Internal consistency reliability: this is used when several observations are made to obtain a score for each participant for example, if participants complete a test with several items, or if the researcher makes several independent observations of their behavior. internal consistency is high if each item or behavioral observation correlates with the other observations that is, if all the items are measuring the same thing. an internal consistent scale measures one construct with several independent observations. the more observations we make to obtain a score, such as a classroom test, the greater will be the internal consistency reliability of the score. using only one or two questions will not provide the same level of consistency compared to a multi choice test.