Class Notes (835,428)
Canada (509,186)
Psychology (7,782)
PSYB32H3 (614)
Lecture

PSYB32 - Lecture 04 Notes.docx

15 Pages
91 Views
Unlock Document

Department
Psychology
Course
PSYB32H3
Professor
Konstantine Zakzanis
Semester
Summer

Description
PSYB32 – Lecture 04 WebOption – Summer 2013  This week we’re talking about Assessment Procedures Slide 2 – Reliability and Validity in Assessment  **Two terms to know for exam: Reliability and Validity o There are different kinds of each ***Reliability: the consistency of measurement o When you give a measure is it consistently measuring the same thing o Different kinds of reliability (all of which are impt to know, know their definitions in textbook well)  Test-retest reliability: if you administer a psychological test to a person once, and then you give it to them again later on, will you get the same result? o If you do get the same result, you have a reliable test (or high test-retest reliability) o *Doesn’t apply to certain tests where different results are guaranteed the next time the patient takes them  ie mood tests where the patient took a depression test while depressed. If they get treated for it, obviously the results won’t be the same the next time they take a depression test  *test-retest reliability wouldn’t be important in this context  Alternate-form reliability: When you administer a test, you want to provide a different form of that test in the future. Why is it important to do this? o So patient doesn’t memorize the answers the first time (no practice effects) o Test has high alternate-form reliability if patient gets the same score (ie depression score) on multiple forms (different versions) of the test  Internal Consistency reliability: All the items on a test measure the same thing o Ex: if you administer a depression test, all of the questions should relate to and measure depression o If a depression test asks “what’s your shoe size” or “what’s your favourite class” then these items are inconsistent with the other items (questions) that actually measure depression ***Validity:*if the test isn’t reliable (low reliability) then there’s no chance that it has validity  Content validity: when measuring something, it’s only valid if we measure as much of the construct (ie disorder) as possible o Ex: if he gave us a depression test and asked us “how’s our sex drive?” and “how’s our appetite?” Is this asking enough in terms of what constitutes depression? No. So the test has Low content validity then. o Depression has many symptoms and this test covers way too few of them o So for a test to have high content validity, the test should measure as many possible symptoms that are associated with the disorder  Criterion validity: does your test measure what you say it measures? o Ex: if prof develops the Zakzanis depression inventory, one way to illustrate criterion validity is to give a group of patients the Zakzanis depression inventory AND another depression test o If his test has high criterion validity, patients would score high on both tests o That would mean his measure (test) is showing that these ppl are depressed because their ratings on his test were similar to their ratings on another test for depression o Another word for criterion validity is convergent validity (their responses/scores on his test converge with another test that also measures depression) o Another aspect of criterion validity is predictive validity: if the criterion is that his measure measures depression, he wants to know that his test is predictive of that  ie if the person fills out the test, does their behaviour (ie how they look) match their answers on the test  Do they look sad or dishevelled in the real world the way they answered on the test?  So the test needs to be predictive of whatever construct it’s measuring  Construct validity: hard to define so he just explains it o Disorders are simply constructs. They aren’t tangible things. For ex, anxiety isn’t something we can touch, it’s something we’ve defined (heightened worry, stress, fear) o So how do we really know if patients with anxiety actually have anxiety? o To illustrate construct validity, take a bunch of patients who have anxiety and compare their scores against people who don’t have anxiety o If you can demonstrate that there’s a difference b/t the groups, then we know that this measure is measuring this construct (anxiety) Slide 3 – Psychological Assessment  An essential component of any psychological assessment is the clinical interview  We’ve all been through clinical interviews (ie when we visit the Drs and he asks us questions)  They ask these questions to figure out what’s wrong with you  The types of questions the interviewer will ask depends on the Paradigm they’re working from o Ex: if it’s a neuropsychologist, the types of questions they ask will relate to the biological paradigm o “Are you using any medications?” “Did you hit your head?” o Ex: if it’s a psychoanalyst, the types of questions they’d ask: o “did you have a crush on your mom in childhood?” They’d ask about your dreams  Psychologists often use 2 types of interviews: o Structured interview: the interviewer asks you question after question in a specific order  *Prof says to keep an eye out for the SKID (a type of structured clinical interview) in the textbook o Semi-structured interview: The questions aren’t as structured. The psychologist will more so just ask questions if something they find interesting about a patient’s answer comes up  Importance aspects of clinical interviews aren’t just what the patient tells you based on your questions, but how the patient behaves. So psychologist may also conduct: o Behavioural Observations: Aside from the answers they give you, does their behaviour indicate what type of disorder they may have?  