Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
59 Cards in this Set
- Front
- Back
concept |
abstraction used to describe characteristics of a phenomenon, group, or individual based on a set of criteria or qualities |
|
measure |
a procedure for determining whether or to what degree a concept applies to specific cases based on observation of those cases |
|
yardstick |
specific measure that depends on why you want to measure something (causal logic?) |
|
absolute measure |
measure that uses measurement units and does not include a comparison |
|
relative measure |
measures that consist of a comparison of cases with one another |
|
categorical |
measures that place cases into discrete groups based on whether characteristics are present or absent |
|
continuous |
measures that place cases along a specctrum from more to less (equal distance between consecutive levels) |
|
binay |
categorical variable with just two categories
|
|
multichotomous |
placing attributes, properties, or behaviour into a pre-defined list of more than two categories |
|
ordinal |
categories have a natural order distances between categories is not always the same
|
|
types of categorical measures |
binary and multichotomous |
|
types of multichotomous measures |
nominal and ordinal |
|
nominal |
categories have no intrinsic order |
|
bias |
systematic error produced when our measurement procedure produces scores that are either too high or too low |
|
measurement error |
measurement value differs from true value |
|
upward bias |
overestimation or overstatement by a statistical measure of the event it is attempting to describe |
|
downward bias |
underestimation or understatement by a statistical measure of the event it is attempting to describe |
|
sources of bias |
researcher subjectivity, gaps between concepts and measurements, obstacles to observation |
|
researcher subjectivity |
researcher may apply different yardsticks to different cases taking into factors outside the concept |
|
hawthorne effects |
individuals may behave differently when observed in a research setting |
|
difficulty of observation in the social world |
social norms, strategic action, incentives to hide info, hawthorne effects |
|
experimental blindnesss |
subjects do not know they are being observed (tackles the hawthorne effect) |
|
population |
full set of cases we're interested in learning about |
|
sample |
the subset of the population that we actually take measures of |
|
random sampling |
selecting cases from the population in a manner that gives every case an equal probability of being chosen |
|
law of large numbers |
as a random process is repeated an increasing number of times, the values generated will converge on the true value of the underlying process |
|
validity |
the degree of fit between a measure and the concept it is intended to measure (how well a measure captures the concept) |
|
reliability |
how consistently a measurement procedure produces the same result when the procedure is repeated |
|
threats to validity |
measure does not cover enough of concept, covers outside of the concept, captures different things in different units |
|
threats to reliability |
subject to researcher subjectivity/interpretation, instability over time |
|
causal theory |
a set of general claims about the cause or the effect of a class of phenomenon |
|
hypothesis |
statement on what you should expect to see if your causal claim is true (includes in/dependant variables) |
|
variables |
a measurable property of a phenomenon that can potentially take on different values |
|
independent variable |
cause |
|
dependent variable |
effect/outcome |
|
operationalization |
rationalizing an abstract theory into a concrete variables for hypothesis |
|
unit of analysis |
a description of the type of cases you will study for your analysis |
|
comparative method/ method of difference |
analysis of two cases that are similar in all aspects except take on different x (independent) value to see x's effect on y (if x is present is y?) |
|
correlation |
Arelationship across cases between the values that two variables (X and Y) takeon. |
|
postive correlation |
cases with higher x values have higher y values |
|
negative correlation |
cases with higher x values have lower y values |
|
why correlation may not mean causation (3) |
spurious relationships(z), randomness, reverse causation |
|
third "z" variable |
third variable not taken into consideration that is a cause of C AND E that creates a correlation |
|
spuriousness |
a correlation between twovariables that is not a result of a causal relationship between these twovariables |
|
intervening variable |
third variable that influences x's effect on y (not spurious=step in causal logic) |
|
antecedent variable |
variable comes before the independent variable in the causal logic and influences x |
|
spurious antecedent variable |
variable that comes before x but isindependently related to the dependent variable |
|
How do you control for more than one z variable? |
multiple regression analysis |
|
multiple regression analysis |
examine correlation holding all z's constant |
|
how to test if a correlation is spurious |
random assignment, multiple regression analysis, comparative method |
|
random assignment |
procedure for assigning x to cases that ensures that the difference in the value of the z proven not spurious by law of large numbers affirmation of z to be irrelevant |
|
reverse causation |
correlation between x and y could arise because y is a cause of x |
|
random correlation |
correlation observed purely by chance (patterns are favourable to us) |
|
how to avoid being fooled by random correlation |
statistics: use probability to tell us the likelihood of chance, compute how closely correlated the variables are, take into account amount of cases |
|
when are correlations more likely to be real not random? |
when the correlation is stronger/steeper and holds across more cases (law of large numbers) |
|
statistical significance |
indicator of how likely it is that correlation is random, higher=less likely |
|
p-value |
indicates the probability that the observed correlation was due to chance |
|
process tracing |
method for assessing whether c is a cause of e that moves beyond the logic of correlation based on clues to support a causal logic |
|
social science vs natural science |
social: more abstract, less quantifiable, concepts harder to measure, fewer opportunities for random assignment/variable control bc dealing with human beings and the world |