill and general population individuals) were compared among each other, in terms of Kruskall Wallis and Mann Whitney U tests. SAS Cronbach's alpha equals 0.897 while ICC regarding its test-retest reliability equals 0.913. Spearman's rho concerning validity: (a) when SAS is compared to STAI-Gr.-X (state), equals it 0.767, (b) when SAS is compared to STAI-Gr. X (trait), it equals 0.802 and (c) when SAS is compared to ZDRS, it equals 0.835. The mentally ill scored significantly higher in SAS compared to both the healthy and the general population. In conclusion, the SAS Greek version presents very satisfactory psychometric properties regarding its reliability and validity as well.
由宗氏提出的自评焦虑量表（SAS）已在研究和临床实践中广泛用于焦虑症的检测。本研究旨在对SAS的希腊语版本进行标准化。SAS由20个项目组成，采用1-4李克特量表评分。SAS总分范围为20分（无焦虑）至80分（严重焦虑）。254名参与者（114名男性和140名女性），包括精神科患者、躯体疾病患者和普通人群，年龄为45.40±11.35岁，完成了以下内容：（a）一份人口统计学特征问卷，（b）SAS希腊语版本，（c）斯皮尔伯格修订的希腊状态-特质焦虑量表（STAI-Gr.-X），以及（d）宗氏抑郁评定量表（ZDRS）。7名参与者在12天的中位时间内对SAS进行了两次回答。计算了以下参数：（a）根据克朗巴哈α系数计算SAS的内部一致性，（b）根据组内相关系数（ICC）计算其重测信度，以及（c）通过其得分与STAI-Gr X的状态和特质子量表以及ZDRS的斯皮尔曼等级相关系数计算其同时效度和收敛效度。此外，为了评估SAS的区分效度，根据克鲁斯卡尔-沃利斯检验和曼-惠特尼U检验，对三组参与者（精神科患者、躯体疾病患者和普通人群）的量表得分进行了相互比较。SAS的克朗巴哈α系数为0.897，其重测信度的ICC为0.913。关于效度的斯皮尔曼等级相关系数：（a）当SAS与STAI-Gr.-X（状态）比较时，为0.767，（b）当SAS与STAI-Gr. X（特质）比较时，为其0.802，（c）当SAS与ZDRS比较时，为0.835。与健康人群和普通人群相比，精神疾病患者的SAS得分显著更高。总之，SAS希腊语版本在可靠性和效度方面均呈现出非常令人满意的心理测量学特性。
