Study of Validity and Reliability in - Research Paper Example

Validity in research is achieved through the internal and external validity of the study. Internal validity refers to the outcome of the study was based on the function of the program. A study has internal validity if the study's outcome is a function of the approach being tested rather than results of the causes not systematically dealt with (Kumar, 2004). Internal validity is justified by the conclusions we have as researchers when we can control the threats of other variables (intervening variables, or moderating variables or extraneous variables).

The more you reduce the nuisances (other variables) affecting the study, the more you attain internal validity. There are four forms of validity, and to which extent many authors agree. Colorado State University (2009) lists the forms of validity as the face, content, construct, and criterion-related Face validity is simply a criterion in which you are to determine whether the instrument appears to be able to measure what it intends to measure. It is the most basic and the first point of determining validity.

You only need to determine if it can measure or not. For example, by looking at it, the tape measure gives you the impression that the instrument can measure height. When it comes to behaviour, we can determine that one’s aggressiveness is a valid way of measuring their anger. Once this first stage has been determined, the researcher can go ahead to determine content validity. In content validity, the main concern is how adequately this instrument covers all facets of the concept? I will use the example of a questionnaire that seeks to determine whether a child is suffering from Attention Disorder.

The questions must be able to serve as indicators of the problem. The questions must be able to cover all aspects that are related to the disorder. If it does not, then it has low content validity. Columbia Center for New Media Teaching and Learning uses a driving test to illustrate the next step in determining the validity of a measuring scale. Criterion-related validity reflects the success of measures used for prediction or estimation.