Validity in Research - Essay Example

Qn. 1

Variables will predict an outcome based on the information that these variables, or a set of variables received from other variables. Once this is achieved, the measurement of this prediction, becomes criterion validity in research, and there are various ways that it can be established. First, a researcher can standardize this measure against another standard that is conventional and known. For example, a new test that measures the intelligence of humans can be measured against the IQ test, which is a known standard for measuring intelligence levels in humans. Secondly, a researcher can establish criterion validity by measuring the tool against itself. In this case, a researcher can establish criterion-based validity by investigating the way that the tool will be able to predict measures in the future.

Qn. 2

There are four forms of reliability. Among these four forms of reliability is internal consistency reliability, which investigates the level to which disparate instruments that are investigating the same phenomenon produced results that were different. Consequently, this form of reliability is further divided into two sub-types with one of the sub-types known as split-half reliability. As the name suggests, spilt-half reliability is achieved through splitting into two halves of the items in a test whose intention is to establish an investigation in the same area of knowledge. In effect, this forms two sets of items that will undergo testing in order to establish the reliability of the tool. In this case, the two sets of items undergo the testing with similar tools with their scores then computed, and the researcher finally obtains the split-half reliability by a correlative determination of the ‘set’ scores between this set of items.