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;
107 Cards in this Set
- Front
- Back
3 poli sci traditions before empirical
|
formal legal tradition, historical/descriptive form, institutions
|
|
behavioralism
|
political behavior of individuals & groups.
|
|
normative
|
knowledge that is evaluative, value-laden and prescribing what should be. subjective goals & moral rules to guide us in applying what we learned.
|
|
non-normative
|
knowledge not evaluative or prescriptive but factual/objective determinations
|
|
empirical
|
dealing with how/what we know by using common objective language to describe and explain political realities
|
|
empirical research
|
systematic in nature, descriptive and causal inferences about political world, goal of inference, generalizable & public.
|
|
inference
|
tangible observations. descriptive or explanatory inference is based on systematic collection of empirical info
|
|
scientific method
|
look for patterns in human relationships or political world to make educated guesses about future. Intersubjective
|
|
scientific research
|
designed to make descriptive or explanatory inferences. method of testing theories and hypothesis by applying certain rules of analysis to the observation & interpretation of reality under strictly delineated circumstances.
|
|
inference
|
conclusion reached on the basis of evidence & reasoning
|
|
operationalization
|
the conversion or redefinition of relatively abstract theoretical notions into concrete terms that will allow us to measure what we are after
|
|
scientific method
|
scientific research that is explicit, systematic & controlled
|
|
political science
|
process of gathering and interpreting data, generally following standard progression but when new info alters understanding may need return to earlier stage
|
|
quantitative
|
based on statistical comparison of characteristics of various cases/variables being studied
|
|
qualitative
|
based on researcher's informed understanding of cases
|
|
empiricism
|
every knowledge claim based on systematic observation . still involves assumptions, but with most accurate info to guard against bias
|
|
generalizability
|
ability to generalize or extend our conclusions with some confidence from observed behavior of few cases to presumed behavior of an entire population
|
|
concept
|
a universally descriptive word that refers to something directly or indirectly that is observable
|
|
conceptualization
|
allows us to see the particular as something more general
|
|
empirical reference
|
can observe items empirically
|
|
positive
|
what properties concepts hold not what they lack
|
|
theoretical import
|
when a concept is related to enough other concepts in the theory that it plays essential role in the explanation of observed events
|
|
theories
|
sets of logically related symbols that represent what we think happens in the world. Intellectual tools.
|
|
What is a theory? (3)
|
1. ties concepts together by stating relationship between them
2. consists of a set of propositions that are logically related 3. explain political phenomena - common frame of reference |
|
3 Functions of Theories
|
1. Explanation: explain political phenomena by showing how & why they are related to other phenomena
2. Organization of Knowledge: explain phenomena that cannot be explained by existing generalizations 3. Derive New Hypotheses: enable us to predict phenomena beyond those those motivated creation of theory |
|
Evaluating Competing Theories (5)
|
1. simplicity (parsimony)
2. internal consistency 3. testability 4. predictive accuracy 5. generality |
|
propositions
|
posit two major relationships between concepts:
1. covariational 2. causal |
|
inductive model
|
starts with set of observations searches for recurring regularities in way phenomena related to another
|
|
deductive model
|
starts with set of axioms and uses logic to derive propositions about how and why phenomena are related to one another (explanation)
|
|
deductive theory building
|
process moving from abstract statements about general relationships to concrete statements about specific behaviors
Axioms --> proposition --> hypothesis |
|
inductive theory building
|
process of observations about recurring regularities in the way that phenomena are related to one another
observation --> empirical generalization --> hypothesis |
|
What is a variable? (3)
|
1. a concept's empirical counterpart
2. any property that varies 3. empirically observable property that takes on different value |
|
What is a hypothesis?