Prof gives ex of the first assessment he ever conducted where he tested the patient for dementia. He brought the man in the room, he sat down, and the man got back up and started to masturbate right in front of the prof  The second patient he saw, he was testing them for memory loss. Patient asked if he could go to the washroom, and he never came back. Prof found the person outside and he said he forgot where he was.  So sometimes these behavioural observations can be as diagnostic as questions  In his first ex, the patients behaviour is hallmark of frontal lobe dementia  In the second ex, prof doesn’t even need a test to prove that the patient has severe amnestic disorder  Limitations of Clinical Interviews: o 1. Are they telling the truth? In other words, is there any evidence of negative impression management?  If all you’re doing is asking questions, you don’t know if they’re faking, lying o 2. Sometimes patients can have no insight, and interviews may not account for this  ie the first patient who Prof saw didn’t think he was doing anything wrong when he started masturbating. He had no insight  So in an interview, he’s going to tell you he has no problems because that’s how he actually feels, but obviously he does have problems based on his behaviour  So if you stop there and write a report saying patient’s got no problems because he said so, this is obviously a huge problem o 3. Sometimes patient may engage in positive impression management  Whereas a patient with no insight has no conscious awareness of their symptoms (that something is wrong with them), someone engaging in positive impression management is consciously denying their symptoms  ie in a job interview when we half tell the truth. We’re intentionally leaving things out  Prof doesn’t clarify the difference b/t negative and positive impression management (so look in book) Slide 4 – Psychological Tests  What do we do beyond Clinical Interviews? Engage patients in Psychological tests  Diff kinds: listed on slide Slide 5 – Screening Measures  These are basically checklists of symptoms that a patient may or may not have  Why may Screening Measures be unhelpful? o (Slide 6) Prof gives ex of the BAI (Beck Anxiety Inventory), a measure of anxiety that psychologists would employ o On the left, you read the symptom. On the right, you rate the severity of that symptom (not at all, mild, moderate, severe for ex) o Prof took the test himself in front of class and started answering them based on how he was feeling in his current situation (in front of 700 students, hungry) o His answers would suggest he’s suffering from an anxiety disorder. But he really isn’t. Shows that someone’s 1. responses may depend on the situation they’re in while answering the question (or how they’re feeling at the time of the questionnaire) o 2. Responses are highly subjective.  The patient, like the prof, may see one of his symptoms as severe when maybe the questionnaire wants severe to mean something else o So Screening Measures aren’t much of a step up from Clinical Interviews. Just as limited. o Some psychologists won’t engage a patient beyond screening measures which prof believes is malpractice. Instead, what they should be doing is employing… (next slide) Slide 7 – General Personality Inventories  …Omnibus Measures: measures that not only have good construct validity, but also contain different measures of validity, ie different scales that determine many things like if the person is exaggerating, whether they have no insight, etc o The book talks about the MMPI which is a good example of an Omnibus Measure, but prof is gonna talk about the PAI (Personality Assessment Inventory) Slide 9 – Personality Assessment Inventory  A 344 item questionnaire which measures many kinds of different psychopathology (ie depression, anxiety, ocd), and it has some validity scales in it Slide 10 – Interpreting Test Results  Before these tests are published, they’re administered to lots of different people, people with and without disorders. This allows for Normative comparison  That way, when you get a patient’s test results, you have something to compare them against  So if you ask a patient to complete the Beck Depression Scale, you know already that healthy people score around 100 on it.  If your patient scores 70 or below, then you can confidently suspect that their behaviour demonstrates ***Statistical Infrequency, or in other words, there may be abnormality present because their score is statistically rare Slide 11+12 – Test Scores  One way to express statistical infrequency is by using T-scores  (slide 12) shows a PAI output (someone’s results) o After the person fills out their 344 questions, the psychologist plots their scores against standardized scores to see how they do against other patients and normal people  Anything that is above 70 (you can see where 70 is on the graph) means that it is statistically infrequent o Healthy people don’t get scores above 70 on this test  Along the bottom (x-axis) are different measures of psychopathology  ie on the first column, you see ICN which is inconsistency  Did the person just sit there and check things off at random? Or we they actually answering questions in a consistent manner so we know they were paying attention? o Ex of questions on the ICN scale would be similarly worded questions/statements but asked in the opposite direction: ie “I feel depressed” and then later on you’ll find the question/statement: “I am not depressed”  If they answered yes for the first one and yes for the second one, their answers are inconsistent o On a Screening Measure you don’t have this kind of consistency scale, so you wouldn’t even know if they were paying attention to what they were checking off  The second column (INF) measures Infrequency – items that most patients and healthy people would never endorse. o Ex of something people would never endorse: “My favourite