All cards Flashcards
Beat Dr. Askari's final exam
Dependent Variable
The variable measured in a study. Its value is determined by the behavior of the subject and may depend on the value of the independent variable.
Independent Variable
The variable that is manipulated in an experiment. Its value is determined by the experimenter, not by the subject.
Scientific Theory
A theory that goes beyond simple hypothesis, deals with verifiable phenomena, and is highly ordered and structured.
Hypothesis
A tentative statement, subject to empirical test, about the expected relationship between variables.
Applied Research
Research carried out to investigate a real-world problem.
Basic Research
Research carried out primarily to test a theory or an empirical issue.
Experimental Research
Research which has manipulation of one or more independent variables and control over extraneous variables.
Confounding
Two variables that vary together in such a way that the effects of one cannot be separated from the effects of another.
Random Sample
A sample drawn from a population such that every member of the population has an equal opportunity to be included in the sample.
Random Assignment
The process of randomly assigning subjects to experimental treatments.
Counterbalancing
A technique used to combat carryover effects in within-subjects designs. Counterbalancing involves assigning the various treatments of an experiment in a different order for different subjects.
What are the two ways of imposing control in an experiment?
- Hold extraneous variables constant.
2. Randomize the effects of extraneous variables across groups.
Internal Validity
The extent to which a study evaluates the intended hypotheses.
External Validity
The extent to which the results of a study extend beyond the limited sample of the study.
Null Hypothesis
The hypothesis that there is no significant difference between specified populations, and that any observed difference is due to sampling or experimental error.
Operational Definition
A definition of a variable in terms of the operations used to measure it.
Test-Retest Reliability
A method of assessing the reliability of a questionnaire by repeatedly administering the same or parallel form of a test.
Pilot Study
A small, scaled-down version of a study used to test the validity of experimental procedures and measures.
Face Validity
How well a test appears to measure (judging by its contents) what it was designed to measure. Example: A measure of mathematical ability would have face validity if it contained math problems.
Floor Effect
Range effect where a dependent variable reaches its lowest possible value.
Ceiling Effect
Range effect where a dependent variable reaches its highest possible value.
Deception
A research technique in which participants are misinformed about the true nature and purpose of a study.
Hawthorne Effect
The alteration of behavior by the subjects of a study due to their awareness of being observed.
Informed Consent
Agreeing to serve as a research participant after being informed about the nature of the research and the participant’s rights and responsibilities.
The participant typically reads and signs a form specifying the purpose of a study, the methods to be used, requirements for participation, costs and benefits of research participation, that participation is voluntary, and that the participant is free to withdraw from the study at any time without penalty.
Between-Subjects Variable/Design
Where different groups of subjects are exposed to different levels of the independent variable.
Within-Subject Variable/Design
Where each subject is exposed to all levels of an independent variable.
Latin Square Design
Method of counterbalancing in a partially counterbalanced design, where the number of treatment orders equals the number of groups.
The result of a Latin square design is that each treatment appears at each ordinal position an equal number of times.
What are the advantages and drawbacks of experiments?
Advantages: Ability to draw causal relationships; high internal validity
Drawbacks: Unusable where independent variable cannot be directly manipulated; high requirement for control of extraneous variables; lower external validity
What are the advantages and drawbacks of correlational studies?
Advantages: Easy to carry out; usable when researcher cannot manipulate variables; able to study variables in messy real-world environments; high external validity
Drawbacks: Third-Variable Problem; Directionality Problem; causality cannot be determined; low internal validity
What is the difference between complete and partial counterbalancing?
Complete counterbalancing provides every possible order of treatments and assigns at least one subject to each; Partial counterbalancing provides only a subset of possible treatments
Carryover Effect
A problem associated with within-subjects designs in which exposure to one level of the independent variable alters the behavior observed under subsequent levels.
Order Effects
Effect of position within a study on the outcome of a question
Mixed Design
An experimental design that includes between-subjects as well as within-subjects factors. Also called a split-plot design.
Factorial Design
An experimental design in which every level of one independent variable is combined with every level of every other independent variable.
Time Series Design
A research design in which behavior of subjects in naturally occurring groups is measured periodically both before and after introduction of a treatment.
What are the advantages and drawbacks of using unobtrusive observation?
Advantages: higher internal and external validity
Drawbacks: Often requires special equipment; not always possible
Solomon 4-group Design
An expansion of the pretest-posttest design that includes control groups to evaluate the effects of administering a pretest on your experimental treatment.
What are the major threats to internal validity?
