Home / Data Collection / The Concept of Validity in Qualitative Research

The Concept of Validity in Qualitative Research

The concept of reliability, generalizability, and validity in qualitative research is often criticized by the proponents of quantitative research. Though it is difficult to maintain validity in qualitative research but there are some alternate ways in which the quality of the qualitative research can be enhanced. The aim of any research regardless of whether it is a quantitative research or a qualitative research should be to maintain rigor and control in the research in order to seek the truth. For a novice researcher, it is hard to seek rigor and control in the qualitative research. To bring validity to the qualitative research he should ask two questions from himself: first, who will decide that the instrument used to, and second, how to establish that the instrument is measuring what it is designed for?

The answer to the first question is the researcher himself and in case he is a novice in this field he will get help from the research supervisor or an expert in the field. The answer to the second question can be obtained by applying statistical procedures to the data measured. In qualitative studies, if statistical analysis cannot answer the second question than logic is used to maintain that the research tools are valid and reliable.

Definition of validity in research

According to Kerlinger the best definition of validity is in the question that “are we measuring what we think we are measuring” and according to Babbie (1990) “validity refers to the extent to which an empirical measure adequately reflects the real meaning of the concept under consideration”.

Types of validity in research

There are three types of validity in research:

Face validity

Face validity measures the validity of the instrument by creating a logical relationship between the questions on the scale and the objectives of the study. The questions on the scale should have a logical reason why they have been developed, i-e; they should provide answers to the objectives. It should also be noted that the questions on the scale should cover all the objectives of the study.

The problem with face validity is that the logic on the basis of which the researcher develops questions is subjective and hence it varies from person to person. If different questions will be developed the link between the objectives and the instrument will not remain same. No objective conclusion can be drawn thus as any change in questions will yield a different result.

Concurrent and predictive validity

Concurrent validity measures the extent to which the results of a scale are concurrent with the results of a similar study in the past. It can also refer to testing the same test on two groups concurrently to establish the validity of the test. It also predicts how valid the scale can be in the future, this is a very fast way to measure or establish the validity of a test. Its use is very common in the field of psychology and personal attributes can be very accurately tested with this validity.

Construct validity

Construct validity determines the validity in a more objective manner. It measures the instrument validity by developing a link between the questions and the outcome variables by testing the coefficient of variation. To establish construct validity the researcher should have good knowledge of the statistical procedures.

How to enhance validity in qualitative research

To enhance or maintain validity and trustworthiness in the qualitative research the researcher needs to adopt the following techniques:

  1. In qualitative research, the researcher cannot adopt an objective manner and hence he is unable to prove the validity by using statistical procedures. He can adopt some methodological techniques to develop and ensure validity in the research.
  2. He needs to make sure that he has avoided personal biases to a minimum to establish the validity of the research.
  3. He should also use the most appropriate sampling technique to avoid the sampling bias.
  4. A novice researcher can seek help from the research supervisor as well as other experts in the field to maintain validity in the research.
  5. The investigator should have to be responsive and adaptable to situations and should clarify and summarize things in the best possible manner to the audience.
  6. The investigator should maintain the credibility, confirmability, and trustworthiness of the study by adopting the right tools in his research.
  7. The reliability and validity of the research tool can also be enhanced by using different techniques to collect data are used to get in depth and reliable results.

Most importantly, many qualitative researchers oppose the idea that validity and reliability cannot be achieved in the qualitative research and hence rigor and control can be achieved by avoiding bias in the investigators’ actions throughout the course of research. These researchers suggest that the terminology validity and reliability should not be removed from the qualitative research. In one study conducted by Morse, et al (2002) the authors noted that nowadays the qualitative researchers focus on rigor and control through the verification of investigators actions in the qualitative research, they argued that the qualitative researchers should reclaim responsibility for reliability and validity by implementing verification strategies during the course of research itself. In their opinion, qualitative research should follow the scientific procedures that are inherent to their research design to bring valid, reliable and trustworthy outcomes.

References

  • Morse, M. J. et al, Verification Strategies for Establishing Reliability and Validity in Qualitative Research, Sage publications, 2002

Comments

comments

Check Also

https://www.flickr.com/photos/neccorp/16437481572

Ethics in Data Collection in Research

Ethics in data collection is not only an important but a sensitive issue, the respondent …

Leave a Reply

Your email address will not be published. Required fields are marked *

Please Answer *