區辨效度 (discriminative validity / known-group validity)
*區辨效度之概念:量表施測結果可區辨不同屬性病人/病人與健康對照組的程度 (between group difference) (McCarthy et al., 2002; Netemeyer, Bearden, & Sharma, 2003) 。
*區辨效度之價值:
1. 偵測出功能正常與不同程度失能之個案
2. 利於臨床人員「對症下藥」
3 於橫斷面研究中,可比較個案間之差異 (Brock, Goldie,& Greenwood, 2002; Mancuso, Peterson, & Charlson, 2001)
*區辨效度之研究設計:
1. floor/ceiling effects:Small≦15% (McHorney & Tarlov, 1995) 或≦20% (Hobart et al., 2001);Moderate≧15% (McHorney & Tarlov, 1995) 或≧20% (Hobart et al., 2001)
2. t檢定:|t|≥ 1.96 (p ≤ 0.05)
3. ANOVA:F值達顯著 (p ≤ 0.05)
1. 將量表施測於一組人
2. 依據某一特質將受測者分群:可依據另一測驗之分數或臨床事件 (Guyatt, Feeny, & Patrick, 1993; Mancuso, Peterson, & Charlson, 2001)
3. 檢定不同群體是否在測驗結果上是否有所差異
*區辨效度之資料分析方法:
1. floor / ceiling effect
˙floor effect:得到量表最低可能分數之樣本數 / 總樣本數 *100%
˙ceiling effect:得到量表最高可能分數之樣本數 / 總樣本數 *100%
2. 獨立樣本t檢定 (independent t test)
˙變異數同質:
˙變異數不同質:
3. 變異數分析 (ANOVA)
*區辨效度之SPSS分析步驟:
1. floor/ceiling effect:
˙點選「分析 (Analyze)」→「敘述統計 (Descriptive Statistics)」→「次數分配表 (Frequencies)」
˙進入對話框,選取欲分析的題目移至清單中
˙點選「確定 (OK)」
2. 獨立樣本t檢定:
˙點選「分析 (Analyze)」→「比較平均數法(Compare Means)」→「獨立樣本T檢定(Independent Samples T Test)」
˙進入對話框,選取欲分析的「檢定變數 (Test Variable)」與「分組變數 (Grouping Variable)」移至清單中
˙於「定義組別 (Define Groups)」中輸入欲進行對比的分類變數類別→點選「繼續 (Continue)」
˙點選「確定 (OK)」
3. ANOVA:
˙點選「分析 (Analyze)」→「比較平均數法(Compare Means)」→「單因子變異數分析(One-way ANOVA)」
˙進入對話框,選取欲分析的「依變項清單 (Dependent List)」與「因子 (Factors)」移至清單中
˙點選「確定 (OK)」
*區辨效度之數值判別標準:1. floor/ceiling effects:Small≦15% (McHorney & Tarlov, 1995) 或≦20% (Hobart et al., 2001);Moderate≧15% (McHorney & Tarlov, 1995) 或≧20% (Hobart et al., 2001)
2. t檢定:|t|≥ 1.96 (p ≤ 0.05)
3. ANOVA:F值達顯著 (p ≤ 0.05)
三、主要參考文獻
2. Guyatt, G. H., Feeny, D. H., & Patrick, D. L. (1993). Measuring Health-Related Quality of Life. Annals of Internal Medicine, 118, 622-629.
3. Hobart, J. C., Lamping, D. L., Freeman, J. A., Langdon, D. W., McLellan, D. L., Greenwood, R. J., & Thompson, A. J. (2001). Evidence-based measurement: which disability scale for neurologic rehabilitation? Neurology, 57, 639-644.
4. Mancuso, C. A., Peterson, M. G.,& Charlson, M. E. (2001). Comparing discriminative validity between a disease-specific and a general health scale in patients with moderate asthma. Journal of Clinical Epidemiology, 54, 263-274.
5. McCarthy, M. L., Silberstein, C. E., Atkins, E. A., Harryman, S. E., Sponseller, P. D.,& Hadley-Miller, N. A. (2002). Comparing reliability and validity of pediatric instruments for measuring health and well-being of children with spastic cerebral palsy. Developmental Medicine & Child Neurology, 44, 468-476. 6. McHorney, C. A., & Tarlov, A. R. (1995). Individual-patient monitoring in clinical practice: are available health status surveys adequate? Quality of Life Research, 4, 293-307. 7. Netemeyer, R. G., Bearden, W. O., & Sharma, S. (2003). Scaling Procedures: Issues and Applications. London: Sage Publications.
3. Hobart, J. C., Lamping, D. L., Freeman, J. A., Langdon, D. W., McLellan, D. L., Greenwood, R. J., & Thompson, A. J. (2001). Evidence-based measurement: which disability scale for neurologic rehabilitation? Neurology, 57, 639-644.
4. Mancuso, C. A., Peterson, M. G.,& Charlson, M. E. (2001). Comparing discriminative validity between a disease-specific and a general health scale in patients with moderate asthma. Journal of Clinical Epidemiology, 54, 263-274.
5. McCarthy, M. L., Silberstein, C. E., Atkins, E. A., Harryman, S. E., Sponseller, P. D.,& Hadley-Miller, N. A. (2002). Comparing reliability and validity of pediatric instruments for measuring health and well-being of children with spastic cerebral palsy. Developmental Medicine & Child Neurology, 44, 468-476.
1. 具區辨效度的評估工具並無法讓臨床人員瞭解個案之失能嚴重程度。區辨效度是between group之比較,因此無法知道個別樣本之評估結果代表的嚴重程度。除非此量表再經過sensitivity之驗證,找到cut-off point。
2. 驗證區辨效度的方法還包括:取前20%與後25%的樣本,求此2組樣本之量表平均分數有無差異。但我個人認為,除非收案之樣本變異太小,否則此方法可能很容易達顯著,是否有高估量表區辨效度之虞?
分析方法 looks good!
回覆刪除二者所代表的概念/意義,如何解釋(或限制)?可找時間補充。
謝謝老師的建議,由於概念與價值是前一次報告主題,本次未再重複。
刪除我已將二主題區分開來,並綜合兩次的報告內容。