2011年11月29日 星期二

項目間一致程度

Item internal consistency is tested by examining the correlations between an item and the scale score computed from all other items in that scale (item-scale correlation after correction for overlap).

Reference: Ware, J. E., & Barbara G. (1998). Methods for testing data quality, scaling assumptions, and reliability: The IQOLA project approach. J Clin Epidemio, 51, 945-952.

中文解釋:計算項目分數與量表分數(不包括此項目分數)之相關。

判斷標準:每個項目和所屬量表間之相關 > 0.4

臨床意義:分析項目是否和潛在構念呈現線性關係,假如每個項目和潛在構念之項目間相關程度高於0.4,表示每個項目測量同一個潛在構念,所有項目分數的加總有其意義,總分可代表潛在構念。

Model paper: Sullivan, M., Karlsson, J., & Ware, J. E. (1995). The swedish SF-36 healthe survey-I. Evaluation of data quality, scaling assumptions, reliability and construct validity across general populations in Sweden. Soc Sco Med, 41(10), 1349-1358.

研究設計:讓一群受測者填寫量表,分析項目與所屬量表間相關。

2 則留言:

  1. 請問Item internal consistency與Internal consistency的差異或關連性?

    回覆刪除
  2. 謝謝提問。

    我的想法:Item internal consistency與Internal consistency的差異在於Item internal consistency為屬於效度,而Internal consistency屬於信度。

    回覆刪除