PSY3213C Exam 2 Study Guide PDF

Preview

Summary

exam 2 study guide based on the class worksheet ...

Description

PSY3213C Research Methods in Psychology Exam 2 Study Guide (Chapters 6-8) -

“Check Your Understanding” Questions from the Chapters covered in this book. http://wwnorton.com/college/psych/research-methods-in-psychology/

Chapter 6 – Surveys and Observations: Describing What People Do ● Recognize the strengths & weaknesses associated with different question formats on surveys. (E.g. open-ended, forced-choice, etc.) o Yes/No ▪ Respondent answers yes or no to item ▪ Strength: straight forward , easy to ask ▪ Weakness: bias, overly simplistic o Forced Choice ▪ Respondent must select b/w two alternative responses (often agree or disagree) ● Ex: is the position of women better or worse than it was ten years ago? ▪ Strengths: easy, fast ▪ Weaknesses: there's no middle ground o Multiple Choice ▪ Respondent must select most suitable response from among several alternatives ▪ Strength: quick, easy to score ▪ Weakness: wording of questions could be confusing, guessing o Open-Ended ▪ Question is asked to which the respondent must construct his/her own answer ▪ Strengths: Information is more complete ▪ Weaknesses: Participant may not understand what you’re looking for, some answers may be omitted inadvertently, summarizing data is difficult ● What is a Likert-Type Scale? o A type of question where the individual answers by selecting a response from a designated scale. ▪ Typical scale might be: 1-strongly disagree, 2- disagree, 3-undecided, 4-agree, 5- strongly agree ● What is the advantage of “exact-point labeling” instead of using two anchors on a scale? o It reduces ambiguity of answers, because it gives a more specific response ● What is a “response set,” and what are ways that you can design a survey to reduce the likelihood of it? o The tendency for participants to answer all (or most) of the questions the same way whenever questions in a series all have the same choices for responding ▪ To reduce response set: make sure items include a mixture of positive and negative statements including some alternate phrasing of the same item.

● This forces respondents to move back and forth b/w opposite ends of scale so they can’t fall into a single response set. ▪ Example: Today’s teenagers are rude and disrespectful vs. Today’s teenagers are polite and courteous. ● What is socially desirable responding, and how can you reduce it? What is an implicit measure? o Giving “socially desirable” answers to a survey. Because respondents are embarrassed, shy, or worried about giving an unpopular opinion, they don’t tell the truth on a survey or other self-report measure. ▪ Can reduce it by ensuring that participants know that their responses are anonymous, or asking people’s friends to rate them (if we need to know personal information) o Implicit measure= designed to detect the strength of a person's automatic association between mental representations of objects (concepts) in memory. ▪ Ex: Implicit Association Tests sees if people associate white-negative/blackpositive combination (computer test) ● Be able to recognize different examples of “observer bias” o Observer bias: expectations influence their interpretation of the participants behaviors or the outcomes of the study ▪ Example: a teacher pays more attention to the students that sit in the first row Chapter 7 – Sampling: Estimating the Frequency of Behaviors & Beliefs ● What is the difference between a sample and a population? Understand the importance of both randomization and reducing bias in selecting a sample. o Sample: the group that is selected to represent the population, smaller set taken from that exact population o Population: the complete set of individuals or events that we want to represent or are interested in ▪ randomization is important so that you have a representative accurate study that can generalize to a larger population - people to have an equal chance of being in a study so they are not a bias sample ▪ reducing bias is important so that you have representative results about the population that you're trying to study, not just the people that are most convenient or initially interested in being in the study ● Be able to differentiate between the different types of both random and nonrandom sampling ( generate your own examples ) o Random Sampling: ▪ Probability sampling (AKA RANDOM SAMPLING): every member of the population of interest has an equal chance of being selected for the sample ▪ Systematic sampling: using computers or random # table, the researcher starts by selection two random numbers (if 4 and 7, start with the 4th person and count off by 7) ▪ Simple random sample: randomly selected from the population (like drawing from a hat) ● do you or do you not have a chance to be picked again