|
a conjectural (conjoins) statement of the relationships between two variables
A hypothesis is logically implied by a proposition. More specific than proposition and clearer implications for testing. concept-->proposition--> concept variable --> hypothesis --> variable |
|
Variables & hypotheses
|
variables classified according to role play in hypothesis
|
|
Dependent variable (conquest)
Independent variable (antecedent) |
the phenomenon that we want to explain
the factor that is presumed to explain the dependent variable |
|
Formulating hypotheses (4)
|
1. Can be arrived at inductively or deductively
2. State a relationship between two variables 3. Specify how variables are related 4. Carry clear implications for testing |
|
formulating hypothesis: If IV + DV both comparative or quantitative
|
state how the values of the DV will change when IV changes
|
|
formulating hypothesis: If IV + DV categorical
|
state which category of the DV is mostly likely to occur with which category of the IV
|
|
formulation hypothesis: if IV is categorical + DV is comparative/quantitative
|
state which category of the IV will result in more of the DV
|
|
If IV is comparative or quantitative and DV is categorical
|
state which category of the DV is most likely to occur when IV increases
|
|
Common errors in formulating hypotheses (7)
|
1. statement contains only 1 variable
2. statement fails to specify how variables are related 3. hypothesis is incompletely specified 4. hypothesis is improperly specified 5. hypothesis contains values statements 6. hypothesis contains proper names 7. hypothesis is tautology |
|
Why are hypotheses so important? (7)
|
1. provides a bridge between theory and observation through testing
2. are predictions of the form: if A. then B. 3. derive empirical expectations that can be tested against reality 4. direct investigation 5. provide a priori rationale for relationships 6. can be tested, and confirmed or disconfirmed, independently of any normative concerns 7. are useful since they may suggest more fruitful lines for future inquiry |
|
What is testing a hypothesis?
|
showing that the IV and DV vary together (covary) in consistent patterned way. BUT not enough to just demonstrate empirical association, need to look at other variables that might alter or eliminate observed relationships
|
|
What are control variables?
|
variables whose effects are held constant while we examine the relationship between IV and DV
|
|
What is a source of spuriousness?
|
variable that causes both IV and DV. Remove common cause and observed relationship between IV and DV weaken or disappear.
|
|
How to identify a potential spurious relationship? (2)
|
1. Is there any variable might be a cause of both IV and DV?
2. Is there any variable that acts directly on IV and DV? |
|
Definition intervening variables (3)
|
1. mediate relationship between IV and DV
2. provide explanation of why IV affects DV 3. corresponds to assumed causal mechanism |
|
Identifying Intervening variables
|
ask why IV would ahve causal impact on DV
|
|
Conditional variables affect (2)
|
variables conditioning the relationship between IV and DV by affecting:
1. strength of relationship between IV and DV 2. form of relationship between IV and DV |
|
Identifying plausible CV...
|
ask if there is some sort of people for whom IV not have predicted effect on DV...maybe have particular value on DV regardless of value on IV
|
|
three types of variables that condition relationships (3)
|
1. variables that specify relationship in terms of salience - interest, knowledge, concern --> attends church v not attending church for religious affiliation --> view on abortion
2. variables specify relationship in terms of place or time 3. variables that specify relationship in terms of social background or gender |
|
Why is research design so important? (3)
|
1. Allows us to impose controlled restrictions on observation of empirical world.
2. Allows researcher to draw causal inferences with confidence 3. Defines the domain of generalizability of those inferences |
|
3 requirements for demonstrating causality
|
1. demonstrate co-variation (IV and DV vary together in patterned consistent way...if A, then B)
2. eliminate source of spuriousness (rule out any possible SS) 3. establish time order (show change in the IV preceded a change in the DV) |
|
classical experimental design
|
experimental group
control group equivalent except experimental group exposure to IV Pre-test exposure Post-test |
|
3 essential components of classical experimental design to meet requirements for demonstrating causality
|
comparison > covariation
manipulation > time order control > non spuriousness |
|
Internal validity to research design means?
