Research & Measurement 2
Minnesota Multi-Phasic Personality Inventory (MMPI-2)
567 true/false statements; Research, diagnosis, clinical use; Clinical scales (different aspects of the personalities) & validity scales (looking at if the person is answering in an honest way); Most widely used personality inventory used in clinical; Describes someones personality
Reliability
A test must have this; does it get consistent results?
Validity
A test must have this; does it measure what it says it measures? can be examined in a number of different ways, how do you decide you are measuring what you want to measure
Biological data
Another type of data also taken, but less likely used in clinical
Define what we are studying
First step in research
Apply; To diagnose if there is a personality disorder, identify what inventions or treatments would be appropriate and effective. By standardize tools or just getting to know that person
Fourth step in research
Initial interview/assessment
Personality assessment in clinical practice; talk to people, ask them about themselves and gather certain types of information; you get a feel or sense on the persons personality; some clinicians will use assessment tools as well to help them understand the person sitting across from them.
Ongoing sessions
Personality assessment in clinical practice; you get to know their personality more and how they present themselves as they get more relax in sessions with you; you get a feel of who they are and how they behave; May include testing
Measure/assess; What tools are being used to measure or assess what is being defined. Very important for the researcher and the clinician to determine the tools that they are using are appropriate, so they can be confident that they are measuring what they are supposed to be measuring
Second step in research
Nomothetic
Source of data; (fixed or standardized) simple more objective structured
Idiographic
Source of data; (flexible) more unique tell me about yourself, fill in the blank, answer this question, or write me an essay on Describing the unique individual, what is distrinct
Construct
Test must have this to be valid; abstract concept, theoretical abstraction (intelligence); validity is essential to show you are measuring what you are supposed to be measuring; do items on the test correlate with eachother, compare one test with another test that is highly valid.
Discriminant
Test must have this to be valid; between one construct and another, what a measure should not correlate with another, especially what is opposite of it. A test measuring shyness, should not correlate with a test measuring something else.
Face
Test must have this to be valid; not very convincing, when a test looks like it measures what it measures; if there hasn't been any studies to check validity, you can't have much confident that it is measuring what it is suppose to be measuring
Predictive
Test must have this to be valid; predict what a person is like or willing to do or engage in some behavior. predicts what behavior should be expected. Demonstrating how a person responds to the measure predicts something external to it.
Causality
Test must have this to be valid; relatively new, measures a quality that already exists, like test anxiety. Variation between this, everyone should have some level of that which should influence what happens. Certain levels of a particular characteristics and influences this.
Inter-rater
Test should have this to be reliable; has to do with observers, technique to make sure that a number of people gathering data have similar ratings
Internal consistency
Test should have this to be reliable; used to see if different items on the test or measurements correlate with eachother, internal because it is looking at the test, if there is a high correlation you have high this and it there is low correlation, you have low this, which probably shouldn't be on the test.
Test-retest
Test should have this to be reliable; you give the test one time and then give the test another time and you should expect the results to be very similar.
Split-half (odd & even)
Test should have this to be reliable; you split a test in half, compare responses of odd number questions to even number questions; are the answers similar in the way the person responds to questions. You would expect there would be no difference between odd and even.
Alternate Forms
Test should have this to be reliable; you want more than one version of the test have very similar questions and items, in reading level and what they are addressing, looking for consistency between alternate forms.
Predict; Allows us to say something about what we are measuring
Third step in research
Social desirability
Type of Non-content responding; When the person answers in a social desirable way, they aren't answering what is actually true. Sometimes people don't want to admit to things and want to make a good impression, so they don't answer truthfully about themselves. Good impression good citizen- they want to answer in a way that the researcher wants them to answer, what the clinician wants to hear Sometimes conscious or sometimes unconscious
Extreme responding
Type of Non-content responding; an individual always responds in an extreme way (strong agree or strongly disagree); that's not valid data about the person
Acquiescence
Type of Non-content responding; the person who is being interviewed is saying yes to everything
Observer
Type of data; focuses on the behavior on the individual, provided by people who know the person well.
Life record
Type of data; most common data taken in personality psychology; school records, medical records
Test
Type of data; obtained from experiments, using computer increase the use of test data, ready source of information of data
Self-report
Type of data; you ask the person to talk about themselves, most commonly used source of data in personality psychology, done by questionnaires or interviews, there are limits because the person may or may not be aware of their characteristics, or they may want to present themselves in a different way.
Correlational
Type of research design; Comparing responses from two or more variables; benefit: can gather a lot of data from a lot of different people; can measure personality variables, different characteristics or traits, how they go together or how they don't go together; you can use very large samples, there are a lot of personality tools with strong validity and reliability; cannot talk about cause or effect; not able to address causality; because you are using primarily self-report data, may get social desirable responses
Case Study
Type of research design; come from clinical treatment, can also come from research; intensive investigation on a single individual is done, can be very valuable to gather information on particular personality characteristics or the particular person; you cannot generalize a case study to anyone else but that person; valuable for gathering information and making research questions
Experimental
Type of research design; only way we can investigate cause and effect; gather subjects, divide into control and experimental group; you need to collect a representable sample; compare results on dependent variable due to the independent variable; you can talk about cause and effect; limit is you are doing it in an artificial setting which can affect the data you get
Variability in behavior
Type of target; how does this persons behavior differ, are they always this way or are their certain situations where they are different
Average behavior
Type of target; understand what this person is like, what is their average behavior like, what is most typical about this person; thought to reveal their true inner personality
Conscious thought
Type of target; what you can bring to mind readily
Unconscious thought
Type of target; what's outside of ones awareness
Non-content responding
We want to reduce this because it doesn't give us reliable or valid information
Life record, observer, test, and self- report
What does L.O.T.S data stand for?
Targets
areas of interest of assessment
Forced Choice Format
combat non-content responding; the good and bad component is removed; Maybe what you like isn't really in there so it doesn't really describe you, which is a limit to this
Response sets 'Non-content' responding
getting response that are not about the context you are asking about; you are getting responses that isn't what you are asking about
Assessment tools
items, tests, measures, instruments...and observations, interviews...Giving someone something to gather data and information about them
Interviews
most common source of data in practice and theory will guide the choice of the data to be collected and what aspect(s) or personality are most important to focus on. testing is always imperfect, and have limitations