- History - events other than the treatment which occur between observations
- Maturation - performance changes due to age or fatigue
- Testing - performance changes due to familiarity with test
- Instrumentation - poorly calibrated equipment or unobserved changes in observer criteria
- Statistical regression - those with extreme scores tend to have more average scores if tested again
- Biased selection of subjects - groups are not equivalent prior to treatment
- Experimental mortality - loss of subjects at different rates results in different groups
- Diffusion - participants know what to expect from the experiment due to leaked information
- Selection by maturation interaction - one group develops more quickly due to extraneous factors
Selection Bias
Nonrandom sampling which leads to groups which are unequal or which do not represent the intended population.
Single-Blind Technique
The person testing subjects in a study is kept unaware of the hypotheses being tested.
Double-Blind Technique
Neither the participants in a study nor the person carrying out the study knows at the time of testing which treatment the participant is receiving.
Generalization
Applying a finding beyond the limited situation in which it was observed.
Demand Characteristics
Cues inadvertently provided by the researcher or research context concerning the purposes of a study or the behavior expected from participants.
Experimenter Bias
When the behavior of the researcher influences the results of a study.
Experimenter bias stems from two sources: expectancy effects and uneven treatment of subjects across treatments.
Replication
Establishing the generality of results by exposing other subjects or samples to similar treatments and demonstrating similar results.
Placebo
A harmless pill, medicine, or procedure given to the control group.
Univariate Study
A study with only one dependent variable.
Multivariate Study
A study with more than one dependent variable.
Descriptive Statistics
Statistics that allow you to summarize the properties of an entire distribution of scores with just a few numbers.
Inferential Statistics
Statistical procedures used to infer a characteristic of a population based on certain properties of a sample drawn from that population.
What are the parts of a research paper, and what goes in each?
- Title page
- Abstract - brief description
- Introduction - lit review, statement of hypotheses
- Method - participants/subjects, apparatus/materials, procedure
- Results - inferential and descriptive statistics, no interpretation
- Discussion - restatement of findings; link results to hypotheses; explain what this link means to theory and study; new research which has come up since writing introduction; confounds and limitations; implications; suggestions for future research
- References
- Appendices, Tables and Figures
What is the difference between a within subjects vs. between subjects t-test?
Within subjects: one value is calculated for each subject. These values are compiled into a single distribution and t-value is calculated based on the standard deviation and number of subjects.
Between subjects: also called an “independent samples test”, uses means and standard deviations from both samples to establish t-value.
Correlational relationship
A relationship in which the value of one variable changes systematically with the value of a second variable.
Analysis of Variance
An inferential statistic used to evaluate data from experiments with more than two levels of an independent variable or data from multifactor experiments. Versions are available for between-subjects and within-subjects designs.
Interaction Effect
When the effect of one independent variable on the dependent variable in a factorial design changes over the levels of another independent variable.
Repeated Measures Variable
Experimental DV which is measured several times for each participant. For instance, in a longitudinal study, the same subjects are measured several times. DVs from these assessments are repeated measures variables.
Main Effect
The effect of a single independent variable in a factorial design on the dependent variable. There are as many possible main effects as there are independent variables.
Construct Validity
Validity that applies when a test is designed to measure a “construct” or variable “constructed” to describe or explain behavior on the basis of theory (e.g. intelligence).
A test has construct validity if the measured values of the construct predict behavior as expected from the theory (e.g., those with higher intelligence scores achieve higher grades in school).
Parsimonious Explanation
An explanation or theory that explains a relationship using relatively few assumptions.
How many possible types of significant outcomes are there for a 2x2 Factorial Design?
8 - two possible main effects and one possible interaction exist. 2 x 2 x 2 = 8.
Cohort-Sequential Design
A developmental design including cross-sectional and longitudinal components.
What are the steps in the research process?
- Idea
- Idea becomes hypothesis
- Research design chosen
- Subjects chosen
- Measures chosen
- Study conducted
- Data analyzed
- Results reported
Cross-Sectional Design
A developmental design in which participants from two or more age groups are measured at about the same time. Comparisons are made across age groups to investigate age-related changes in behavior.
Longitudinal Design
A developmental design in which a single group of subjects is followed over a specified period of time and measured at regular intervals.
Parametric Design
An experimental design in which the independent variable is systematically varied across several levels.
Nonparametric Design
Experimental research design in which levels of the independent variable are represented by different categories rather than different amounts.
Criterion-Related Validity
The ability of a measure to produce results similar to those provided by other, established measures of the same variable.
Reliability
Whether a measure or questionnaire produces the same or similar responses with multiple administrations of the same or a similar instrument.
Parallel-Forms Reliability
Establishing the reliability of a questionnaire by administering parallel (alternate) forms of the questionnaire repeatedly.
