Three claims, four validities research methods for tge consumers of research Flashcards
empiricists
psychologists, drawing a coclusion about behaviour, emotion and cognition from systematic observations
variable
something that varies
measured variable
a variable that varies but cannot be changed by the researcher (height)
manipulated variable
a variable that can be changed by the researcher (amount of water)
conceptual variable
abstract concepts that are not visiable, also called constructs (intelligence)
conceptual definitions
a conceptual variable that needs to be carefully be defined theoretically.
operational variable/definitions
after a concpetual variable, converting an abstract concept into a varaible that can be measured or manipulated
claim
an argument that someone is trying to make
frequency claim
describes a certain percentage or degree of a single variable ( hom common or frequent something is ) only focusing on one variable and are always measured and not manipulated (example; two out of five americans say they worry everyday)
association claim
representing the relationship between at least two variables. Stating that one level of a variable is likely to be associated with a certain level of another variable. When one changes the other also tends to change (correlated or covariant)
correlation study
measure association between two variables
associations can be represented in a …
scatterplot
relationships on a scatter plot
positive; vary in the same direction
negative; vary in different direction
zero; no relationship
causal claims
a casual relationship between two variables indicated that one variable is responsible for change in the other.
criteria for causality
two variables correlate, relationship cant be zero
must show that the causal variable came first and the outcome variable came later
no other explanations for their relationship, confounders must be excluded
validity
appropriatness of a conlcusion or decision
construct validity
how well the variables in a study are measured or manipulated. (am i really measuring what i want to measure ?)
external validity
the extend to which the results of a study generalise to a larger population or to other times and contexts.
internal validity
the extend to whcih one variable is responsible for changing the subsequent variable (to what extend is the variation in variable X responsible for the variation in variable Y)
statistical validity
the extent to which the data support the conlusion.
Deepening of frequency claims
construct validty; how well a concpetual variable is measured, the variable must be measured reliably (in such a way they produce similar results)
external validity: generalisability
statistical validity: how accurate the statsical results are and how reasonable (how well the figures support the claim)
Deepening of association claims
association claims measure two variables instead of one, to evaluate them , we need to see how well they correlate.
Construct validity: the consutruct validity of each variable must be ensures, if one isnt reliable neither are the results.
external validity: how well the association claim can be generalised to the population
statistical validity: measure two aspects. Measure the strength of the association. When the association is strong, the prediction can be quiet accurate. Measure the statistical significance. if the sample is too small, the results may due to a chance association
margine error of the estimate
(statistical validity in frequency claims) an interval, based on the size of the sample for the study, in which to attempt to include the true value in the population
error in statistcal conlcusion : type one error
when a study incorrectly conlcudes that there is a relationship between two variables, when in fact there is no relationship (false positive)
error in statistical conlcusion: type two error
when a study incorrectly concludes that there in no relationship between two varaibles, when in fact there is a realtionship (false negative)
Deepening causal claims
covariance ; the degree to which two variables are related
time sequence: the cause perecends the effect
internal validity: the third variable criterion: there is no third alternative explanation
random assignment
assigning particiapts to groups are randomly as possible, which isncreasis internal validity
prioritising validty
normally its impossible to carry out a study that fuull fills all four validities simutaneously. Depending on the purpose of the study and the advandtages and disadtvantagesof the method, it will determin which validity has priority for the researcher.
