Explain how theories and data interact to form empirical inquiry.
A cycle: have a theory, gain data, add to theory
-Theory: leads researchers to pose particular...
-Research Questions: which lead to an appropriate...
-Research Design. In the context of the design, researchers formulate...
-Hypotheses. Researchers then c
� Describe how the peer-review process affects the communication of scientific ideas.
Rigorous process; ensure that the articles published in scientific journals contain innovative, well-done studies.
Learn to avoid the two common pitfalls of basing conclusions on intuition:
1.) When we think, we often take the easy way
2.) When we think, we are biased to see what we want to see
Understand why researchers should include a comparison group
control for confounds, and strive to evaluate information without bias.
Describe a variable in terms of both its conceptual definition and its operationalization and recognize the difference.
- Conceptual: something that varies and can be measured
- Operational definition: taking an abstract concept (i.e.:depression) and define it so that it is measurable.
Understand the three criteria that are used to evaluate a causal claim: covariance, temporal precedence, and internal validity.
- Covariance- there is an association between A and B
- Temporal precedence- A comes before B in time
- Internal Validity- there are no other possible causes for B except A (in other words, there are no confounds)
Identify three types of Reliability
-test-retest- take the test again and get a similar score,
-interrater- two observers ratings are consistent with each other,
-internal- people get consistent scores on every item in a questionnaire (part similar to whole))
Face Validity
measures what it looks like it would measure
Content Validity
measure everything it should
predictive measure is correlated with a relevant behavior outcome that will be observed in the future
Criterion
measure correlated with a relevant behavior outcome right now (predictive, postdictive, concurrent)
Convergent
measure related to other measures of similar constructs
Discriminant
validity measure unrelated to other measures of the dissimilar constructs
Construct validity
how questions are worded and such (convergent and discriminant)
leading questions
non-neutral q's that can push participants in a certain direction.
double- barreled questions
asks two questions at once (participants don't know which to answer)
double negatives
difficult to answer
question order
order can affect later responses
respondent shortcuts (such as nay- saying)
the person taking the test doesn't try
biases in answering quesitons
faking good- to be socially acceptable
Define sampling problems that lead to biased samples.
-Explain biased sampling and what sampling problems lead to it Biased- Sample is unrepresentative of the population
Causes of biased samples:
- only those who are easy to contact: Convenience
- only those who you are able to contact
- only those who invit
Explain five techniques for random sampling:
- Simple random sampling: put everyone's name in a hat and draw a few
- Systematic sampling: pick every 10th person (or another number)
- Cluster sampling: randomly pick a few clusters and then sample all members of the cluster
- Multistage sampling: rand
Understand how the correlation coefficient, r, represents strength and direction of a relationship between two quantitative variables.
How strong was the result:
r = about 0.1 Small or weak effect
r = about 0.3 Medium or moderate effect
r = about 0.5 Large or strong effect
Interrogate the statistical validity of an association claim, asking about features of the data that might distort the meaning of the correlation coefficient, such as outliers in the scatterplot, effect size, subgroups, or curvilinearity.
- Spurious (subgroups): finding statistical significance between subgroups, but it is not real
- Restriction of range (effect size): a variable does not vary enough; some possible values for the variable are left out
- Curvilinearity: a relationship might
Identify and interpret data from either a multiple regression or a longitudinal design.
Longitudinal (especially good when an experiment is unethical or impractical):
-cross-sectional: correlation of the two variables measured at the same time points
-autocorrelation: Correlation of a variable with itself across time
-cross-lag correlation:
Understand the difference between mediators, third variables, and moderating variables.
Mediation:- Variable 1 influences Variable 2 because of Mediator.
- the "how" behind a correlation test...
- younger age leads to divorce: younger age�> less maturity�> divorce
Moderator:- Variable 1 influences Variable 2 differently for different levels
the why in the relationship:
a. moderator
b. mediator
c. subgroup
d. third variable
b. mediator
Describe how the procedures for independent-groups and within-groups experiments are different. Explain the pros and cons of each type of design.
- Within-subject better than between subject : - Better control for systematic variability.
- Control for unsystematic variability.
- Require fewer participants.
- Disadvantages of within-subject design:
- harder to isolate what is really causing an effec
-Describe random assignment and explain its role in establishing internal validity.
