Abstract:

Several authors have pointed out the frequent misuse of the term hypothesis in scientific texts, where such term is often conflated with predictions. In scientific practice, a hypothesis is a tentative explanation that accounts for an observed phenomenon. Testing a hypothesis involves hypothetical–deductive reasoning: from a proposed explanation, scientists derive predictions that can be contrasted through planned experiments or systematic observations. Thus, while predictions are expected outcomes derived from hypotheses, not all predictions are grounded in an explicit explanatory framework. This distinction is particularly relevant in school science, where students’ statements are often labelled as hypotheses even when they merely anticipate results without articulating underlying mechanisms. Within inquiry-based science education, formulating hypotheses constitutes a central scientific practice. Research shows that children and adolescents are capable of making predictions from an early age and progressively developing the ability to construct hypotheses that incorporate causal reasoning. However, classroom practices do not always explicitly differentiate between predicting and explaining, which may limit students’ engagement in authentic scientific reasoning. This distinction becomes especially significant in the context of STEAM education. The STEAM framework (Science, Technology, Engineering, Arts, and Mathematics) promotes interdisciplinary learning and the development of STEM competencies, including modelling, investigation design, data analysis, and evidence-based argumentation. From this perspective, generating hypotheses should not be understood merely as anticipating outcomes, but as constructing explanatory models that can be empirically tested. When students design experiments—such as investigating the presence of microorganisms on different surfaces—the quality of their hypotheses can reveal the extent to which they mobilize scientific models, in this case the model of living beings, to explain observable phenomena. In this context, the current study examines the development of scientific skills through the integration of inquiry-based practices. 82 students of 1st year of secondary education (12-13 years old) participated in the design of an experiment related to the microorganisms during winter 2026. Students worked in groups (21 groups of 3-5 people in total). The written reports produced by each group were examined. In the report, students were explicitly asked to formulate a hypothesis and a prediction separately, and to explain the experimental procedure. This distinction was intentionally required in order to analyse whether students were able to differentiate between a hypothesis (as a possible explanation) and a prediction (as the expected outcome of the test). The results show that although the majority of the groups were able to make predictions (17 groups) only 5 groups formulated a proper hypothesis. In all this cases, these hypothesis referred to the living being model. It is concluded that the main difficulty students face is in distinguishing hypotheses from predictions. Given that students had been shown concrete examples of both hypotheses and predictions but still struggled to distinguish them, future work should explore instructional designs that more effectively support students in linking predictions to underlying explanatory models within inquiry-based STEAM activities.

Keywords:

Hypotheses and predictions, scientific reasoning, inquiry-based learning.