Methodology & Stats Flashcards
What defines pseudo science
Ideas driven by ideological, cultural or commercial goals
Presented as scientific even though they aren’t
May be consistent with the truth but this doesn’t make a good theory
Scientific law
Something that happens without exception and has been established by repeated testing.
It specifies under such-and-such conditions, such-and-such will occur
Often describes a mathematical relationship e.g. The law of gravity
Scientific theories
A conceptual framework the explains existing observation (and the laws based on them) and predicts new ones
What’s the difference between scientific theory and what theory means in layman terms
Layman - theory is something unproven, an educated guess or assertion
Science - must be supported by substantial evidence and has been proven beyond reasonable doubt
What’s better a scientific law or theory
Trick question, they are equal
What 6 things are scientific theories evaluated on
Principle of parsimony (occams razor) Falsifiability (generates hypothesis that are falsifiable) Does it require any leaps of faith Consistency with know data Internal consistency Utility in generating new hypothesis
What is parsimony
The preference for the least complicated explanation
What is falsifiability
Is it possible to Meade and observation that would show that the theory is wrong (even if that observation hasn’t been made)
Is it clear what would justify rejection of the theory and is it possible to actually test and challenge the basic premises of the theory
What is a hypothesis
A working assumption whose merit is to be evaluated
Ideally based on evidence but could be an educated guess
How is a hypothesis phrased
A statement which may or may not be true relating the independent and dependent variable
What two pairs do hypotheses always come in and what do they mean
Null - no relationship
Alternative - a relationship exists
What would make a test one or two tailed
One tailed - a directional prediction which specifies the nature of a relationship
Two tailed- non-directional prediction which only states there is a relationship
When can it make a one tailed hypothesis
When based on a well-researched theory
When based on previous research showing consistent trends in that direction
what is a variable
how must scientists define variable
anything that varies and can be measured
scientists must provide an operational definition
what is an operational definition
this defines a variable in terms of the procedures used to produce or measure it
Confounding variables
one or more variables which are not under the experimenters control which vary systematically with the independent variable. this means the experimenter cannot claim the independent variable was only at play in causing change in the dependent variable
how to remove the effect of confounding variables
counterbalance
eg if time of day is an issue then run half of each group in the morning and half of each group in the evening
random variable
variables other than the independent variable that effects the dependent variable something that is simply impossible to control
eg participants mood or IQ
how to remove the effect of the random variable
random assignment
what are the two broad avenues of scientific study
quasi-experiments and true experiments
name 4 types of quasi-experiments
observational studies
interviews and questionnaires
case studies
correlation studies
what are the three essential components of a true scientific experiment
one variable is manipulated
the researcher measures whether or not this manipulation results in changes in a second, dependent variable
extraneous variable are controlled so that causal relationships can be established
what is a between subjects-design
each participant is randomly assigned to only one of the conditions
so each participant is only used once
what is a within subjects design
each participant is measured under both the experimental and control conditions
each participant is used more than once
design looks for differences from condition to condition within individuals
what is matched design
type of between-subjects design
each pair is matched on one or more relevant characteristic
this reduces between groups variability
requires considerable effort and resources but exploits the advantages of both within and between subjects designs
what is a population
a complete set of individuals with come common characteristic
what is a sample
a small number of scores selected from the whole set of scores
name two sampling techniques
random
opportunity
what is random sampling
every member of the population has an equal chance of being selected
nice idea but rarely possible
opportunity sampling
a sample based on subjects who are available and willing to participate in the study