Choosing a Topic and Searching the Literature

• Read broadly before choosing a topic. This will allow you to establish some background knowledge on the topics, before focussing on the area in which you wish to design and evaluate a new test/ test battery.

• You are strongly encouraged to go beyond the provided references to do well in this assignment. These references are given as a broad and general background to provide you with a starting point to help generate ideas for your assignment.
o Try searching for keywords in databases such as Google Scholar and Web of Science.
o Another route to finding more articles is to look up references cited in the reference sections of the recommended readings and use databases to see what they cite and who has cited them.

• Find an empirically-supported relationship between the skills or abilities that you want your test to measure and what this measure actually indicates in a real world domain (i.e., upon what aspect of performance this skill or ability impacts).
o Do a little background reading and general searches for all skills or abilities that might impact on performance in the domain you are investigating. From this, you can get ideas for specific abilities for which you would like to design a test or test battery.
o One skill or ability is not expected to predict overall performance in a domain. Therefore, you should specify your argument (e.g., it would be incorrect to say that hazard perception skills alone indicate someone is a better driver; a better way to argue this would be to specify what aspect of performance this impacts e.g., safer drivers generally have good hazard perception abilities).
o You must find empirical research to support the design of your test or at least lead the reader to understand why you have employed a particular method (e.g., if you were proposing a hazard perception test, then find other research that has used hazard perception tests and any evidence that hazard perception tests are reliable and valid indicators of actual real world driving ability).

• Search for other tests that assess your selected skills/ abilities/ traits:
o Look at the current test(s) of the skill/ ability/ trait. What properties can you improve?
 Can you aggregate a few existing tests into one comprehensive battery, or just improve elements of existing tests?
• Within this, consider the way the test is currently administered (including format of presentation, response requirements etc.). Are there any issues that should be addressed that may interfere with this test accurately gauging what it claims to measure?
o Do these tests display adequate validity and reliability?

o Can you develop a way of testing this skill/ ability/ trait that will address any psychometric limitations/ shortcomings of current tests in this area?

• Focus the research question/ aim of your research proposal in order to ensure you know exactly what you are looking at and why.
o For example, ask yourself questions such as these:
 What domain am I investigating? (E.g., driving).
 What performance outcome measures am I interested in? (E.g., preventing car crashes).
 What skills/ abilities/ traits are related to these performance measures? (E.g., fast and accurate anticipation of potential traffic conflicts).
 What tests of these skills/ abilities/ traits already exist? How can I improve them or improve the evidence for their effectiveness with my knowledge of psychometric principles (i.e., reliability and validity)? (E.g., more evidence for the predictive criterion-related validity of hazard perception tests but specific to older drivers, current hazard perception tests do not include scenes of night-time driving hazards, or more alternate forms of hazard perception tests are needed).
 What are some risk factors that could potentially limit the effectiveness of my new measure? (E.g., if the hazard perception test has audio instructions, people with a hearing impairment will not be able to take the test unless they are provided with subtitles).