The 2013 curriculum uses a classroom-based assessment. The curriculum contains various supporters, one of which is an assessment. Assessment by teachers is used to measure the achievement of learning objectives. But in the real world, a teacher has difficulty in developing assessment planning, assessment implementation, and processing of assessment result, and teacher has not analyzed comprehensively about validity and reliability in question, and analysis of evaluation instrument conducted by the teacher is only limited to analyze difficulty level. Thus, the problem that is made to know the cognitive level of students has not known the quality. The objectives of this study were (1) to know the reliability of cognitive assessment instruments developed, (2) to know the level of difficulty, (3) to know the differentiating power, and (4) to know the effectiveness of the cultivators. The method used in this research is research and development (R & D) developed by Borg & Gall. Data analysis technique that is done is instrument validation test with construct validity through judgment expert, reliability test using ITEMAN with formula CR-20. This study resulted in (1) the reliability of the question items of 0.740, (2) the degree of difficulty there are three levels, namely difficult 23%, medium 70%, and easy 7%, (3) distinguishing power on the grain matter analysis resulted in 20%, 7% of the questions are sufficient, 60% is good, and 13% are very good, and (4) the effectiveness of the deception is 82% working and 18% not working.