Randomly assigning participants to the different conditions turns systematic variability into unsystematic variability, thereby eliminating confounds
-Describe counterbalancing and explain its role in establishing internal validity.
present the levels of the independent variable to the participants in different orders. (in within-subjects designs, participants are randomly assigned to these list conditions, not independent variable groups.)
Identify the following three threats to internal validity: observer bias, demand characteristics, and placebo effects, and identify ways to counteract these threats.
- Observer Bias- You see what you expect to see
(fix-blind study)
- Demand Characteristics- Participants guess what the study is supposed to be about and change their behavior in the expected direction
(fix- double blind)
- Placebo Effect - Participants r
Understand the reasons that a study might result in null effects
- there were no real effects to be found
- operationalization of the variables poor
- too much unsystematic variability
( not enough variance between groups, too much variance within groups, or a true null effect. )
Identify potential causes of insufficient between-group variance and too much within-group variance.
between-groups: so many differences, can't focus on what is important (need a more sensitive measure)
within-groups:- not enough power, not enough participants (need to increase power with participants)
Understand and be able to explain/apply the three ethical principles of the Belmont Report.
Respect for Persons (informed consent/children & prisoners are not able to give)
Beneficence- welfare of the subject in mind. Harm vs. Benefit
Justice - fair across kinds of people.
Explain informed consent and the protection of vulnerable groups (applying the principle of respect for persons).
Informed consent means that participants must have enough info about the research and its risks and benefits.
Protection of vulnerable groups: children, prisoners, mentally ill. Anybody with reduced autonomy must be protected.
Explain how researchers might evaluate the risks and benefits of a study (applying the principle of beneficence).
Ask if the benefits outweigh the risks
If your research won't produce good results, it doesn't matter how little risk is involved
Explain how researchers would apply the principle of justice in selecting research participants.
All participants should be equally exposed to the risks and benefits of research
Those that will benefit from the study should be involved in the study
Know when and how Psychologists may ethically use deception in their research.
It may be needed in order to manipulate the variables in the study
May be necessary when the participants knowing the hypothesis may ruin the results
In order to use deception you must prove that the research couldn't be done without it.
The research must
Understand what the IRB is, what its purpose is, and what a researcher's responsibilities are relative to the IRB.
IRB is an ethics committee that reviews research ideas.
It is for living human research only
Responsibility: report all risks before to get it passed, report misconduct, report unforseen risks
Become familiar with the points in the APA's Ethical Standard 8 (the standard that most closely applies to research in psychology). Know what behaviors constitute research misconduct.
Data fabrication- researchers invent data to fit hypothesis
Data falsification-researchers influence studies results, influence test subjects, delete results
Plagiarism
Only about one time in 10,000 allegations is true misconduct confirmed.
Describe how research using animals is regulated and what ethical principles apply to animal research.
IACUC -
replacement: find alternatives to animals where possible
refinement: modify procedures to reduce animal distress
reduction: use as few animals as possible
Describe the differences among direct replication studies, conceptual replication studies, and replication-plus-extension studies.
Direct Replication Studies: exact replication- repeat exactly
Conceptual Replication Studies: same question different procedure
Replication-plus-extension Studies: Same question, same procedure, add an element
Explain what a meta-analysis does and what it has in common with direct and conceptual replication.
Mathematically averaging a bunch of studies together.
Direct and conceptual: more of a conceptual replication: main trend of a body of literature
Give examples of how external validity applies both to other participants and to other settings.
other participants: needs to be true even when applied to other people not sampled other settings: needs to be true even in other settings that are not part of the sample (if these are not met, results were only because of the population in that setting a
Describe the difference between generalization mode, in which external validity is essential, and theory-testing mode, in which external validity is less important than internal validity and may not be important at all. Reevaluate two common assumptions f
Theory testing mode- testing of association or causal claims to investigate support for a theory
Contact Comfort Theory: Artificial situation to test the theory, no random sample of monkeys
basic research
Generalization mode- want to generalize findings f
Understand and be able to recognize and apply different measures of central tendency.
Central Tendency: A measure of values the individual scores tend to center on.
Understand what Standard Deviation represents.
how far, on average, each score in a data set is from the mean, square root of variance
Variance- how spread the scores of a sample are around their mean SD^2, average distance from the mean squared
Understand what a z-score is.
describes where an individual's score is above or below the mean and how far it is from the mean, in standard deviation units
express the data in number of standard deviations from the mean,
allows for comparison of different scores with different scales
Be able to interpret numeric and graphical representations of different measures of relationship size/strength, including the correlation coefficient and Cohen's d. Understand when each would be appropriate to use.
