How to Cross Validate Qualitative Research Results

Cross-validation is a popular validation strategy in qualitative research. It’s also known as triangulation. In triangulation, multiple data sources are analyzed to form a final understanding and interpretation of a study’s results. Through analysis of methods, sources and a variety of research theories, cross-validation can make a powerful contribution to support existing research, hypotheses and researcher hunches by presenting interpretations from multiple perspectives.

Validate the methods used to collect the data by examining the data-collection methods. In cross-validated studies, it is common to have data from both qualitative and quantitative studies. Question and verify what data collection methods were used. For example, were surveys taken or were live interviews conducted? Was information obtained over the phone or in person? Who asked the questions? Each method of collection must be confirmed and validated for any bias, unforeseen impediments or errors. Consider how the sources were obtained, such as surveys, for example — were they obtained legally and ethically from a population sample that represents the study at hand, or was the population of a protected class, such as children under the age of 18? Methods can only be validated if they were obtained legally and ethically.

Validate the consistency of the data sources by examining them at different points in time, along with the settings they evolved from, such as public versus private. Compare the differing viewpoints of people interviewed and survey participants. Validating the source assures that the research was taken from a reliable network of individuals and places.

Use different researchers to analyze the data to review the findings. This will also minimize any potential bias findings later on. Encourage multiple perceptions to develop an understanding of the data. The goal is not to have all of these analyses be in agreement, but rather to develop multiple ways of looking at the data, from multiple perspectives, by involving a variety of researchers in the analysis stages.

Develop a final theoretical understanding of the research, based upon all of the aforementioned cross validation strategies — the research methods, the data sources, using multiple researchers and the resultant theoretical underpinnings. Once all has been considered, multiple perspectives will inform stronger data sets because when taken together versus individually, results become more powerful.