301 midterm

statement about the possible casual relationship between two or more concepts

theory based statements about a realtionship

what we see if the theory is wrong

how we evaluate hypotheis

thing to be explained, the effect, y

thing that does the explaining, cause, X

1. make thoeries casual
2. dont let data drive theory
3.consider emprical evedince
4.avoid normative statements
5. pursue generalilty and parasiomny

casual
testable
genreral
parasaimounis
new
nonobvious

implying casuaulity, that these tings cause one another

statisical measure of covarration which summarizes the direction and strength of the linear realstionship between two varribales

when two varriables varry together

varriable for which at least some of the varration is theorized to be caused by one or more indepednent varriables

based on real world observation

the act of evualauting the empicrical evedince in order to determine the level of support for the hypotheis verus the null hyptheis

process by which abstract concepts are turned into real world observations

higher values of the independent varriable tend to coincide with lower values of the dependent varriables

scetific research that is carried out under the shared set of assumptions and accepted theories of a paradigm

statements about how the world ought to be

another word for measurement, when a varribale moves from the concept level in a theory to the real world measure for a hypotheis test it has been operationilized

when new fidnings challenge the conventioal wisdom of a paradigm to the point where the set of shared assumptions and accepted thoeries is redefined

simple

higher values of the independent varriable tend to coincide with higher values of dependent varribale

the combination of independent varribales, the dependent varribable, and the casual relationship that are theorized to exist between them

measure for which the time dimension is the same for all cases and the cases represent muliple spatial units

the physical units on which the varriable is measured

the point or points in time at which a varribale is measured

measure for which the spatial dimesion is the same for all cases and the cases represent muliple time units

mathimaticial relationship such that if A is greater than B, and B is greater than C, A must also be bigger than C

involving just two varribales

varriable that is correlated with both the independent and dependent varriables and that somehow alters the realstionship between those two varriables

not what it appears to be, false

1. is there a credible casual mechainism that connects X to Y
2. can we rule out the possiblilty that Y causes X
3. is there covarration between X and Y
4. have we controlled for all confunding varribales, that might make the relationship spurious

reserch desgins that focuses on varration acorss spatial units at a single time unit

the spatial and time units that define a data set

research desgins in which the researcher both controls for and radomly assigns values of the independent varribale to the particants

fake stimlous given to a control group

when the particpants for an expirement are radomly assigned to one of the several possible values

method for selecting indivual cases for a study in which every member of the underlying population has an equal probabbility of being selected

the stragies that a researcher employs to make comparsions with the goal of evaulating causal claims

the physical unit that froms the baisis of obsevration

research desgins that focuses on varration within a single spatial unit over muliple time units

the systematic over-reporting or under-reporting of values for a varribale

the extent to which applying the same measurment rules to the same case or observation will produce identicial results

the degree to which a measure accurately represents the concept that it is supoosed to measure

typical values for a particular varriable at the center of its distrbution

the spread or range of values of a varribale

visual depection of the distbrution of a single varribale that produces two-dimensional figure in which the horizontal displays the varriables and vertical displays the relative frequency of cases

visual depection of the distrbution of a single varribale based on a smoothed calcualtion of the denisity of cases across a range of values

property of the mean value for a single varribale y, which means that the sum of the sqaured diffrences between each y values,yi, and some value for c

average of a varribale equal to the sum of all values of y across indiviauls cases of y

value of the case that sits at the exact center of our cases when arrnaged from lowest to highest

the most frequnelty occuring value of a varribale

varriable for which we can make universally holding ranking distctions acorss the varriable values, but metric does not have equal unit diffrences

statisical measure of indicating the symmtery of the distrbution around the mean

statsicial measure of dispersion of a varribale around its mean

statisical measure of the dispersion of a varribale around its mean

property of the mean value for a single varriable y which means that the sum of the differnce between each y value, yi, and the mean value of y, ybar, is equal to zero

data for every possible revleant case

isolate, account for

directional, (x-y,y-x, both, neither)

what youre comapring, comapring indivauals to indivuals, comparing aggerate quantaties

statisical contorls. (if we cannot dorectly equate groups the way an expiremnt does with radom assignment, in observational studies we work to contorl for other possible causes of y)

at one point in time

1. have a clear goal
2. develop a measure which is both reliable and valid
3. know your data

1. realiablilty
2. validity

catergoical
ordinal
continous

mode
median
mean

zero-sum
least squares

range (high score-low score)
IQR (rnage between 25 and 75)
standard devation