It might be reliable but is it valid?

When we want to research something we often test our ideas or hypothesis using a range of experimental tests and designs. One of the most important things a researcher should ensure is that their test actually measures what it says it is going to measure, this means making the test valid.

There are two main types of validity, construct and criterion. Construct validity refers to theoretical concepts or ideas such as intelligence or stress. If we wanted to measure a person’s intelligence and ensure our test has construct validity, the scale or measurement being used to measure intelligence would have to be a true measurement of this concept. Before even testing any participants on their intelligence the researcher would need a clear understanding of intelligence and know how they are going to measure it. One problem the researcher could face is defining intelligence because intelligence can be displayed through many ways such as art, maths, logic or language.

Criterion validity refers to a correlation between the test and the criteria. The criteria will be representative for the construct. For example, we would expect to see a correlation between IQ tests and academic performance. Predictive validity branches from criterion validity. An example of this would be a person taking an IQ test then predicting the academic performance of the person in the future. If test and criteria data are collected at the same time this is concurrent validity, so if IQ test data and academic performance data are collected at the same time this is referred to as concurrent validity evidence.

If research tests or experiments are not valid, data can be seen as useless because it has not measured what it is supposed to measure. Results will show nothing as there has not been an accurate measure. It is important to remember that reliability alone does not make a successful experiment. A researcher could observe aggressive behaviour over and over again amongst children across different schools, however, if what they interpreted as aggression was really just harmless play results will not be valid. Children often like playing games involving physical contact and these games could be analysed incorrectly by the observer (Shaffer & Kipp, 2010).

Shaffer, D. R., &, Kipp, K. (2010). Developmental Psychology: Childhood & Adolescence, Psychology, Cengage Learning, 10-11.


One thought on “It might be reliable but is it valid?

  1. Both reliability and validity need to be high in order for the results of a study to be respected by others. Liek you have said there are two main tupes of validity which are important. However, there are also other types of validity such as ecological validity which are often considered to be equally important. If your results can’t be applied to the real world, it also poses a big problem.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s