o ex: jury duty → some people who have done it and others who have not ▪ Cluster/multistage sample: divide the population of interest into clusters, then randomly sample from each ● ex: cluster: 5 colleges, include every student from those 5 colleges in the study ● Multistage: instead of including every student from those 5 colleges, researcher selects random sample of students from within each of the 5 selected colleges ▪ Stratified sample: researcher selects particular demographic categories on purpose then randomly selected individuals within each of the categories ● ex: 50% women in a class and 50% men in a class ▪ Oversampling: researcher intentionally overrepresents one or more groups ● ex: asking more people to participate in the study than are represented in the population ▪ Proportionate sample: the proportions of different groups in the population are reflected in the sample from the strata ● ex: 13% men and 87% women o Non- Random Sampling: ▪ Purposive sample: want to study only certain kinds of people they recruit only particular participants ▪ Convenience sample: available and willing - may result in bias sampling ▪ Self-selection: participants volunteer themselves to participate (internet poll) ▪ Snowball sampling: ask participants to recommend a few acquaintances for the study ▪ Quota sampling: identifies subsets of the population of interest and then sets a target number for the category in the next sample Chapter 8 – Bivariate Correlational Research ● Be able to identify if a Pearson’s correlation coefficient (r) is being correctly or incorrectly used to evaluate a correlation (i.e. are the two variables both on an interval or ratio scale?) o interval and ratio data use scatter plots because they are continuous o bar graphs are used for categories ▪ most widely used, can range from +1 to -1 where 0 is a non significant relationship ● How do r values help determine an effect size? Can an effect size be large even in cases with small r coefficients? o “r”: tells you the strength and direction of the linear relationship ● Would a bar graph or a scatterplot be more appropriate to examine data when one of your variables is measured on a nominal (categorical) scale? o A bar graph→ it would show the mean of all the people for those categorical groups and you would see if there was an association

● Understand p values as they related to statistical significance as well as what a p value less than .05 means. (How does it relate to the probability that your results were caused by random error?) o “p” value is reported with a correlation and helps evaluate the probability that the sample association came from a population in which the association is zero ▪ if the “p” is less than 5% it is statistically significant and there is a high probability in difference of the population ● reject the null (original) and accept the alternative ▪ if the “p” is greater than 5% it is non-significant and you accept the original hypothesis and reject the alternative ● How can looking at “subgroups” help explain unusual scatterplot results? o subgroups: what if the relationship is true for men not for women , look if its there for the whole population- if not it might only be there for a subgroup ▪ how you divide a factor within a study ● ex: how well do you study ? subgroup: does sleeping affect this , study people who sleep for 4 hours versus 6 hours ● What is the effect of an outlier on a correlation? o outlier: single case that stands out far away from a pack ▪ can have a strong effect of correlation coefficient, it can make something weaker or stronger depending on where it is located ▪ matter most when a sample is small ● Understand why correlations cannot establish causation (temporal precedence, thirdvariable problem, etc). o Why correlation cannot establish causation: Just because things are alike doesn't mean that one caused the other ▪ temporal precedence: causal variable must precede the effect variable - cause happened before the effect ▪ third variable problem:we can come up with an alternative explanation for the association between two variables, the alternate variable is the third ▪ directionality problem: we don’t know which variable came first ▪ covariance: how much the variables vary, the correlation/association ▪ internal validity: make sure that the only variable that is effect the study is the one that is being tested ▪ spurious association: two things that have a weird connection usually due to the third variable ● When is a third variable a moderator? o It is a moderation when the association relationship changes depending on the level of the third variable and the second variable becomes the moderator ● Understand the relationships between correlation and regression. How are r and R2 related? What is the line of least squares? o regression line helps you to calculate the correlation and predicts how individual people will fit into the scatterplot ▪ regression is a tool that tells you how apply a correlation to a scenario o r^2: relationship between the things that are being tested, how much of your outcome variable does it explain ( middle of a venn diagram) o r: direction and the strength of the line, pearson correlation coefficient

o line of least squares : the regression line...