research methods Flashcards
naturalistic observation
viewing people in their natural environments
only able to record visible behaviours
observation bias- being watched may change behaviours
experimenter bias-observer records what they want to see
confounds
survey
convenient
social desirability bias- lie to look good
systematic observation
collection of behaviours in controlled tasks ex. implicit association test
high internal validity but low external validity
archival data
using available records
collect data quickly
experience sampling
longitudinal, participants report over time
correlation research design
examines relationship among multiple variables
does not address causality
longitudinal
tracking participants for extended period
addresses causality and direction
experimental design
manipulate at least one variable to cause effect on another
control group and experimental group
random assignment
between subjects
each participant experiences only one level of independent variable
within subjects
each participant experiences all levels of the variable
counterbalancing
quasi experimental
matched groups rather than random assignment
cover story
pro: behaviour will be more natural
con: more work to pass through ethics, debrief, people hearing about what the study is actually about
Nominal/categorical
ex. male, female
ordinal
ex. letters, rank
interval
temperature, year, iq
ratio
weight, # of clients
internal consistency reliability
consistency across questions
parallel forms reliability
consistency across forms of a test (splitting items into two tests)
average inter item correlation
how correlated are the questions with eachother on average (correlate every single item with every single other item)
average item total correlation
how correlated are the items with the total on average (average totals)
split half reliability
if you randomly split the items into two sets, how correlated are they
cronbach’s alpha
what is the average of all possible split half correlations
measurement error
when measurement isnt consistent
classic test theory
our observed score is a combination of the true score and error