|
internal validity when enables us to infer with reasonable confidence that the IV indeed does have causal influence on DV
|
|
Threats to internal validity: extrinsic
|
usually arise from case selection. Selection bias causes experimental group and control group to differ before experimental group exposed to IV
|
|
Counter extrinsic threats to research design (3)
|
ensuring equivalence
1. Precision matching (case by case matched with someone identical combo of characteristics) 2. Frequency distribution matching (distribution of characteristics within each group matched) 3. randomization (cases assigned in such a way probability equal being assigned to either group) |
|
Intrinsic threats to internal validity arise from (3)
|
1. Changes in cases being studied
2. flaws in measurement 3. reactive effects of being observed |
|
intrinsic threats to internal validity (6)
|
1. History - events may occur while study is under way which affect values on DV independently of exposure to IV
2. Maturation (psychological processes (say as you mature) affect values on DV independent exposure IV) 3. Mortality (selective dropping out may cause groups differ on post-test) 4. Instrumentation (measuring instruments perform inconsistently) 5. Regression effect (atypical pretest scores appear more typical when posttest apart from exposure to IV) 6. Reactivity - "test effect" fact being pretested may cause people's values to change apart from exposure to IV |
|
countering intrinsic threats (6)
|
1. History - both groups exposed
so differences in post test still result of exposure to IV 2. Maturation: both groups mature 3. Mortality: selective dropping out affect both groups equally 4. Instrumentation: Both groups affected by random errors in instrumentation 5. Regression: both groups will regress 6. Reactivity: if pre-test does affect values on post test, both groups react. |
|
Threats to external validity (3)
|
1. unrepresentative cases
2. artificiality of the research setting 3. reactivity |
|
Other research designs
|
1. post test only control group design
2. quasi experimental design |
|
how do you convert a proposition into a testable form? AKA operationalization
|
concept > proposition > concept
variable > hypothesis > variable indicator > working hypothesis > indicator ** it is possible and desireable to represent one variable by several different indicators |
|
What is a research problem? What do you want with it?
|
a properly formulated research problem form of question:
how is concept A related to concept B? Want maximize generality > aim for abstract and comprehensive formulation instead narrow specific one |
|
Why is generality important? (3)
|
1. goal of empirical method is to come up with generalization
2. findings will have implications beyond the particular puzzle motivating research 3. access to diverse theoretical and empirical literature in developing an answer |
|
The research process
|
find a puzzle
formulate research problem develop a hypothesis (conceptualize) identify plausible SS, intervening or conditional variables Choose indicators to represent the IV, DV, and control variables Collect and analyze the data |
|
stages in data analysis
|
test the hypothesis
test for spuriousness if non spurious test for intervening variables test for conditional |
|
what is a measurement? (2)
|
1. the process of assigning numerals to observations according to rules
2. can be qualitative or quantitative |
|
Rules and level of measurement (4)
|
1. determine what kinds of statistical tests can be performed
2. depend on nature of property being measured 3. depend on the choice of data collection procedures 4. provide a basis for classifying, ordering or quantifying |
|
Levels of measurement (4)
|
nominal, ordinal, interval, ratio
|
|
nominal level of measurement
|
lowest level of measurement
classifying variable into two or more categories and sorting out observations into appropriate category. Numerals serve to label categories no hierarchy categories interchangeable but must be exhaustive and mutually exclusive |
|
ordinal level of measurement
|
classifying variable into set of ordered categories then sorting out observations into appropriate categories according to whether they have more or less of property being measured
categories hierarchical relationship, numerals indicate order of categories only one observation more property than another, can't say how much more |
|
interval level of measurement
|
classifying variable in set of ordered categories with equal interval between them
sort observations into categories according to how much of property they possess fixed and known interval between each category so numerals have quantitative meaning can say some have more of property than another and can say by how much more zero point is arbitrary cant say has 2x as much |
|
ratio level of measurement
|
classifying variables in ordered sets sorting observations based on how much of the property they possess.
fixed and known interval between categories = quantitative meaning and can say how much more is one over another. non-arbitrary zero point - zero indicates the absence of the property being measured. Now can say observation has twice as much of property as another |
|
transforming data (2)
|
collapsing and dropping
nominal can collapse and drop ordinal/interval/ratio can collapse but NOT drop |
|
Descriptive statistics
|
used to describe characteristics of a population or a sample
|
|
inferential statistics
|
used to generalize from a sample to the population from which the sample was drawn.
|
|
statistics variates
|
univariate - describe (descriptive), make inferences (inferential) about values of a single variable
bivariate 2 variables multivariate 3+ |
|
descriptive statistics process (3)
|
1. Distribution > how many cases take each value?
2. Central tendency > which is most typical value? 3. Dispersion > how much do the values vary? |
|
frequency distribution
|
a list of the number of observations in each category of the variable. it displays the frequency with which each possible value occurs > called absolute or raw frequencies.
