Validity, reliability and item analysis are critical to the process of evaluating the quality of an educational measurement. The present study evaluates the quality of an assessment constructed to measure elementary school student's achievement in English. In this study, the survey model of descriptive research was used as a research method. Students of both genders (male/female) constitute the population of the study. To measure the achievement level of students, a test of English language as a subject was administered to the sample of 150 students (75 males and 75 females) of 8th grade. They were selected by convenient sampling method from 10 (5 private sectors and 5 public sectors) schools in Pakistan. English Achievement Test for elementary students comprises of 50 multiple-choice items (developed by the author), was used. The collected data were analyzed through descriptive statistics, Z-test and item analysis. The results of the study indicate that male student's achievement level in English is better than the female students at elementary level in selected schools. (Contains 5 tables and 4 figures.)