Jette A M, Cummings K M, Brock B M, Phelps M C, Naessens J
Health Serv Res. 1981 Spring;16(1):81-98.
A wealth of research using the Health Belief Model provides empirical evidence of the model's utility in predicting health, illness, and sick role behaviors. Until recently, however, little attention has been paid to the important issues of the validity and reliability of measures used to assess various health belief dimensions. Using factor analysis, our study demonstrates that moderately reliable indices covering a wide spectrum of distinct health beliefs can be constructed and then replicated across independent samples. The factor analysis approach revealed that condition-specific measures of perception of susceptibility and severity and situation-specific measures of perceived barriers are empirically distinct from general measures of these beliefs. We therefore recommend caution in mixing general and specific questionnaire items within the same index when measuring these beliefs. A factor representing perceptions of health threat emerged, but its composition require further clarification. The degree of similarity between the factor structures in the two independent samples provides support for the existence of independent health belief dimensions.
大量运用健康信念模型的研究为该模型在预测健康、疾病及患病角色行为方面的效用提供了实证依据。然而，直到最近，用于评估各种健康信念维度的测量方法的有效性和可靠性这一重要问题却很少受到关注。通过因子分析，我们的研究表明，可以构建涵盖广泛不同健康信念的适度可靠指标，然后在独立样本中进行复制。因子分析方法显示，特定状况下对易感性和严重性的感知测量以及特定情境下对感知障碍的测量在实证上与这些信念的一般测量有所不同。因此，我们建议在测量这些信念时，在同一指标中混合使用一般和特定问卷项目时要谨慎。一个代表健康威胁感知的因子出现了，但其构成需要进一步澄清。两个独立样本中因子结构的相似程度为独立健康信念维度的存在提供了支持。
