This study compares the scientific and alternative conceptions of energy and momentum of university first-year science students in Ethiopia and the US. Written data were collected using the Energy and Momentum Conceptual Survey developed by Singh and Rosengrant. The Concentration Analysis statistical method was used for analysing the Ethiopian data from 72 students and extending the analysis of the American data from 352 students in the Singh and Rosengrant study. Low levels of scientific conceptions of students' responses were identified for both groups of respondents. A three-level categorisation of students' responses showed that the distribution of responses to 80% (ETH) and 52% (US) of the items represented the null-model or random state. The distribution of responses to 20% (ETH) and 48% (US) of the items represented the bi-model state and none of the items was responded to in a purely correct or alternative conception one-model state. The analysis of the data revealed a variety of students' alternative conceptions and phenomenological primitives (p-prims) regarding energy and momentum. The findings imply that teachers should explore the use of the Concentration Analysis method in identifying alternative conceptions and p-prims in physics to support their students' learning.