Running head: THE USE OF RELIABILITY AND VALIDITY IN RESEARCH
The Use of Reliability and Validity in Research
Dina Jones
The Use of Reliability and Validity in Research
The article “The Use of Reliability and Validity in Research”, really gave a better understanding of how to test reliability and validity. These were two areas that I felt would be a challenge for me to comprehend just as statistics was a bit confusing. Luckily, the article broke things down in laymen terms for readers and learners like me.
From the article, I was able to understand that the reliability of a test meant that a test could only be relied upon to produce similar results if administered to a person on a different occasion (Goshan, 2003). In other words, you need to get the same results from a test if you’re testing the same individual on the same test as if you were to test them a week from today. Through further reading I discovered that there are three techniques that can be used to determine the reliability of a test: re-testing, administering a second test, and split halves testing.
Re-testing using the same test is not always the most accurate option, because any person who takes the test again will become familiar with everything on the test which as a result will make the test easier to pass (Goshan, 2003). Personally, I feel that this concept is true, because some people may memorize their answers from the first test which could cause them to have either a higher score or a lower score. A higher score could be obtained if they had the correct answers rather then a low score following that of incorrect answers.
Another way of testing to establish reliability is through administering a second test that “contains items which are different to those of the first, but which measure the same qualities” (Goshan, 2003). This is called alternate form reliability. Although this technique would reduce the test taker from having familiar items on the...