Research Design Final - Chapter 10-14

complex experiments solving primary issue of basic experiments

can examine presence of curvilinear relationships

factorial design

experiments with more than one factor on dependent variable

main effects and interactions

effect of one factor while ignoring others vs effect of one factor depends on levels of other factors

IV x PV design

includes both a manipulated independent variable and a participant variable

purpose of single case designs

determine whether an independent variable affects a single participant

design features of quasi experiments

manipulate variables like true experiments but do not have random assignment (cannot control for extraneous variables)

cross sectional

compare different ages at one point in time

longitudinal designs

same group of people at multiple time points

sequential design

combine longitudinal and cross sectional

history threat (threats to internal validity)

anything that happens during a study that may affect the dependent variable besides the independent variable

maturation threat (threats to internal validity)

changes that occur among participants due to the passage of time

regression threat (threats to internal validity)

extremely low or extremely high performance at time 1 is likely to be less extreme at time 2

testing threat (threats to internal validity)

taking a test once can affect future performance

instrumentation threat (threats to internal validity)

changes in the measures, or how measures at interpreted, can affect results

discrete vs continuous

distinct, countable values with gaps vs can take on any value with infinite values between two

descriptive vs inferential statistics

organizing and summarizing data vs using statistics to make conclusions about population parameters

statistics vs parameters

sample vs population

positively skewed distributions

tail on right side, mode median, mean (smallest to largest), more scores at low end

negatively skewed distributions

more scores at high end than low end, tail on left side, smallest to largest: mean median mode

null vs alternative hypothesis

no significant difference vs significant difference

ecological validity

whether results from a lab setting generalize to real-world settings

direct replication

only change participants

conceptual replication

use same conceptual variables but change operational definitions of manipulations and/or measures

replication-plus-extension

add other variables to test other questions

underreporting null effects

use open materials

p-hacking

using many ways to analyze data until you get statistically significant results - solution is open data

HARKing

hypothesizing after results known - solution is preregistration

using small samples

require larger samples do thorough power analysis before study