Psych 325 1st EditionExam # 3 Study GuideChapter 5 Multiple Choice Question Topics- Internal validityo Have you identified the true cause and effect- External validityo Does your study apply to other people and settings?o Threats to validity: Tracking people over time When not using an experiment Almost any time Tattoo example- Individual differences threat to validity. Why it’s a problem. What’s the solution?o People vary in many wayso These differences may threaten your research- Pseudo-experiment. o One group gets treatmento Do they behave as expected?o Expose people to treatment, assess reaction, no control group i.e. new depression therapy- measure depression level on a 0 (lowest) to 10 (highest) scaleo average score =2 i.e. test a new pain medication- measure pain ratings on a 0-10 scaleo average score=2 i.e. try a new “pick-up” line at a bar- measure number of phone numbers revealedo number=20- Selection bias. Why it’s a problem. What’s the solution? o Selecting population to study o You may favor one type of person over anothero Sampling people from an unrepresentative sample (by using imperfect sampling techniques)- Non-response bias. Why it’s a problem. What’s the solution?o The respondents themselves are the source of the biaso Because people who choose to answer surveys are systematically different from people who choose not to do so, surveys that have low response rates may yield information that is highly misleading.o Solution: weigh individual respondents based on known population values.- History and maturation threats to validity. Why they’re a problem. What’s the solution? o History:o Example: you hypothesize that job loss increases depression levels You measure depression before and after assessment of job loss in NYC residents Also, 9/11 occurred at the same time as your assessment Can you be confident that it was really job loss that caused changes in depression scores?o Maturation: Example from p. 112 in text: mental imagery training with adolescent ballet dancers- Measure dance ability before and after 24-week mental imagery trainingprogram- The group improves in dance ability. Was it due to the training?o Maybe not: dance ability is known to improve with physical development- Regression toward the mean. How measurement error causes it. Why it’s a problem. What’s the solution? o Extreme scores are likely to be followed by less extreme scoreso Why? Measurement error All scores reflect true score and measurement error Measurement error is normally distributed around the true score Extreme scores reflect chance and are unlikely to repeat True score may stay the sameo What is it? RTM is a statistical trend that is a reality of life in statistics. It basically happens on the posttest where the measures (for example test scores) on the average move back toward the mean on average. The net effect of RTM is that the lowest scores on the pretest tend to be higher on the posttest, and the higher scores on the pretest tend to be lower on the posttest.  It is important to note that regression is always to the population mean of a group But really, there is no change that takes place from the pretest to the posttest due to the treatment or the dependent variable. RTM is like a number of everyday terms for example: “law of averages”, “things will even out” or “we aredue for a good day after a string of bad ones”  Essentially what all these slogans are statin g is that “extreme occurrences tend to be balanced by less extreme occurrenceso When does it occur? When the sample is selected on the basis of extreme pretest scores.- Examples: workers with the lowest drive, students with the lowest math skill, patients with the worst symptoms- RTM is anticipated on these cases where there is non-random sampling or assignment in experimentso How does it happen? When one takes a measurement in a study, the measurement consists of two components; the true score/measure and the error- Hawthorne effect.o When you’re in an environment where a supervisor or leader is watching your performance automatically improves- Testing effects. Why it’s a problem. What’s the solution? o On most tests, you’ll improve if you take it a second timeo Why? Practice or learning- Types of Mortality / attrition. When is this a problem for internal validity? What’s the solution?o The failure of some of the participants in an experiment to complete the study: experimental mortality or attritiono Example: Study over 24 weeks, recruit 100 people for fitness assessment You think your exercise intervention is going to improve people’s fitness who are“unfit” Pretest, treatment, posttest Measure fitness on 0-10 scale (100 people) exercise intervention  measure fitness on sample scale (50 people) This is homogeneous attrition because in the control group the same amount of people drop out Heterogeneous attrition- After posttest: n=15 but in control group after posttest n=90- You want to do everything you can to reduce attrition and mortality - Participant expectancies. Demand characteristics. What’s the solution? o Occur when participants consciously or unconsciously try to behave in ways they believeto be consistent with the experimenter’s hypothesiso Reasons: in an effort to please the experimenter, feel normal, part of a desire to fill an implicit social contract with the experimentero Characteristics an experiment itself that subtly suggest how people are expected to behave: Demand characteristics: play a role in the operation of participant expectancies.- Participant reactance. What’s the solution?o Tendency of participants to try to disconfirm an experimenter’s hypothesis Main reason participants might work hard to disconfirm an experimental hypothesis s is probably grounded in people’s basic desire for autonomy or independence- Evaluation apprehension. What’s the solution?o Refers to people’s concerns about being judged favorably or unfavorably by another person- Experimenter bias. Why it’s a problem? What’s the solution? (including double-blind procedure)o An unintentional bias on the part of experimenterso Takes two distinct forms:  When experimenters make biased observations in an experimenters When experimenters actually treat their participants different according to their expectations about how their participants should