Midterm 1 Flashcards
What is a test
A standardized process or device that yields information about a sample of behavior or cognitive processes in a quantified manner.
- We never experience any cognitive process on its own, we only assess external behaviors that indicate that process.
- This can be tricky because we are taking a set of traits from a human and putting them into a manner that can be quantified so we can use statistical analysis
- There is a large assumption that we take that we take a reductionistic stance on what it is to be human. We assume that we have enough parts and that if we put them together we know what it is to be human.
What is the difference between validity and reliability? Why are these important concepts in psychological measurement?
- Validity- How well out measurement taps into the construct (if we are measuring what we think we are measuring).
- Reliability- Consistency in measurement over time. (If your observation of the behavior will give you the same concept of the behavior at different points in time.)
- These are important because we need both for a strong measurement, we need to make sure we are actually measuring what we think we are measuring and that its consistent over time.
What are the four critical assumptions of testing and what do they mean?
- Deviation- People differ in important traits.
- We can QUANTIFY these traits: Need operational definition. We need to have some level to be able to judge a behavior-operational definition. We as a field have not measured happy, motivation, attitude, etc but we measure behaviors that underlie the psychological construct.
- The traits are reasonably STABLE We are variable and so are emotions. Naturalistic observation, we act differently in a lab.
- MEASURE of the traits relate to actual behavior: We do not see the constfuct, but we use correlated behaviors to quantify the traits Remember: Dairy Queen Strawberry Milkshake
What are societal concerns of psychological assessment?
- Extent to which tests invade privacy
- Fair use of a test
- Justice: Impact of testing on society- We have a responsibility to know that research is for something ethical,
- We have an ethical code- way to help you as a professional (ethics is a mindset)
- We have not grown into a view of a holistic person because we are stuck into a behavioristic standpoint.
- We typically look at the average and look at the individual scores
How do ethical standards and legal standards differ? Which are we held to as professional psychologists?
- Ethical standards are what one should or should not do according to the principles through norms of professional conduct
- The law is what one must or must not do according to legal dictates.
- Examples: CEO hiring to weed out people with mental illness, interrogation – legal but not ethical, reporting client abuse.
Idea of ethical themes and the need for competence
• We are held to both and have a strong need for a high standards of ethics and administering and using tests.
Our field is based on reputation.
We need to use ethics in choosing, administering, interpreting, and communicating test results.
We need to utilize tests responsibly, and we should develop competence in assessment concepts and methodology
Competence:
- An understanding of norms, reliability, validity, and test construction
- Knowledge of specific procedures applicable to a particular test (administration, scoring, etc.) •The psychologist is responsible for continually updating his or her knowledge and assessment skills •Recognizing the boundaries of competence
What are ethical responsibilities of the test developer?
- The test developer should define clearly what the test measures and who it applies to.
- Know the present characteristics and limitation of the test.
- Review test items for insensitive content and language
- Define clearly what the test measures and who it applies to.
- Accurately present characteristics and limitation of the
- Review test items for insensitive content and language.
Ethical responsibilities of test administration
- Select tests only after they have conducted a thorough review of all tests available
- Have and maintain a thorough knowledge of all test materials (including the manual)
- Avoid using the test for purposes that are not recommended by the test developers
- Provide the test-taker with information about their rights
- Inform the test-taker how long scores obtained will be kept on file and to whom they can (and will) be released
- Explain the results of the test in language the test-taker can understand
APA test taker rights
- Be treated with courtesy, respect, and impartiality, regardless of age, disability, ethnicity, gender, national origin, religion, sexual orientation or other personal characteristics
- Be tested with measures that meet professional standards and that are appropriate, given the manner in which the test results will be used
- Know, in advance of testing, when the test will be administered, if and when test results will be available, and if there is a fee for testing services ¨
Also, the test taker has the right to…
- Have the test administered and the test results interpreted by appropriately trained individuals who follow professional codes of ethics
- Receive a brief oral or written explanation (prior to testing) about:
- The purpose(s) for the testing
- The kind(s) of tests to be used
- If the results will be reported to them or to others
- The planned use(s) of the results
Where can you find information on ethical principles and professional guidelines, and information about specific tests?
- The ethical principles of psychologists and code of conduct
- The standards for Educational and Psychological Testing
- About tests: Mental measurement yearbook, Test in print, Test Critiques, Google Scholar
What is measurement?