|
|
central tendency
|
measure of central tendency indicates the most typical value > the value that best represents the entire distribution
|
|
measure of dispersion
|
a measure of dispersion tells us how typical value is by indicating extent to which observations are concentrated in a few categories of the variable or spread out among all categories
|
|
measuring central tendency & dispersion nominal data/variation ratio
|
1. mode - most frequently occurring value. Only operation required is counting
2. variation ratio. v= 1- # in modal category / sample size |
|
measuring central tendency & dispersion ordinal data
|
1. median - middle case in distribution. same number of cases above and below it.
2. range > indicates the highest and lowest values taken by cases |
|
measuring inter quartile range
|
range of values taken by middle 50% of cases = inter quartile b/c endpoints are a quartile above and below median value
|
|
Central tendency for interval/ratio-level data
|
1. mean - preferred measure of central tendency because it takes into account the distance (or intervals) between cases
When interval level distribution few cases with extreme values, median should be used instead Mean is subject to distortion, mean value should always be presented with appropriate measure of dispersion |
|
measuring dispersion for interval/ratio level: standard deviation
|
appropriate measure of dispersion at the interval level because takes account of every value and distance between values in determining amount of variability.
Standard deviation will be zero if and only if each case has the same value as the mean. The more cases deviate from the mean, the larger the standard deviation will be. |
|
Measuring dispersion in interval/ratio level
|
z score:
tells us the exact number of standard deviation units any particular case lies above or below the mean. |
|
descriptive statistics: level of measurement, central tendency, measure of dispersion
|
1. nominal: mode, variation ratio
2. ordinal: median, range 3. interval & ratio: mean, standard dev/zscore |
|
2 key functions of concepts
concept formation 1st step to treating phenomena in general class of phenomena |
building blocks of theories
data containers - tools for data gathering |
|
what are concepts? (2)
|
universal descriptive word that refers directly or indirectly to something that is observable
universal descriptive word refers to class of phenomena particular descriptive word refers to particular instance of that class |
|
conceptualization (2)
|
allows us to see partiuclar as example of something more general. involves:
generalization: classifying phenomena according to properties have in common abstraction: represent a class of phenomena by labeling them |
|
nominal definition of a concept
|
*every concept needs nominal and operational definition
describes the properties of the phenomenon that the concept is supposed to represent. provides basic standard against which to judge operational definition. not true or false. |
|
operational definition of a concept
|
identifies specific indicators that will be used to represent concept empirically
|
|
four requirements for a nominal definition of a concept
|
clarity
precision non-circular positive |
|
concepts used to describe political phenomena and can provide basis for (3)
|
1. classification - sorting political phenomena into classes or categories
2. comparison - ordering phenomena according to whether they represent more or less of the property 3. quantification - measuring how much of the property is present |
|
criteria for evaluating concepts (2)
|
empirical import
systematic import |
|
concepts can be linked to observables in 3 ways
|
directly
indirectly via relationship within a theory to concepts that are directly or indirectly observable |
|
what are the hallmarks of the scientific method?
|
empiricism
intersubjectivity explanation determinism |
|
empiricism
|
requires that every knowledge claim be based on systematic observation. use assumptions with most accurate data because obtaining info systematically through our senses helps to guard against bias
|
|
assumption
|
using the most accurate and reliable info about what is happing around us
|
|
intersubjectivity
|
essential safeguard against bias by requiring our knowledge claims be transmissible and replicable
|
|
explanation
|
political phenomenon explained by showing HOW it is related to something else
Empirical research involves search for recurring patterns in why that phenomena are related to one another |
|
determinism
|
recurring regularities in political behavior
can't be proved valid to extend that research produces knowledge claims that withstand empirical testing |
|
nature of scientific knowledge claims (3)
|
never true or false
must be testable/potentially falsifiable impossible to test all possible empirical implications |
|
when using the scientific method we (4)
|
1. make systematic observations and establish criteria of relevance
2. avoid over generalizing 3. avoid selective observation by testing for alternatives 4. address contradictory evidence by making additional observations |