Introductory physics lab instructors’ perspectives on measurement uncertainty

Introductory physics lab courses serve as the starting point for students to learn and experience experimental physics at the undergraduate level. They often focus on measurement uncertainty, an essential topic for practicing physicists and a foundation for more advanced lab learning. As such, measurement uncertainty has been a focus when studying and improving introductory physics lab courses. There is a need for a research-based assessment explicitly focused on measurement uncertainty that captures the breadth of learning related to the topic, and that has been developed and documented in an evidence-centered way. In this work, we present the first step in the development of such an assessment, with the goal of establishing the breadth and depth of the domain of measurement uncertainty in introductory physics labs. We conducted and analyzed interviews with introductory physics lab instructors across the US, identifying prevalent concepts and practices related to measurement uncertainty, and their level of emphasis in introductory physics labs. We find that instructors discuss a range of measurement uncertainty topics beyond basic statistical ideas like mean and standard deviation, including those connected to modeling, another lab learning goal. We describe how these findings will be used in the subsequent development of the assessment, called the Survey Of Physics Reasoning On Uncertainty Concepts In Experiments (SPRUCE).

Figure 1

A diagram illustrating our coding scheme. The green boxes represent the types of a priori codes that were applied in the first pass. These codes were used to identify excerpts for later analysis. The blue boxes represent the emergent codes that were applied in the second pass. In that pass, the excerpts pertaining to each green box were coded using the emergent code book(s) of the blue box(es) in which they appear in this diagram.

Figure 2

The number of interviews in which instructors mentioned concepts (a), and only those concepts they identified as important (b), broken down by the emergent codes in Table 4. As a reminder, we interviewed 22 instructors ($N_{\mathrm{tot}}=22$).

Figure 3

The number of interviews ($N_{\mathrm{tot}}=22$) in which instructors mentioned practices broken down by the emergent codes in Table 5.

Figure 4

The number of interviews ($N_{\mathrm{tot}}=22$) in which instructors mentioned topics students already know, separated by whether these topics supported (a) or hindered (b) learning, broken down by the emergent codes in Tables 4 and 5.

Figure 5

The number of interviews ($N_{\mathrm{tot}}=22$) in which instructors mentioned topics students find difficult, broken down by the emergent codes in Tables 4 and 5.

Figure 6

The number of interviews ($N_{\mathrm{tot}}=22$) in which instructors mentioned topics that they chose not to cover or emphasize, broken down by the emergent codes in Tables 4 and 5.

Figure 7

The number of interviews ($N_{\mathrm{tot}}=22$) in which instructors mentioned ways of assessing measurement uncertainty, broken down by the emergent codes in Tables 4 and 5.

Figure 8

The number of interviews ($N_{\mathrm{tot}}=22$) in which instructors mentioned various types of lab activities involving measurement uncertainty.