This paper focused on the analysis of test items constructed in the paper of teaching Physical Science for B.Ed. class. It involved the analysis of difficulty level and discrimination power of each test item. Item analysis allows selecting or omitting items from the test, but more importantly item analysis is a tool to help the item writer improve an item. There were 100 objective type questions in the achievement test and was administered to a sample of 100 students from B.Ed. colleges. The result of the work showed that most of the test items fall in the acceptable range of difficulty index and discrimination index. However, 17 of the test items was excluded due to very high or very low difficulty level and poor discrimination power. Purposive sampling was adopted for this study and 100 B.Ed. Physical Science students were taken from private B.Ed. Colleges in Pondicherry, India. A test of 100 items was used for data collection. Using the findings relationship between the difficulty index of each item and the corresponding discrimination index is carried out using the Pearson correlation formula. From the analysis, it was found that item discrimination power increased with the increase in difficulty value but got decreased for very easy and very difficult test items. This work is significant for the researcher in framing test items with optimum difficulty and discrimination power. The findings also show that, most of the items were falling in the acceptable range of difficulty and discrimination level, however, some items were rejected due to their poor discrimination index.