青少年預防事故傷害發生之自我效能量表的發展與評估

The Development and Evaluation of the Self-Efficacy Scale for the Adolescent Unintentional Injuries

自我效能是Bandura社會認知理論最主要的重點所在，在過去的十幾年中已逐漸地為社會、健康與行為科學研究者應用於各種健康行為與問題的探討上，並作為發展健康促進計畫的重要依據。雖然事故傷害是目前青少年所面對最嚴重的健康問題，但是將自我效能實際應用到防範青少年事故傷害發生之研究卻很少。然而要瞭解自我效能在青少年事故傷害問題上所扮演的角色，就必須先發展信、效度俱佳的青少年預防事故傷害發生之自我效能量表，此乃本研究之主要目的。此自我效能量表是以我國青少年常見的五類事故傷害問題（即交通事故、墜落、中毒、火災、溺水）為範圍所發展出來的，以臺灣地區八十四年度公、私立國中、高中與高職之日間部學生為研究對象。對於此量表的效度分析，除了考慮了專業上的內容效度之外，亦選擇了建構效度與關聯效標效度。至於此量表的信度分析，主要是以內部一致性Cronbach's α信度係數為主。結果發現此自我效能量表是由三個彼此具有密切關聯的分量表所構成，分別為「防範個人事故發生的自我效能」、「防範交通事故發生的自我效能」， 以及「防範常見事故發生的自我效能」。在效度分析上，對於各級學校與男女等不同的研究對象，不論是在試探或驗證階段中，此量表所顯示的建構效度均令人相當滿意，而至於關聯效標效度，此量表對於行為意向與行為都表現出很高的預測力，並且與其他相關的素質變項（認知、情感、價值觀念與信念）也呈現了很強的關聯性。在信度分析上，此量表在試探與驗證階段亦均具有相當高的內部一致性信度，而且這兩個階段之結果也指出此量表具有相當程度的穩定性。因此，此自我效能量表能有效地應用於探討青少年事故傷害問題。

Self-efficacy, one of the most important components in Bandura's social cognitive theory, has been widely applied by social, health, and behavioral scientists to various health behavior research, and served one guide for developing effective health promotion interventions. Though the unintentional injuries are the most serious issue facing the adolescent, relatively few studies concerning self-efficacy are found in the field of the adolescent unintentional injuries. The major purpose of the study is to develop a scale that measures self-efficacy validly and reliably for the adolescent unintentional injuries in order to understand the impact of self-efficacy on the adolescent unintentional injuries. After having been developed to include the five common unintentional injuries for the adolescent, the scale was administered to the junior high, senior high, and vocational school students in Taiwan. In addition to the content validity, the construct validity and the criterion-related validity were considered to establish the validity of the scale, and the Cronbach's α coefficients were selected for the internal consistency reliability of the scale. It is found that the multidimensional scale is mainly composed of mutually related three factors. The construct validity of the scale is better in both of the exploratory and confirmatory stages. Not only has the scale much higher predictive power for behavioral intentions and behaviors, but also it has stronger associations with those predisposing variables (cognitive, feeling, value, and belief). The reliability of the overall scale and three subscales is very satis-factory, and so stable on two occasions. In conclusion, this scale is so effective that it can be used to understand the influence of self-efficacy on the adolescent unintentional injuries.