- Quantification
- The process of assigning numbers to persons in such a way that some attributes of the persons being measured are faithfully reflected by some properties of the numbers.
(We want to maintain a faithful representation of what we are studying. )
What are the scales of measurement?
Nominal Scales – numbers take on the meaning of a verbal label, but don’t signify any particular amount of a trait •Least useful, pretty much just a label that it attached to an individual. Ex: football jerseys, we don’t analyze quantities
Ordinal Scales – numbers denote order or ranking, but not amount of a trait, and there is no consistent difference between numbers •Rank ordering of information. No distinction between each of the numbers (2 to 3 distance is not the same as 3 to 4) ex: runner 1:10 min, runner 2: 15 min, runner 3: 2 days
Interval Scales – numerical differences in scores represent equal differences in trait being measured •Ex: on a scale of 1-100, how happy are you. Identifiable and known difference between points on the scale. Likert scale- not a true interval scale. It is interval-ish, we do it because we don’t have the precision knowledge to make it more precise. This may have a 0 but it does not mean the absence of the trait, it is just on the scale (ex: temperature)
Ratio Scales – have a true zero point, with zero= total absence of the trait being measured AND can make proportional statements, with twice the score= twice the attribute •Theoretically this can have a 0 as long as it is a true 0 and when it goes in the negatives it is in line with the interval for the positives. (ex: money and debt) **The scale you use is important because it will help you draw different conclusions
Which of the scales of measurement meets the minimum criteria for statistical measurement?
•Interval is the basic assumption for all statistics to work. (Can’t have an absence of an emotion.)
What are the different measures of central tendency?
A Singular representation of a lot of data
Mode: most frequently occurring number
Median: the literal middle score, the number that separates the top half from the bottom half of a score distribution, the 50th percentile
Mean: the arithmetical average score, calculated by summing the scores, then dividing by the total number of scores. This is the most common!
\+: It takes into account every point of data
-: It is susceptible to outliers
What is the normal curve and why is it important?
- Theoretical distribution of human traits in nature
- Also called the normal distribution or bell curve
- Mean, median, and mode are the same value in a normal distribution
- Same proportion of scores can always be found within the same standard deviation limits We would have no statistical analysis if we did not have the concept of what is normal.
- It is a BIG assumption that data has this curve.
- Normal curve is the perfect standard to which we will compare other information and data we have. We would have no statistical analysis if we did not have the concept of what is normal.
- Science does not prove anything, we have hypothesis but we don’t ever have definitive proof of anything.
- The more non-normal your distribution is the less you should trust it
- This curve assumes that within the population there is a normal distribution.
What is variability and how is it calculated?
Variability- reflects the extent to which individuals differ and how far our points are from the mean.
Variance is the core of statistics. We need variance, it reflects the extent to which individuals differ. Ex: taco bell.
Calculate: sum of squared deviations / number in sample
What is the standard deviation and how is it calculated?
Standard deviation: average of how far things are away from the mean.
Calculate: Square root of the variance
What benefits are derived from use of a z-score? How do you calculate it?
- We use a z-score to normalize a weird distribution. Allows us to compare across different scales and put them on the same curve. Apples and oranges.
- Provides meaning into insight on test scores of being high, medium, or low. Interpreted in standard deviation units. z = divide an individual’s score by the standard deviation.
What is correlation?
- Reflects the degree to which a score on one measure or variable is associated with a score on another measure or variable.
- Range from +1 to -1, with numbers closer to +/-1 indicating a strong relationship and numbers closer to zero indicating a weak relationship
- Relationship can be positive, indicating scores vary in the same direction, or negative, indicating scores vary in opposite directions
Why do we say that we cannot assume causation from correlational data?
We cannot assume causation because:
- We do not know if there is another variable driving the correlation
- We do not know which caused which because we are not looking at a pre-existing state.
- We can make prediction in regression, but cannot make causal claims.
When can we assume causation?
• We can assume causation in a methodological situation. Need experimental manipulation to infer causation
When can we predict using correlation?
If you have a strong correlation where in a positive correlation as one increases we see a corresponding increase in the other variable, and the negative is the inverse of this.
If you have a moderately strong correlation in either direction, simply knowing the value of one variable will give you an idea of what the other variable is.