Split-Half Reliability
A method of assessing reliability of a questionnaire using a single administration of the instrument. The questionnaire is split into two parts, and responses from the two parts are correlated.
Name the three points of the Belmont Report.
- Respect for persons - participants must be able to make their own choices, and vulnerable persons deserve protection.
- Beneficence - do no harm to participants, and maximize benefits to participants
- Justice - researcher and participant share costs and benefits of research
Name the ten points of the Nuremburg Code
- Participants must give informed consent
- Research should yield beneficial results which can’t be obtained another way
- Research must have a sound footing in animal research and the natural history of the problem studied
- Avoid unnecessary harm to subjects
- Research should not be conducted if it will possibly result in death or disability of subjects
- Risk to subjects must be proportional to benefits of results
- Plans and facilities must protect subjects against harm
- Research should only be done by highly qualified scientists
- Subjects are free to withdraw from the experiment at any time
- Research must stop if it becomes evident that the research causes harm
What information must be given in order to show informed consent?
- Purpose, duration and procedures
- Right to withdraw
- Consequences of withdrawing or declining
- Foreseeable factors that might induce a subject to decline, such as risks or discomfort
- Research benefits
- Limits of confidentiality
- Incentives for participation
- Contact info for questions about research and participants’ rights
What extra information must be provided for informed consent in an experimental situation?
- Experimental nature of the treatment
- Services that will or will not be available to the control groups
- Method of assignment to group
- Available treatment alternatives if a subject declines or withdraws
- Compensation or costs of participating
When can informed consent be bypassed for visual or audio recordings?
- Naturalistic observations in public places, with no possibility of identification or harm to subjects
- Deceptive research where informed consent is gained after the recording
When can informed consent be completely dispensed with?
- When there is no distress or harm to subjects, including:
1a. Educational settings
1b. Anonymous settings or archival research
1c. Work settings where confidentiality is protected
- Where otherwise permitted by law or regulations
Matched-Groups Design
Between-subjects experimental design in which matched sets of subjects are distributed, at random, one per group across groups of the experiment.
Matched-Pairs Design
A two-group matched groups design.
Third-Variable Problem
Problem of establishing causation in a correlational study.
It is possible that two correlated variables are each caused by the level of a third unknown variable, and thus causation cannot be established.
Directionality Problem
Problem of establishing causation in a correlational study.
Given a correlation between causally connected variables, it is not clear which direction causality goes. A -> B and B -> A are both plausible, and thus causation cannot be established.
What are the advantages and drawbacks of a matched-group design?
Advantages: Better control over variables which may affect outcome measures
Drawbacks: Lower power of statistical tests due to altered distributions; more demanding and time consuming to implement; larger subject pool to find matches
What are the advantages and drawbacks of a within-subjects design?
Advantages: Lower variance of outcomes; higher power in statistical tests; fewer subjects necessary
Drawbacks: More demanding on subjects; higher possibility of subject attrition; carryover effects
What are some sources of carryover effects?
- Learning
- Fatigue
- Habituation
- Sensitization
- Contrast (change in motivation due to changed rewards)
- Adaptation
What are some ways of dealing with carryover effects?
- Counterbalancing
- Take steps to minimize carryover (i.e. washout periods or practice sessions)
- Separate carryover effects from treatment effects by making treatment order an independent variable
Nested Design
Experimental design where a within-subjects variable is nested underneath a between-subjects variable.
Quasi-Independent Variable
A variable resembling an independent variable in an experiment, but whose levels are not assigned to subjects at random (e.g., the subject’s age).
Quasi-Experimental Design
A design resembling an experimental design but having at least one quasi-independent variable.
What are the advantages and drawbacks of quasi-experimental designs?
Advantages: Allow the researcher to evaluate effect of QIVs under naturally occurring conditions
Drawbacks: Researcher does not have control over the variables influencing behavior; little or no control over when an event will occur; lower internal validity than a true experiment due to confounding
What are the advantages and drawbacks of longitudinal designs?
Advantages: Permits researcher to see developmental changes clearly
Drawbacks: Attrition; generation effects; multiple-observation effects (i.e. carryover effects); confounds with age; confounds with prevailing attitudes at time of assessment
Criterion Variable
Outcome variable in a correlational study (e.g. height in correlation of age and height)
Predictor Variable
Input variable in a correlational study (e.g. age in correlation of age and height)
When is deception ethical?
Deception is ethical if the researcher can demonstrate that important results cannot be obtained in any other way.
When is deception unacceptable?
Deception is not used if the research is reasonably expected to cause physical pain or severe emotional distress.
When must deception be revealed?
Deception must be revealed during debriefing, and the experimenter must permit the participant to withdraw his or her data after being informed of deception.