sport is high jump.” It’s almost certain no one will answer yes to this so it’s infrequent o So if you see an elevation on this scale you know that the person may not have been paying attention to the questions  He very briefly runs through a few more. The point is how in depth these tests are, how much they cover  According to this patient’s scores (it’s the prof’s scores), nothing is wrong with him. None of his scores are statistically infrequent (above 70) Slide 13 – Next graph  The results on this graph are of a patient the prof saw and as you can see, many of his scores break the statistically infrequent mark of 70  In particular, his Negative Impression Management is through the roof o Thus it’s safe to assume that this person may be faking having a psychological disorder Slide 14  Based on the graph it looks like they have depression, because there is that elevation on the DEP scale. It’s pretty clear cut depression too because it’s the only scale that goes above 70 o So this is what someone with depression’s scores would look like on one of these tests Slide 15  Cases like these are not so clear cut. The patient suffers from Schizophrenia (as it’s the highest scoring scale), but their negative impression management and suicide scores are pretty high too  The point is that it’s not like the previous slide where it was easy to determine that the patient has depression. o Shows that diagnosing using this method can be difficult at times if patient scores high on multiple, unrelated scales  Today’s psychologists really rely on these Omnibus measures Slide 16 – Projective Techniques  These are tests that were much more common in the 50s – 70s. Not so much anymore, especially for diagnostic purposes  Some clinicians may use them as part of therapy however when a diagnosis is already known  Knowing the Projective Hypothesis (definition on slide), you can determine that these tests/techniques are based on the Psychanalytic paradigm  Examples of these tests on following slide(s) Slide 17 – Rorschach Inkblot Test  An example of the test on slide 18  Prof asks what do we see on the slide? Everyone had different answers  Shows that this technique has issues of Reliability – won’t get the same answers each time you employ the test  The way psychologists score patient responses to this test is by commonality of response, but if everyone in the class had different answers to what they saw then it shows how unreliable these tests can be o These tests are hard to score because their Test-retest reliability sucks  However, what psychologists generally do with these tests is use to patient’s response (as to what they see) and dig deeper to touch on some inner conflict that it may represent Slide 19 – Thematic Apperception Test  Ask the patient to look at a set of pictures (ex Slide 20) and tell the psychologist what you think is going on in each of them Slide 21 – Specific Psychological Inventories  There are thousands of other more specific measures assessing each different type of psychopathology.  So these are more specialized types of tests focusing in on a specific disorder, for ex:  Trauma Symptom Inventory: A measure that is used on a person who may have PTSD  Pain Patient Profile: Measures a person’s perception of pain (are they able to sleep because of their pain, are they disabled by their pain) Slide 22 – Intelligence tests Intelligence tests: different types  Most commonly used is the Wechsler Adult Intelligence Scale-III  Impt points to keep in mind about intelligence tests: Why are they useful? (4 points):  1. When combined with achievement type tests, it can help diagnose learning disabilities o Then you are able to direct an academic plan for the patient  2. Helps determine if someone is mentally retarded  3. Helps identify gifted children o Then you can help them by directing a school plan for them  4. Employing an intelligence test as a neuropsychological evaluation o Intelligence tests test cognition (memory, concentration, problem solving), all the things neuropsychologists are interested in studying o So sometimes intelligence tests are part of a neuropsychological assessment because they can tell us about someone’s cognition, and thus, their brain functioning Other notes about Intelligence tests:  They’re highly reliable o If you take the same intelligence test 10 years from now, your results will likely be the same  For those worried about getting poor results on an intelligence test, the construct validity is questionable o These intelligence tests may not capture the entire construct of what intelligence is o The tests may claim to define intelligence, but depending on the questions asked, the test may only define intelligence to a certain extent, or in a limited way Slide 24 – Wechsler Intelligence Scale Subtests  What an abbreviated version of a Wechsler Intelligence Scale looks like  The particular test on this slide is measuring the person’s visual, perceptual and constructional functions  The test is separated into verbal and non-verbal (performance) subtests o If there are discrepancies in one or both it may be indicative of a learning disorder or brain damage o ie if you do really well on one but terrible on the other then it’s a possible sign of brain damage on the part of the brain related to the section you did poor on  Combining their verbal and performance IQs gives us their Full Scale IQ  Summarizing their scores into a single Full Scale IQ however causes us to miss out on important information o ie there could be subtests that could have gone horrible badly but as long as you did good on the other subtests and end up with an average score, it covers up the one really bad test you had Slide 25 – How to Conduct a Psychological Assessment  There are t
More Less

Related notes for PSYB32H3

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit