Research Methods 4
14 December 2020
question
Cross-Sectional Designs:
answer
A research design in which investigators compare groups of subjects of differing age who are observed at a single point in time.
question
experimental contamination:
answer
a situation that occurs when participants in one experimental condition are indirectly affected by the independent variable in another experimental condition because they interacted with participants in the other condition
question
generational effects:
answer
when a particular generation has had certain experiences, their experiences make them politically extinct
question
interrupted time series design with a reversal:
answer
a quasi-experimental design resembling the one-group pretest-posttest design but with a series of observations (measurements) before and after the treatment manipulation.
question
interrupted time series design with multiple replications:
answer
a quasi-experimental design resembling the one-group pretest-posttest design but with a series of observations (measurements) before and after the treatment manipulation.
question
local history effect:
answer
something else may happen to one group that does not happen to the other group (also called a selection-by-history interaction)
question
longitudinal design:
answer
A research design in which investigators observe one group of subjects repeatedly over a period of time.
question
nonequivalent control group design:
answer
used when equivalent one cannot be isolated
question
nonequivalent groups posttest-only design:
answer
a between-subjects design with at least two groups of participants that uses a subject variable or that creates groups that are nonequivalent
question
nonequivalent groups pretest-posttest design:
answer
a between-subjects design with at least two groups of participants that uses a subject variable or that creates groups that are nonequivalent
question
one-group pretest-post test design:
answer
A quasi-experimental design in which the subjects in the experimental group are measure before and after the treatment is adminstered, but there is no control group
question
pre-experimental design:
answer
When a single group is used in a research or two groups are not equivalent.
question
quasi-experimental designs:
answer
Research method similar to an experimental design except that it makes use of naturally occurring groups rather than randomly assigning subjects to groups.
question
quasi-independent variable:
answer
Variable selected (not manipulated) to create groups to compare (groups are selected)
question
regression to the mean:
answer
The tendency for extreme scorers on one measurement to move (regress) closer to the mean on a later measurement causing a change that would normally not happen in the population.
question
selection bias:
answer
Errors in the selection and placement of subjects into groups that results in differences between groups which could effect the results of an experiment.
question
time series design:
answer
A research design where subjects are measured at specific times before and after the treatment has been administered in order to determine the long term effects of the treatment
question
natural settings:
answer
involves the observation of ongoing behavior as it occurs naturally with no intrusion or intervention by the researcher
question
contrived settings:
answer
involves the observation of behavior in settings that are arranged specifically for observing and recording behavior
question
reactance:
answer
subjects know they are being observed
question
How to minimize reactance?
answer
Researchers may conceal the fact that they are observing and recording behavior. (disguised observation)
question
structured observations:
answer
the right choice: check lists, categories of behavior, ratings, time measurements (latency, duration)
question
3 ways to record behavioral observations:
answer
1. Does it occur in natural or contrived setting? 2. Will participant know they are being observed? 3. How will participant’s behavior be recorded?
question
experience sampling:
answer
measures the frequency of mental experiences; thought, mood, mind
question
unstructured observations:
answer
a) Field notes: a running record, a continuous record (more-like anthropology) b) Narrative: Not quantified; how to turn into #s?
question
probability sample:
answer
a sample in which every element in the population has a known statistical likelihood of being selected
question
sampling error:
answer
Random error due to the chance that a sample may not accurately reflect the characteristics of the population.
question
What factors influence the size of sampling error?
answer
1. # of groups (difference among means) 2. lack of experimental control 3. Intraparticipant variance 4. Unsystematic, uncontrolled, unidentified
question
drawbacks of phone surveying?
answer
not deep, trusting self-reports
question
stratified sampling:
answer
-Involves dividing the population into distinct sub groups (layers) and selecting a separate from each layer in the same proportion that they occur in the target population
question
What is the problem of non-response?
answer
non-completion decreases representativeness; participants must also complete survey once selected.
question
external validity:
answer
The extent to which the results of a study can be generalized to other populations or settings than the sample that was studied.
question
Type of sampling most used in psychological research?
answer
1. non-probability 2. non-random sampling 3. convenience sampling
question
Convenience samples of college students?
answer
1. Choosing random participants readily available 2. Student’s differ from average person 3. Large sample size
question
Validity?
answer
Test is supposed to measures what it was intended to, if it does not then the test is not meaningful or valid. Strict procedures, ensuring patient effort, performance & equipment accuracy & calibration
question
Types of response format on surveys?
answer
face to face, web, phone, mail
question
social desirability:
answer
responding in ways you believe the experimenter will approve of
question
case studies:
answer
Intensive investigation of a single or a few individuals. This is generally a non-experimental method. Ex post facto: after the fact or maybe during the fact.
question
4 uses of case studies:
answer
1. Source of ideas for further research 2. Describe rare phenomena 3. Test theories
question
Why are people reluctant to trust case studies?
answer
retrospective, ex post facto, possible experimenter bias, demand characteristics, observer bias
question
Quasis differ from real experiments if one or both of the following occur:
answer
•The subjects are not randomly assigned to conditions •Usually the experimenter does not control the IV
question
When/Why use a quasi?
answer
to study real world situations where experimentation is not possible and for program evaluation.
question
Threats to internal validity:
answer
history, maturation, pretesting effects, statistical regression, DV changes, groups are unequal, differential attrition and history.
question
Why doesn’t author like pretest-post test design?
answer
It has a bad, poor, low, internal validity design.
question
longitudinal:
answer
follow a group for a long period of time, tracking changes; often used in developmental research
question
cross sectional:
answer
compare age-effects, compare groups of different ages at the time the study is conducted.
question
cohort effects:
answer
Cohort effect enables us to understand older adults from different cultures and different geographical areas
question
Time series:
answer
make a series of measurements of the DV; often uses records or statistics that are otherwise being collected
question
program evaluation:
answer
evaluate effectiveness of a program
question
Why use quasi for program evaluation?
answer
To study real world situations where experimentation is not possible
question
Donald Campbell’s contributions?
answer
popularized quasi-designs; social & educational programs; internal & external validity
question
Criteria of cause-and-effect?
answer
1. The presumed causal variable preceded the effect in time. 2. The Cause and Effect co-vary. 3. All other alt. explanations of results are eliminated through randomization or experimental control.
question
patched-up designs:
answer
rather than adhering blindly to 1 particular design, helps provide meaningful data.
question
what does patch-up refer to?
answer
refers to increasing confidence in quasi-experimental results
question
experiments:
answer
active manipulation of the independent variable
question
non-experimental methods:
answer
no control over subject assignment; usually not of IV
question
4 non-experimental methods:
answer
1. quasi 2. surveys 3. behavioral observations 4. case studies
question
quasi experiments
answer
\”as-if\” experiments; sort of experiments
question
Quasi experiments in some ways look like experiments _________
answer
because they are experimental and have a control conditiong
question
A good quasi design uses both a _______:
answer
pretest and control group
question
Bad quasi-experimental design factors:
answer
poorly designed, poor control, low validity
question
Selection Threat
answer
groups may differ in important ways before the study even starts
question
Maturation Threat
answer
short or long term changes in participant’s may change their behaviors (boredom, physical development)
question
Attrition
answer
people may drop out of the study; those who do may differ from those who remain
question
History Threat
answer
some minor or major event may occur during a study that changes the participants’ behaviors