exam 1 Flashcards
Dr. Sadeh is interested in studying the frequency of depression among pre-pubertal adolescents who also have anxiety disorders. Which of the following are variable(s) in this study?
Depression
- something that can be tested
Based on what we learned in class, which of the following is a good example of a scientific hypothesis?
Smaller class size is related to better academic performance; Decreased sugar consumption is associated with decreased physical activity. - something that can be tested
Explain why we do not say that a single study proves a theory or that a single study disproves a theory
A single study does not prove a theory because although a study today may find support for a theory, a study done tomorrow may not find support for that theory. In addition, a single study cannot disprove a theory because the single study may have been poorly conducted. Finally, a disconfirming study may mean that the theory may need to be amended or altered, rather than completely dismissed. Scientific research is iterative and probabilistic.
Your friend is trying to figure out whether drinking caffeine in the morning is good or bad for her health. She decides to drink two cups of coffee every day - one in the morning and one in the afternoon - to see how she feels after upping her coffee intake.
What method of acquiring knowledge is she using in this scenario? Experience
- full of confounds
Your friend decides that her coffee “treatments” are working, and that coffee makes her feel good throughout the day. What biases or errors might your friend be making? A combination of confirmatory hypothesis testing and present/present bias - she needs to compare to how she feels without the two coffees per day.
You develop a new measure of romantic love and find that your measure is strongly related to a second, previously-developed measure of love that you administered simultaneously. You have demonstrated:
Convergent validity
- the degree to which two measures of constructs that theoretically should be related, are in fact related
An experimenter conducted a study where they videotaped people on their first date. Three independent coders later watched these videos and recorded how often they saw each person flirt. Which type of measurement validity or reliability is particularly important in this study?
Inter-rater reliability
- consistency across raters
- cronbach’s alpha
- agreement among independent observers who rate, code, or assess the same phenomenon
How would you define “operational definition” for this high school student?
An operational definition is a specific statement about how a construct will actually be measured or manipulated - it’s taking a conceptual level definition and making it concrete
Conceptual definitions
- “Construct” or “conceptual variable”
- Theoretical, abstract level of description
Operational
- Concrete, specific level of definition
What is the critical ingredient of operational definitions that this student must understand?
Concrete quantification and/or specification
Generate both a good and a weak example of an operational definition of self-esteem and then explain to the high school student why they are good and bad examples.
Good: We will ask people to indicate how good they feel about themselves using a scale of 0 to 100. This is a good example because it is really concrete and quantifies self-esteem.
Weak: Extent to which people like themselves. This is a bad example because this is closer to a conceptual definition, rather than an operational definition.
Dr. Spielberg just generated a new operationalization of anxiety. However, he quickly realized that his operationalization (and the measure that resulted from it) was going to produce results that were reliable but not valid. Generate an example of an operational definition of anxiety that is reliable but not valid, and explain why it meets those criteria. Dr. Spielberg instead decides to operationalize anxiety by asking people to state how many episodes of acute anxiety attacks they have on average per month. Describe whether anxiety is 1) a self-report, observational, or physiological measure; 2) whether it’s categorical or quantitative, and 3) whether it’s an ordinal, interval, or ratio scale.
- anything that would generate consistent responses that doesn’t relate to anxiety would work. For example, if you asked people to self-report whether they are morning people or night people, this would likely generate reliable responses (e.g., consistent over time), but it would be an invalid measure because the morning person/night person distinction does not accurately capture anxiety.
- Self-report, quantitative, ratio
Your research question is “Does taking depression medication improve depressive symptoms?”. What sampling technique(s) will most likely allow you to investigate this question?
Purposive Sampling
- targeting specific types of participants for recruitment
Dr. Ly is interested in understanding whether exercise improves attention deficit problems. She is concerned that people taking ADHD medication might interfere with her ability to answer this question; she reasons that ADHD medication improves attention deficit problems and thus it might be difficult to also figure out the impact of exercise for these people. As Dr. Ly weighs her options, she should seriously consider which type of validity that will be directly affected by her decision? (Note: there are two correct answers - you need to pick one and justify your choice.)
Internal
- Is your IV responsible for the change in the DV, or is there an alternative explanation?
- Compare between conditions; control confounds; randomly assign participants to conditions
- Random assignment: ensuring that each participant in your experiment has an equal chance of being assigned to any given condition in the design
- Random assignment minimizes the influence of individual difference in the sample, particularly when the sample is large enough
External
- To what degree can your results be generalized to other samples or other situations?
- Random sampling
What is the difference between random sampling and random assignment?
Random sampling is used to generate a representative sample and/or is used to enhance external validity. Random assignment is used in experimental designs to make sure experimental groups are equal and/or is used to enhance internal validity.
Jen is a clinical psychologist interested in the link between mental illness and criminal activity. She uses cluster sampling to recruit patients from five inpatient/residential mental health facilities in her state. There are 4,307 patients living in these facilities. She asks patients whether they have ever been arrested for a crime and whether they have ever been convicted of a crime. Across, 1,309 patients, she finds that 27% (+/− 3%) report having been arrested for a crime but that only 13% (+/− 3%) have been convicted of a crime.
- Name a population of interest that Jen could reasonably generalize her findings to.
- Jen is interviewed by a journalist about the findings of her study. The journalist criticizes the sample size as being too small and says because of that, the findings do not really have any real world meaning. How should she respond?
- People with mental illness living in a residential facility in her state.
- Sample size is unrelated to external validity. Further, Jen should also state that this sample size is probably not too small, and the journalist would be wise to focus on effect size, margin of error and sampling technique instead.
Dr. Kubota sends her study to a journal to be published. In it, participants are exposed to an emotion induction which is manipulated between subjects. Some subjects are exposed to emotion-related words, while others are exposed to neutral words. One of the peer reviewers questions the way Dr. Kubota manipulated emotion, arguing that being exposed to emotional words does not make one emotional. The reviewer is questioning which of the following?
Construct validity
- How well have the conceptual variables been operationalized?
- are you measuring what you say you are measuring and are you manipulating what you say you are manipulating
- Subjective assessment: does this scale look like what happiness measure should look like?
- Objective assessment: do scores on this scale correlate with other measures of happiness?
- Criterion validity: the extent to which an operationalization of a construct, such as a test, relates to, or predicts, a theoretical representation of the construct—the criterion
- Discriminant (e.g., divergent) validity: tests whether concepts or measurements that are not supposed to be related are actually unrelated
________ enhances internal validity, while random sampling _______________.
Random assignment to condition, increases external validity
