lecture 9 Flashcards
what is the random groups design
there are two groups and participants are randomly assigned to two groups/conditions
what is random assignment
each participant has equal change of being assigned to each of the groups
the participants in the groups of random group designs are ______________
non-overlapping or independent from one another
how do you remove confounds between the control group and the experimental group
equating the groups
the variable being manipulated is
the independent varaiable
the variable that shows the effect
dependent variable
explain levels of independent variables
there are levels when manipulating the independent variable. the control is a level of IV and the experiment group is also another level of IV
ex. you test chocolates effect on test taking
levels are = small sized vs medium sized vs large sized chocolate
differentiate correlational designs and experimental designs
correlational designs are non-experimental and do not lead to conclusions about causality. ideal when IV cannot be manipulated
what type of design is described as “between subjects”
independent group designs
describe independent groups design
participants are non-overlapping meaning they are either in the control or the experimental group
what is the purpose of the random-group design
to establish causality since you are removing possible confounds through andomization
what type of group is a placebo
control group
what variable is the hypothesized outcome
dependent variable
what is the purpose of random assignment?
to create equivalent conditions and rule out differences between groups that could impact results
why do you need to randomize the order of data collecting
bc if you dont, you add a confound of duration of the test between the control and experimental groups
what is statistical power
probability of detecting an effect (researchers ability to find an effect size of interest)
need enough participants
what are underpowered studies
studies that do not have enough participants to see effect size
what is likely to happen if a study is underpowered
no effect can be seen thus the null hypothesis will be accepted
what is data peeking
checking for statistically significant results after collecting a small subset of data and stopping data collection
what is the purpose of a representative sample
to have external validity where the results of a study is applicable to population
confounding variables are a threat to ________
internal validity
describe representative samles
have characteristics representative of broader population
what are the 3 examples of confounding variables
differences in:
- environment
- individuals
- researcher (researchers do sntg different)
what is a way to avoid the Clever Hans Phenomenon (the horse)
making sure the researcher is unaware of the experiment’s hypothesis
what is the single-blind design
the participants dont know which group they are assigned
what is a double-blind design
both experimenter and participants dont know hypothesis
talk about the clever hans phenomenon
a horse was said to know how to solve math but it was just responding to physical cues from the experimenter - research confounds
what are matched groups
participants in each group arae matched on particular characteristics of interest (i.e. iq, gender, age..)
what is the importance of matched groups
to ensure groups are equivalent
