Final Concepts Flashcards
Abstract
succinct summary of research that includes: RQ, methodology/approach, key findings, and implications of the study/contributions. Single paragraph, usually 150-200 words, mitigate stress of having to read and comprehend entire study
Additive Relationships
the control variable is the cause of the DV but defines a small compositional difference across values of IV. Because relationship between IV and Z is weak, IV retains causal relationship with DV after controlling for Z.
Ha and Rival IVs
Ha= mean< Ho value, mean> Ho value, or mean is different than Ho value. rivals: represented by Z, can pose a large or small threat to IV or X
Anomalous Case
deviating from or inconsistent with the other cases being observed, or irregular; abnormal from the hypothesized phenomenon/theory
Authoritative
official (i.e. govt. docs, press releases, publications of organizations), authored by a knowledgeable source (i.e. scholarly articles, primary accounts of history), or supremely confident (i.e. opinion articles)
Case
(George and Bennett’s definition): a case is an instance of one specific phenomena or class of events
Case Study
A case study is the intensive study of a single unit for the purpose of understanding a larger class of (similar) units. We do case studies in order to better understand certain classes of events or types of phenomena, not just to learn about the specific units that we study. Gerring: A case study looking to investigate causality using a single unit will involve comparing that unit’s values on variables either across time, within subunits, or across time and within subunits. Case Study Type I: Variation in a single unit across time
Ex. The case of France in 1788 and case of France in 1789
Case Study Type II: Variation in subunits of a single unit at one time
Ex. The cases of Williamsburg in 1867, Richmond in 1867, Norfolk in 1867, Alexandria in 1867
Case Study Type III: Variation in subunits of a single unit across time
Ex. The cases of Williamsburg in 1867, Williamsburg in 1892, Norfolk in 1867, Norfolk in 1892…
(Professor Brown’s) Case Study Type IV: Counterfactuals
Ex. The case of Europe in 1914 with the assassination of FF and the counterfactual case of Europe in 1914 with
the failed assassination attempt.
Causal Effect
something has happened, or is happening, based on something that has occurred or is occurring. A simple way to remember the meaning of causal effect is: B happened because of A, and the outcome of B is strong or weak depending how much of or how well A worked.
Causal Mechanisms
the processes and intervening variables that link together the cause and effect.
Counterfactuals and Counterfactual Analysis
reverse of proposed relationship (if x then y changes to if not x then not y), counterfactual analysis used often when we don’t have evidence of counterfactual because it hasn’t happened, parallel or alternate worlds in which key features of real world were not present/ were different values, imagine x didn’t happen, when it did, trying to prove that lack of IV leads to diff. DV to prove causal relationship b/t the two
Content Analysis
Content analysis is a research tool used in document analysis to determine the presence of certain words, themes, or concepts within some given qualitative data (i.e. text). Using content analysis, researchers can quantify and analyze the presence, meanings and relationships of such certain words, themes, or concepts. Researchers can then make inferences about the messages within the texts, the writer(s), the audience, and even the culture and time of surrounding the text.
Control Group
subjects who do not receive the given treatment
Control Variables
variables held constant during experimentation
Cotenability Requirement
(counterfactual analysis): that scholars must also consider what else might have to change, or stay the same, in order for their counterfactual antecedent to be true and make sure they all align
Covariation
(how it applies to causal analysis): as 1 variable changes, does the other variable change in some sort of systematic way?
Deductive Reasoning
general to specific
DV
Y, effect
Descriptive Analysis
Descriptive analysis is an important first step for conducting statistical analyses. It gives you an idea of the distribution of your data, helps you detect outliers and typos, and enable you identify associations among variables, thus preparing you for conducting further statistical analyses.
Deterministic Relationships
invariant/ causal relationships are always true
Empirical Implications
(see process-tracing): The next step of process tracing involves looking for evidence of the empirical implications of the causal steps of your theory. Good process-tracing will also involve looking for evidence of alternative theories, you want to find evidence that other potential causal processes for the DV did not occur
Endogeneity
when you think that IV causes DV but actually DV influences IV, i.e. economic inequality and democracy: some argue economic inequality diminishes democracy, however some argue that lack of democracy actually causes economic inequality
Experiment
a scientific procedure undertaken to make a discovery, test a hypothesis, or demonstrate a known fact.
Experimental Group
treated group
External Validity
degree to which results of a study can be generalized outside of the parameters of experiment and or across diff. populations, times, and settings
Field Experiments
subjects studied in natural environment, introduces possibility that other IVs will affect experiment
Framing
act of communicating info in a way that promotes a particular understanding
IV
X, cause
Inductive Reasoning
specific to general
Inference
process of using known facts to learn about unknown facts
Interaction Relationships
relationship b/t IV & DV depends on value of control variable, direction of relationship + strength may change depending on value of control
Internal Validity
degree to which the results represent true cause and effect relationship b/t IV & DV, did the experiment actually isolate the effect of IV on DV from other potential explanations?
