Developing an instrument to measure tinkering and technical self-efficacy in engineering

An instrument to measure tinkering and technical self-efficacy was developed based on recommendations by Bandura. Self-efficacy is defined as an individual's beliefs about their ability to engage in activities that will result in successfully attaining specific goals. Thus, selfefficacy is context and skills specific rather than a global judgment of ability. Tinkering and technical self-efficacy in engineering are important because individuals with low self-efficacy in these areas are more likely to leave engineering majors independent of their levels of achievement. This is especially true for women. Consequently, the development of an instrument with good predictive power can be a useful tool for creating interventions for retention in engineering majors. The first phase of this study to develop a predictive instrument was to establish the content validity. During this phase, we used two open-ended questions asking the respondents to identify tinkering and technical skills. Eight hundred and seventy-one statements were obtained from a volunteer expert sample of 101 respondents. A count of statements with the same meaning was conducted and the most frequently mentioned were used to write questions. Approximately half were worded positively and half negatively. The tinkering and technical scales each consisted of 30 questions. The second phase of the study started with the development of a Likert-scale survey using these statements. The instrument was given in freshman design classes (n=84 students). Students were asked to rate themselves on a Likert scale from not descriptive of me (0) to very descriptive of me (5). The analysis indicated that students had moderate self-efficacy in terms of technical skills. Mean item scores were between 3.2 and 3.7. Students rejected negatively worded items on both scales as not descriptive with mean scores between 0 and 2. Students reported the least technical self-efficacy on the item "I can statistically model a process". Ratings on the tinkering scale included items with means scores above 4.0. These items were: "I can think outside the box", "I know how to use tools", I want to know how things work and how to make them better", and "I have the persistence to complete a project". The reliability of the tinkering scale was .87 and the reliability of the technical scale was .80. A factor analysis found three factors for tinkering. Factor one was labeled knowledge and experience, factor two creativity and curiosity and factor three knowledge and skills. There were also three factors for technical skills. Factor one was labeled technical knowledge, factor two understanding theories and models and factor three systems and how things work.