-r: at least 2 continuous variables (.1,.3,.5)
-d: at least 1 categorical variable (.3,.5,.7)
Understand and be able to apply the logic of statistical inference.
utilizes law of chance and probability to help researchers make decisions about what their data mean and what inferences they can make
-test hypotheses
Know the 4 steps in statistical inference.
Step 1- Assume there is no Effect (Null hypothesis)
assume nothing is going on, no effect
Step 2- Collect data
Step 3- Calculate the probability of getting such data, or even more extreme data, if the null hypothesis is true
Step 4- Decide whether to reje
Classify different types of errors in statistical inference (Type I, Type II).
Type 1 error- false positive, "We could conclude that Sarah probably has special abilities, when she really does not."
Type 2 error- a miss, "We could conclude that Sarah probably does not have special abilities, when she really does
Know what a t-test is and when it should be used.
allows researchers to test whether the difference between two group means in an independent-groups design is statistically significant,
Sampling distribution of t leads to comparison with p value
Know which statistical tests produce t statistics and which produce F statistics. Know what t and F statistics mean.
The T-test
compares two means together
The F Test (Anovas)
Anova - analysis of variance
ANOVA compared variability between groups with variability within groups
T vs. F
F is t squared
both give p values
T is directional (can be positive or negative), f is
Be able to correctly decide what statistical test is appropriate for a given data set/study design.
Factorial designs: (two or more independent variables) Anova, T-test: two simple variables
Know the standards for reporting statistical results.
Examples:
symbols (such as M, SD, t for t test, and F for ANOVA) are presented in italics, but the numerals themselves are not (M=3.25).
-for p greater than .001 write EXACT value
(also only put a 0 in front of a decimal that CAN be bigger than 0)
Describe how an APA-style paper should be formatted, including title page content and section headers.
Header! (Title, Authors & affiliations, Logos)
-Same sections as a paper (Just much shorter!):
Intro
Methods
Results
Discussion
References
Correct APA table:
title on top, title and abbreviations: italicized
No vertical lines, few horizontal ones
Label units
Explain any abbreviations
Correct APA figures:
Bar graphs for categorical data: title on bottom
Scatterplots w/ trendline for correlations
Label units
Explain any abbreviations
Know what makes a good poster
FEDACLAM
Focus on findings.
Emphasize graphics
Avoid 'chart junk.'
Choose colors wisely.
Leave white space.
Aim for symmetry.
Design for your readers' eyes.
Mind the details.
Which claim is most likely to be tested in :generalization mode" rather than "theory testing mode"?
-Frequency claim
In this study, the authors are interested in student's math effort. which of the following would NOT be a useful operational definition of math effort?
-The number of math classes taken in college
Which of the following is an example of being a producer of research?
-Administering an anxiety questionnaire
Hosea is studying the relationship between caffeine consumption and problem-solving ability. Which of the following is a categorical way to operationalize caffeine consumption?
-Whether the participant drank a soda in the 24 hours prior to the study
Ellie is looking for a summary of research on the effects of childhood abuse on adult functioning. Which of the following scientific sources would be the least helpful?
-An empirical journal article
Known groups paradigm (GA and AA people)
-Criterion validity
How do we talk about interactions?
-The effect of variable A depends on Variable B
when a researcher is in theory-testing mode, which validity is most important?
internal validity
deception in psychology studies should be followed with
a debriefing
*Why do we have a null hypothesis?
Because a single study can show that something is NOT true, but not that something is true.
*P=.02, what are the chances that there really is no difference?
2%
� Articulate how a crossed factorial design works.
Identify different types of factorial designs.
Studies two or more independent Variables
Independent-groups (between subjects)
Within-subjects
Mixed factorial: one of both between and within
Given a factorial notation (e.g., 2 � 2), identify the number of independent variables, the number of levels of each variable, the number of cells in the design, and the number of main effects and interactions that will be relevant.
In a 2x2 design there are 2 independent variables, each with 2 levels.
- 4 cells
Main Effects: find if there are differences "In a factorial design with two independent variables, there are 2 main effects
Dr. Rhodes notices an interaction in his factorial study. Which statement might he use to explain it?
-Variable a cancels out variable b
-Effect of variable A depends on variable B
-Variable a mainly affects variable b
-Effect of variable A is mediated by
Effect of variable A depends on variable B -moderator is interaction (aka depends)