First, let us look at the empirical aspects of the science. What do clinical researchers do? What methods do they use and how do they control for confounds? Secondly, let us take a look at the application of the science to real world practice. What types of reliability and validity should clinical assessment tools display and what are the pro's and con's of structured interviews verses unstructured interviews and of projective tests verses standardized tests? What is the purpose of clinical diagnosis?

The flu, a broken leg, cancer, depression, anxiety, schizophrenia....They've all got something in common. Sometimes, we forget that just like a broken leg, having abnormal bouts of depression, anxiety, etcetera are medical issues that need to be addressed. As with more observable medical research, psychological abnormalities also requires clinical research studies to determine different way to combat the issues or determine the most likely causes for certain issues. In these studies, researchers are able to establish different concepts that allow them to test the validity and reliability of their work for future use on an array of people. During clinical interviews, these tested methods are put to use to establish clinical diagnoses to help providers care for their patients and clients in a useful manner.
Clinical researchers conduct studies and trials for new medications, theories, and applications. The scientific method plays a large role in this particular career field (of every variation). Researchers regularly go through a similar process of the "scientific method" which begins by asking a question related to the topic they wish to run a trial on. Once a question has been asked the researchers will then accomplish a literature review of sorts to increase their knowledge of the former results related to the subject. Utilizing this research, an education hypothesis should be made. A clear set of dependent and independent variables should be established along with a process to test the concept/item they wish to test. Often times, this testing will need to be approved by a particular board associated to the field of study. Of course, once the process has been approved, the researchers will move on with testing. Using statistics, observations, and other types of information the researchers will then analyse the data and draw a conclusion based on the study they designed (Comer, 2014)

In real world applications, tests and treatments must be used in a sense that allows them to be reliable and valid in the circumstance. Reliability is graded by the ability to test and retest a certain item and score relatively similar results each time. The two tests should generally be taken within a short time frame (within a couple of weeks or months of each other, rather than years). Validity, on the other hand, is a bit more complex. Validity can take a couple of forms: face validity, concurrent validity, and predictive validity. Face validity is a sort of logic test; does the item we are using logically appear to fit the scenario it's being associated to? Predictive validity helps to predict future behaviours in relation to the situation. Finally, concurrent validity compares two (or more) items to see if those items with similar issues result in a similar solution using with the treatment chosen. (Comer, 2014)

When determining the use of certain applications in a real world scenario, it's important to decide whether or not to begin the program with a structured or unstructured interview of the patient/ client. Unfortunately, with interviews as a whole, the validity and reliability can be called into question based on the potential biases of the interviewer or the misdirection of the interviewee. Unstructured interviews allow the patient/client to discuss things they feel most comfortable with and what they think may be causing their issue. The problem with a lack of structure is a potential misdirection or avoidance of uncomfortable topics, making many question the reliability of the interview yet again. Structured interviews are slightly more reliable in a sense that the questions asked could be re-asked with the likeliness of gather the same answer. Structured interviews may also bring issues to light that were not considered by the interviewee for a variety of reasons. Again, as with the unstructured interview, the structured discussion could be led by a biased interviewer resulting in skewed results or mislead by an uncomfortable interviewee. (Comer, 2014)

The utilization of all the above (clinical research, reliability/validity, and interviews) are to determine a clinical diagnosis for issues perceived in an individual. A clear and reliable diagnosis is the first step to recovery of the mind or body. Understanding the underlying issues, whether biological, psychological, emotional, or whatever is what allows medical professionals to pursue a road to recovery that's been proven to work in similar situations through trials and testing.