Originally published here.
What is validity? Firstly, there are two types of validity we must consider when investigating how valid the results of a study are. There is internal and external validity. Also, what is reliability? Are our results reliable and valid?
According to Graham R. Gibbs (University of Huddersfield), he defines that internal validity is when the evidence found reflects the reality under investigation. As for external validity, he defines it as when the results have relevance beyond the situation being investigated. This can also be know as generalisability. Furthermore he describes reliability, it’s defined as when an investigation is repeated by different researchers using the same methods and the same results are obtained.
We were told to investigate the best way to cook an egg, however seen as ‘best’ could be interpreted in various ways, as a group, we decided to investigate what would be the tastiest method. Therefore seen as we only investigated one line of inquiry around the word ‘best’ I think that the internal validity of our results is quite low, because we didn’t investigate the question as a whole so it doesn’t represent the reality under investigation.
As for external validity, I believe the generalisability is very limited for our results. The survey we conducted via twitter had 18 responses, therefore only a small sample, so we can’t generalise the results to the rest of the population.
Lastly I need to consider how reliable the actual findings of the investigation are. We did use triangulation, therefore our findings are based on more than one source. However they didn’t all correlate. Smith, Delia (2013) How to fry an egg and Kerrison, Melissa (2013) What’s the best way to cook an egg? [survey] did correlate, but Berry, Mary p84-86, had different results. Therefore our reliability is limited as not all of our sources agreed. In addition, our results have not be tested by any other groups, also reducing their reliability.
It is always very important to therefore plan an investigation first, and try to consider techniques to improve reliability and validity before you begin the research process.