Intervening Variables
mediators b/t cause and effect
Journalism v. Scholarship
both should share goal of truth, but non-scholarly audiences may have other goals such as furthering political debate to fuel coverage. Some misreport, others don’t report at all because research might suggest that what’s in the news isn’t actually important. Scholars can aid journalists by providing structural context for daily news, identifying stories worthy of coverage, countering elite spin, identifying historical comparisons, and identifying q’s that scholars seek to answer but cannot.
Lab Experiments
subjects studied in environment created wholly by investigator, great for max control, may not approximate real world conditions, subjects are aware they are being experimented on
Large-N v. Small-n
n=sample size, large sample size, small sample size/ number of cases
Latent Events
In statistics, latent variables, as opposed to observable variables, are variables that are not directly observed but are rather inferred (through a mathematical model) from other variables that are observed (directly measured).
Least-Likely Case
A least likely case, con- versely, is a tough test for a theory because it is a case in which the theory is least likely to hold true.
Method of Difference
(Most similar cases): two similar cases in all relevant ways except for value on IV, if they also differ in DV, conclude IV causes DV
Method of Similarity
Most different cases): causality addressed by comparing two cases that differ in all ways but one: key IV, if cases take on same Dv, conclude that common IV causes DV
Mixed-Methods Research
involves at least 2 diff. types of research methods (qualitative, quantitative, formal modeling)
Most-Likely Case
A most likely case is one that is almost certain to ‹t a theory if the theory is true for any cases at all. The failure of a theory to explain a most likely case greatly undermines our confidence in the theory.
Necessary Conditions
this IV must happen in order to see the DV
No-Variance Designs
where cases under study all have same value on DV or IV, scholars use this design with anomalous cases
Post-Testing
measures subject’s values on DV after treatment, makes potential causal relationships clear and easy to measure
Pre-Testing
measures subject’s values on DV prior to treatment
Primary Sources
original source material on an event, including all evidence contemporary to the event
Probabilistic Relationships
aren’t always true, case studies aren’t good for assessing these relationships, can study cases where relationship doesn’t occur and discount probabilistic as deterministic
Process-Tracing
identify and follow causal mechanisms through analysis of how a causal process plays out in an actual case, ID and find empirical evidence for the steps that lead the IV to cause/influence DV w/in that case, often in conjunction with case studies. good for understanding causal mechanisms (process through which IV interferes with DV), can help identify intervening variables, start with a theory, look for steps between IV and DV, need really well defined theories, usually only used for small-n, can be good for refining theory
Qualitative v. Quantitive Methods
qual: testing causal mechanisms (ideas, thought processes), quant: estimating causal effects in same study (structural variables) (also consider how this relates to case selection)
Scope Conditions
conditions under which a given theory applies or the universe of cases to which a theory applies
Secondary Sources
written about an event after it occurred (i.e. work by historians)
Selecting on DV
selecting cases based on their values on DV rather than IV, biases the study, limits variation
Spurious Relationships
in terms of Z, spurious relationships b/t IV and DV can mean: control defines large compositional diff. across values of IV or dist. of scores for Z is diff. for each value of IV, Z causes IV and Z causes DV, not Z causes IV which causes DV (intervening variable)
Sufficient Conditions
doesn’t have to happen but if it does, its enough to cause the DV
Theory Building
clearly specify the causal mechanisms and causal process
Theory Testing
Once the researcher has theorized the causal process linking the IV(s) and DV, the next task is figuring out what the empirical implications of each step in the process will be, “If my theory is correct, then what outcomes should I see along the way?”
Treatment
condition researcher applies to participants
Triangulation
1) diversification of sources to bring more info, 2) double-checking one source against another to reduce uncertainty, important for establishing causality and describing intent and motivation
Truth Decay
“Where basic facts and well-supported analyses of these facts were once generally accepted, disagreement about even objective facts and well-supported analyses has swelled in recent years.”
elements: blurred line b/t opinion and fact, increased volume and value of opinion, diminished trust in formerly respected institutions
causes: cognitive biases, changes in info system, polarization, competing demands on educational system
Type 1 Error
rejection of a true null hypothesis. Usually a type I error leads to the conclusion that a supposed effect or relationship exists when in fact it does not. Examples of type I errors include a test that shows a patient to have a disease when in fact the patient does not have the disease, a fire alarm going on indicating a fire when in fact there is no fire, or an experiment indicating that a medical treatment should cure a disease when in fact it does not.
Type 2 Error
failure to reject a false null hypothesis. Some examples of type II errors are a blood test failing to detect the disease it was designed to detect, in a patient who really has the disease; a fire breaking out and the fire alarm does not ring; or a clinical trial of a medical treatment failing to show that the treatment works when really it does.
Universe of Cases
All cases that comprise a class of events or phenomena