Internal reliability is a problem as several different questions are often used to measure the same trait or attitude (e.g scales used by Adorno on fascism, ethnocentrism, etc., contained many items that collectively measured these concepts, but didn't equally measure the same concept)
Use of split-half method to ensure internal reliability, which involves splitting the questions into two halves and comparing the findings from both halves during analysis
This method should be achieve the same score from both halves, and if they do not it means some of the questions measure a different concept