Psych 100 Exam #2 Review PDF

Preview

Summary

Review for the 2nd Exam taken. Talks about observer biases, manipulated vs. measured variables, correlations, etc. ...

Description

PSYCH100 Exam #2 Review ✴ Questions that come earlier in a survey may affect how a participant answers later questions (True) (Order Effects) ✴ Participant is unusually happy because he thinks the study is about emotional expression is… Observer effect (not participant effect because that is not a thing) ✴ Observer effect - change in behavior of study’s participants; effect the observer has on the participant, how participant is effected by the observer ✴ Observer bias - influences observers interpretations of measured data ✴ Online surveys commonly suffer from… Self-Selection (people who feel more strongly on topic are more likely to contribute to survey) ✴ How are quota sampling and stratified random sampling similar? - Both identify subgroups that need to be studied ✴ Quota sampling - you must fit a certain quota in sample and a fixed number of participants is set (like taking first 20 participants that walk by) (Biased) ✴ Stratified Random Sampling - looking at a specific subgroup but randomly select from that subgroup (Unbiased) ✴ In a research study, which would tell you that an association claim is being made? - The measurement of two variables (if there is only two variables) ✴ Correlation Coefficients can be used in casual claims as well as association claims ✴ Statistical significance depends on… Sample Size and Effect Size (Direction of association does not matter) ✴ Advantages of Within-Groups Design: Fewer Participants, Researchers get more power to find differences between conditions, Participants in the treatment/control groups will be equivalent ✴ What is used to control for order effects in an experiment? - Counterbalancing ✴ A researcher would chose a pretest/posttest design to… Determine how groups change over time, to ensure the treatment/comparison groups will have equal numbers, to make a stronger causal claim (Does not allow for the study of spontaneous behaviors) ✴ Random Selection enhances External Validity and Random Assignment enhances Internal Validity ✴ When examining an association claim using a bar graph, what shows an association? - A difference in the height between the bars ✴ A study finds a correlation of r = .52, By Cohen’s benchmarks, magnitude of this is… Large ✴ In true experiments, Measuring is to the dependent variable as Manipulation is to independent variable ✴ What is the primary difference between pretest/posttest designs and within-groups designs.. The number of levels of the IV participants are exposed to ✴ If the p value is low (less than .05), reject the HO Effect size in which Cohen’s d =

Can be described as

Compared to Pearson’s r =

0.2 Small/weak

0.1

0.5 Medium/moderate

0.3

0.8 Large/strong

0.5

Used in causal claims, looks at the average difference between groups

1. A researcher conducted a study of 34 scientists. He reported a correlation between the amount of beer each scientist drank per year and the likelihood of that scientist publishing a scientific paper. R = -.55, p < . 01

- A negative correlation means that the more beer drank, the lessened likelihood that a paper was published, this is a strong negative correlation (more beer = less published)

- P < .01 means that the null hypothesis is rejected. Because it is less than .05, the results are statistically significant that the results showed a correlation. Unlikely to have occurred by chance.

- Popular Press headlined “Suds seem to skew scientific success”, is claim justified? - Covariance: Drinking and Success are usually negatively associated - Temporal Precedence: Hard to tell which came first; Ex. maybe drank more when not getting published - Internal Validity: No because there was no experiment, did not test for confounds - Because the three criterion were not met, the claim is not justified. 2. Drinking Sugared vs. Sugar-Free Lemonade

- Manipulation used as an independent groups design: - Randomly assign participants to a group that either drinks Sugar or Sugar-Free lemonade then have both groups rate the taste

- Manipulation used as a within-groups design: - Participants try a Sugar lemonade, measure taste ratings, then try Sugar-Free lemonade, measure taste ratings

- Disadvantage of using with-in groups design - Contamination, after trying the Sugar lemonade they may favor that over the Sugar-Free lemonade; can use Counterbalancing to cancel out order effect 3. Indicate if each variable below is MAnipulated/MEasured Participant’s age - ME Participant’s number of siblings - ME How long a participant is given to complete a test (30/20/10 min) - MA How long it takes a participant to complete a test - ME Dose of a drug participant is given (1/5 mg) - MA Height of participant - ME Participants score on test - ME Level of test participants receive (Easy/Hard) - MA Level of test participants choose to take (Easy/Hard) - ME Whether participants wore braces - ME Personality of experimenter (friendly/stern) - MA (when purposefully trying to act one way, not natural personality) 4. A cognitive psychologist believes that people learn better when they spread studying over several days, so she creates a study with 3 groups of participants. Each group studies the same Chinese vocab list (no participants have studied this verb before). Group 1 studies vocab for 20 min on the first day. Group 2 studied for 20 min the first and second days. Group 3 studies for 20 min on the first, second, and third days.

On the 4th day all participants were tested on knowledge of the vocab. People in Group 3 scored the best, so the researcher concluded that distributed studying does improve people’e ability to learn. ! What is the internal validity problem (Design confound, Selection Effect, Order Effect)? - Design Confound (Not Order Effect because this is NOT a within-groups design). The 3rd group studied the most so they should know the most. The study wanted to test the effectiveness of distribution of studying, but this study is not accounting for total time studied....