Main Types of Validity
When you conduct qualitative research, reliability and validity are the main criteria to assess the accuracy of the chosen methods and measurements. Therefore, validity can inform how appropriately the method picked by you measures the subject of your research. If the received results correspond to real-life values, they are valid.
The article describes the four types of validity in qualitative research. They are the following:
- Construct validity - shows whether the test measures the subject that needs to be measured.
- Content validity - helps to understand whether the test measures exactly what it is aimed at.
- Face validity - demonstrates whether the test content is applicable to its purposes.
- Criterion validity - shows whether the results can be considered accurate in relation to the concrete outcomes of the test.
These types of validity measure the accuracy of measurements only. However, if you are doing experimental research, you will also need to think about internal and external validity. These types are related to the generalizability of results and the structure of the experimental design.
What Is Construct Validity About?
Construct validity shows whether a measurement tool you have chosen for your research works properly for measuring only those subjects it is meant to measure. This type of validity is the most important for assessing the full validity of the research method.
The Conception of a Construct
A construct is a characteristic of the subject that cannot be observed but is available for measuring via other indicators related to it. Constructs usually represent individual features, for example, depression, satisfaction, or levels of intelligence. They can also refer to group characteristics, such as gender representations, social responsibilities, or freedoms.
So, to obtain the construct validity, you need to check whether all the measurements are accurate and correspond to the existing theoretical backgrounds. Your questionnaire should involve only the questions that relate to satisfaction and not to anything else.
You cannot observe and measure satisfaction with any physical instruments. However, some characteristics and psychological criteria can provide such measurements based on positive attitudes, predictions, and levels of happiness.
What Stands Behind Content Validity?
This type of validity evaluates the correspondence of a test to all aspects of a construct. The content of any test, survey, or measurement tool should involve all the parts of the subject you want to measure. If some essential aspects are not included in the measurement or other aspects that do not relate to the construct are included instead, the resulting validity is low.
An Earth science professor at the college wants to develop an end-of-term test for the course. The test is meant to cover all the questions the students were learning during the term. If some parts of the in-class material are missing, the results will not show how well the students understand the subject. On the other hand, if the questions included in the test do not relate to Earth science, the results cannot be valid for the student's knowledge measurement in this field.
What Does Face Validity Imply?
This validity is meant to show how well the test content corresponds to the construct by its external features. It means how well it is compliant with it on the surface due to more informal evaluation. Such assessment is usually subjective, but the results are pretty apparent.
Since face validity is pretty subjective, it is the weakest type. Still, it can be helpful when you are just starting to formulate your research method and develop the measurements for it.
Suppose you are developing a survey to measure teenagers’ sleeping routines. So, you look through the survey questions to see whether they consider the number of hours teenagers sleep every day, the bedtime hours, what they do before going to bed, and the time for getting up. If you look at the external indicators, they seem quite valid on the surface because they consider the subject you want to test in full. It means that your survey has high face validity.
What Does Criterion Validity Mean?
This type of validity is about how consistently a test can predict the outcomes based on the approximately relevant results of another test that has already been conducted.
This variable is widely considered valid and can be referred to as a “gold standard.” It is difficult to find exact criterion variables, but they are immensely effective and always well-established.
You can assess criterion validity when calculating the results of measurement and compare them to the outcomes of the criterion measurement. If the correlation between these results is high, so your test is valid and measures exactly what you have meant it for.
You are making a new test for assessing students’ knowledge of English grammar. You have found the existing test that is widely applied and considered valid. Then, you give your students the test you have created and the test you have found. If the results your students show are almost the same, the criterion validity of your test is high.
Final Thoughts
Now, you know what the four types of validity are and how to distinguish between them. This information will help you to find the best measurements for your goals and research purposes and develop tests that are highly valid. That will improve the overall validity of your research results.