The study was carried out to evaluate the performance of undergraduate students in Test and Measurement Achievement test with samples from probabilistic and non-probabilistic sampling procedures. Three research questions and one hypothesis guided the study. A sample of 200 students (100 from proportionate stratified random representing probabilistic and 100 from purposive representing non-probabilistic). Test and Measurement Achievement test was the instrument used for the study. It has face and content validities. It has a reliability coefficient of 0.79 obtained through Kuder-Richardson formula 20. The data collected with the instrument were analysed in line with the research questions and hypothesis. Mean was used to answer the research questions while Z-test was used to test the hypothesis at 0.05 level of significance. The results showed that the undergraduate students who were selected through the probabilistic sampling procedure performed better than sample of students from the non-probabilistic procedure. There was a significant difference between the sample of students from probabilistic and non-probabilistic sampling procedure. Recommendations were made. It was recommended that researchers and evaluators should always use probabilistic or random sampling procedure to carry out their studies for generalisation or judgement. In addition, simple random sampling of balloting or random numbers should always be used and applied when using stratified sampling procedure to select individuals for a study.