Research internal validity vs external validity

In this post we will briefly discuss difference between interval validity and external validity when it comes to research methodology. What are they? How they arise? What are the sources of each? How to improve research internal and external validity?

Internal validity

Internal validity refers to the closeness the sample data analysis results is to answering the research question with minimal external effects. This means the researcher carefully minimizes the effect of what is called extraneous variable. Extraneous variables are external factors that provide parallel explanation which rivals the true research variable. Extraneous variables are also called artifacts

During data analysis stage, if the research has artifact present and not detected by the researcher, the research is said to be lacking internal validity. This also impacts the validity of the research hypothesis to draw correct conclusion.

Let us take an example of research which lacks external validity to help understanding. Suppose your research is concerned with answering why students who have internet at home score better than students who dont have. An artifact (extraneous variable) that can impact internal validity of your research might be students who have home internet access might also have home library and the home library variable will challenge the home internet explanation.

With internal validity concept now understood, now let us discuss some possible sources of research internal validity

1. Research participant behavioral change during the study process

An example is if researcher is subjected to long waiting hours to participate the study and feels fatigue and physical drain. This can lead to too many wrong answers which will reduce internal validity

2. Instrument malfunction

Ageing of laboratory equipment in experimental research, faulty data collection tool, participants bad preparation all lead to reduction in internal validity and introduce too many artifacts

3. Same pretest and post-test in multiple group comparison studies

This is another source of reduced research internal validity especially if participant lean the mechanisms of the pretest so that they manipulate the post-test results. A good research design should make sure the participants are unaware of the research process

4. Experimenter bias

A research internal validity might be reduced and results altered by artifacts if for example, the researcher is distracted by the influence of the participant and offers sympathy. Another source of experimental bias and hence reduced internal validity is when the researcher uses inappropriate statistical analysis procedure or inaccurate data collection plan

External validity

External validity refers to when the results of the sample data analysis can be applied and generalized at the population level. Sampling bias (lack of representative sample) or large sampling standard error reduce research external validity.

If the research study course is long-term, then it will have large external validity due to large data observation than short-term reseach

To recap, use probability sampling method if you want to increase research external validity. Non-probability sampling methods introduce sampling bias and hence reduce external validity

I hope you learnt something from this post. Any question or improvement please write in the comment below and subscribe so that you dont miss anyone of future articles and papers