Examining the impact of specific types of item-writing flaws on student performance and psychometric properties of the multiple choice question