Reliability is the measure that shows how often the research presented the same results after the same tests were done. The reliability of a test or research depends on its frequency and consistency. The results of the test or research are considered reliable in case if the same outcome was recorded after the majority of the tests. Reliability is crucial for all kinds of research. This is why Babbie (2014) states that measurement devices are necessary to achieve the most accurate results and the most precise records during the research.
Validity is the measure that shows the accuracy of subject the research is directed to study. Validity of the test is measured according to the degree to which the test sticks to its subject and all of its parts and aspects. Babbie (2014) warns about overgeneralization and selective observation – these two factors influence the research’s validity a lot. Visible patterns should never distract the researchers and make them less objective, as this will ruin the test’s validity.
In sociology perfect validity and reliability are extremely hard to achieve. Sociological researches study humans, human relationships, human interactions, human behaviors – these are very unstable objects to measure. There are multiple factors that could influence a person and make an impact on the kind of data this person will end up presenting to the researchers.
Both validity and reliability are extremely important for the quality of the results of the research. Only valid and reliable results can be applied. Only valid and reliable results will have authority and will be trusted by the public.
After certain research or a test has been performed, an additional test is needed to find out the effectiveness of the previous research and the quality of its results. Medina, Barquera and Janssen (2013) have presented the work based on the test-retest in measurements of the physical inactivity level in Mexico. It is well known that physical inactivity is harmful to health; it damages the muscles of the human body and increases the risk of heart diseases.
The participants of the test were two hundred and seventy-two Mexican adult men and women of different ages, employed at a factory and doing different types of work daily. The participants had to answer the questions about their daily activity, and then they were given accelerometers which the participants had to wear for 12 days, and after that, they were given another questionnaire to fill in. All the results were compared by the researchers and the reliability and validity of the test were calculated. After the calculations have been processed the conclusions were made. The test showed “modest reliability and poor validity” (Medina, Barquera, Janssen, 2013).
The reliability of the test was based on the regularity of the data calculations. The participants of the test were wearing accelerometers daily and registering the time when they had to take the devices off. Validity failed because the participants of the test were employed people whose level of activity was mostly moderate due to their work at the factory, but a big part of the Mexican population is unemployed and suffers from health problems. This aspect was not taken into consideration, and as we know, to achieve maximum validity, the researchers have to pay attention to all aspects of the subject they study.
In conclusion, we can see that validity turns out harder to achieve when the subject of the research is inhomogeneous and has too many aspects that cannot fit into one just research. At the same time, in most cases, reliability depends directly on the use of devices and machines that collect practical data. Machines are much more reliable than humans.
Reference List
Babbie, E. (2014). The Basics of Social Research. (6th ed.). Belmont, CA: Wadsworth.
Medina, C., Barquera, S., & Janssen, I. (2013). Validity and Reliability of the International Physical Activity Questionnaire among Adults in Mexico. Revista Panamericana De Salud Republica, 34 (1), 21-28.