Introduction
Internal validity is a condition of scientific research which depicts the degree to which a conclusion is warranted. It is common that systematic error minimization is the basis of warranting causal conclusions in research. An inference is concluded to be having internal validity if the binding relationship between the independent and dependent variable is described with precision. However, internal validity of research can be jeopardized (Salkind, 2010).
One of the ways that internal validity of a study can be jeopardized is by the selection bias. Selection bias is as a result of a bias involvement of more than a single person of the same type is involved in the experimental design of a study. For instance, it is of importance for people who will have to return the study questionnaires to be different from those who did not or will not return such questionnaires. It is of importance for a quantitative research project student for instance who might have volunteered in a research project to be different from those who did not but are part of the research project (Salkind, 2010). This difference is usually of importance as it is result based and achievement oriented. It is thus clear that the consideration of this difference should be taken into account because their effects to a research project are not known with precision as they are dynamic.
Differentiating these groups of the research study is of importance in reducing the bias that might be translated to the causal conclusion. The most commonly type of study that can be like affected is the quasi experimental studies in addition to convenience samples (Salkind, 2010). The main and solid way of preventing this selection bias which causes a thread to internal validity of causal conclusion of a study is the comparison between groups.
References
Salkind, N. J. (2010). Encyclopedia of research design. Thousand Oaks, Calif: Sage.
