Experimental Psychology part 2 Flashcards
Self report
When people are asked about their attitudes
Computerized
Type of sampling where participant reports via beeper, computer etc.
Interviewer bias
Lead them to specific answers
Respondent bias
Social desirability
Engage in response set (1,1, 1, etc…)
demand characteristics
talk a lot with participants and ignore others etc
Types of sampling
Probability
Non probability
Probability
Everyone has an equal chance of being chosen.
- stratified random sampling
- clusters
Non probability
- haphazard (convenient sampling)
- purposive sampling
- quota sampling: same as stratified but without the
Stratified random sampling
Representative sampling (freshman, sophomore, junior, senior)
Clusters
states, counties, cities, districts, (prisons/school), classes ;; not individual but clusters
Haphazard
convenient sampling
Purposive sampling
The researcher chooses the sample based on who they think would be appropriate for the study
Quota sampling
Same as stratified but without the random sampling (haphazard technique)
Problems with wording survey
yay or nay saying -negative wording
yay or nay saying
negative wording
loaded question
double barrel
yay or nay saying
agree to everything or disagree
reverse questioning is used to avoid this
loaded question
ask a question with an assumption already in mind or desired answer
double barrel
ask two questions in one (ie have you ever had a stroke or a heartattack?)
IV, DV, Control
IV: Manipulate
DV: measure
Control everything else
Internal validity
IV is influencing the DV
External validity
Experiment can be generalized (run study 100x with different population, setting etc, expect similar results)
Why do we do random sampling?
to avoid confounding variables
Mortality
When people drop out of a study
Between subject
participates in one group
Within subject
(repeated measure design) each participant participates in all groups/ every condition
Match pair design
When you give them a test based on characteristics then randomly assign one to condition A the other to condition B
Pretest/post test design
a pretest is given before the experimental manipulation; pretest=control;
Quasi experimental design
a type of design that approximates the control features of true experiments to infer that a given treatment did have its intended effect
Problems with Reserach designs
history effects maturation instrument decay testing regression of the mean