The Difficulties of Creating a Test
In the following post I would like to address certain issues that I faced so far while creating a short emotional intelligence assessment for this class. The most salient challenges include: having to create a test that can be completed in less than 2 minutes, choosing which questions will best assess the construct, and which response format to use.
Firstly, creating a test to assess any construct in less than 2 minutes is an extremely difficult task. Given how long most assessment tests are, I always believed it was absurd that researchers had their participants complete assessments that took over 20 minutes to complete. I was under the impression that researchers were accumulating an excess amount of data than was actually required for the purpose of their study. However, now having been given the opportunity to try and assess emotional intelligence in approximately 15 questions I have realized that I was the one being unrealistic. In order to properly assess a construct, including one as intricate as emotional intelligence, one most definitely needs many questions that will most definitely take more than 20 minutes to complete.
Secondly, it was difficult to choose which questions would best assess the construct I had chosen. Given the limited amount of questions I could use, I was struggling to decide how many questions to dedicate to each of the facets found in emotional intelligence. I also could not choose if I should use multiple questions asking similar things in order to properly asses a single component of emotional intelligence, or ask many shallow questions on all components. In the end I used more general questions that I thought could best assess emotional intelligence more generally. However, had I been given the opportunity to create an emotional intelligence test that was slightly longer to complete, I would have asked more questions on similar components.
Thirdly, choosing weather multiple-choice questions, true or false, or a 5-point Likert scale was the best to assess the responses was also challenging. These three forms of assessment questions have all been used to assess emotional intelligence and can all do so in different ways. This therefore made it difficult to decide what I should choose for my test. I used a 5-point Likert scale because it is most commonly used in emotional intelligence testing, however I believe that I could also have obtained valuable information from those who completed my short assessment by using either multiple choice questions or true or false questions.
To conclude, I have learned a great deal from this experience. Having to create a test for this class has allowed me to understand why certain tests are so long. It also permitted me to gain appreciation for researchers who have the difficult task of creating and choosing items that will best assess their construct as well as choosing a form of assessment that will yield the best responses form participants. My next task of assessing the validity and reliability of my test will surely arise other difficulties.