lecture 5 Flashcards
operational definitions:
definitions of variables in terms of the operations needed to produce them
how to define operations?
replication by other researchers and scrutiny by other researchers
extraneous variables (confounders)
have an unintended influence on the results of experiment by changing difference between groups
when an extraneous variable is present. the experiment is:
confounded
nuisance variables:
unwanted variables that can cause the variability within all groups to increase
example of extraneous variable:
giving encouragement and warning to gaming participant
example of nuisance variables:
not holding ruler a precise distance above participants hands
how to control extraneous and nuisance variables:
produce groups that are equivalent prior to the intro of the IV. reduce the effects of nuisance variables as much as possible
randomization:
distributes extraneous variables equally to all groups. eg those who drink cola A regularly and those who regularly drink cola B
elimination:
completely remove the extraneous variable from the experiment. ex test and remove those who are loyal to cola A or cola B, only test those who are undecided so there are no biases
constancy:
keeps an extraneous variable constant. ex only test people who haven’t drank any of their fav cola that day
balancing:
ensures that all participants receive extraneous variable to the same extent. ex if one always prefers the first cola since they’re more thirsty
counterbalancing can control for:
sequence of order effects
lesson of M&M sampling:
sample size really does matter, there is power in. large numbers
we must have a sample that is:
representative of the population we are studying
a representative sample is most likely with a ____ sample, where each member of the population has an ___ ___ of being selected in sample
random; equal likelihood
types of sampling: (3)
- random sampling without replacement
- random sampling with replacement
- stratified random sampling
basic research strategies:
single strata, cross sectional, longitudinal
single strata:
looking at a single thing of the population of interest, ex percentage of green peanut m&ms
cross sectional:
comparing of two or more groups during the same limited time period. ex percentage of green peanut verses green plain m&ms
longitudinal:
obtaining research from the same group of people over extended period of time
stratified random sampling
the total population is divided into homogenous groups (strata) to complete the sampling process and sample is drawn from each strata
random sampling without replacement:
a subset of the observations is selected randomly, and once an observation is selected it cannot be selected again.
random sampling with replacement
a subset of the observations is selected randomly, and once an observation is selected and can be selected again.
validity of tests and inventories include:
content validity, concurrent validity, criterion validity
concurrent validity:
the results of a particular test or measurement correspond to those of a previously established measurement for the same construct
criterion validity:
evaluates how accurately a test measures the outcome it was designed to measure
reliability for tests and inventors includes:
test retest, split half
pilot testing:
preliminary testing done in advance of the complete study
types of surveys:
mail/email/web survey, personal interview, phone interview
mail/email/web survey benefits and negatives
fast, can get large numbers of participants, return rate can be low
personal interview benefits and negatives
slow, expensive, interviews bias
phone interview benefits and negatives:
good compromise, ability to randomly sample before cell phones
correlational studies:
looks at the relationship between two continues variable , positive, negative and zero
ex posto facto studies:
“after the fact” use an independent variable that has already varied
ex posto studies sometimes called:
subject variables eg, gender, SES, experiences are things you cannot or would not manipulate
