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;
41 Cards in this Set
- Front
- Back
Symmetric
|
a word to describe correlation of DV and IV- if something is symmetric, it doesn't make a distinction between the IVs and DV
|
|
What is to be considered a “strong” relationship and a “weak” relationship? When is this useful?
|
Strong relationship is when the regression has substantive and statistical significance. A weak relationship is when it doesn't. This is useful when calculating the correlation of a DV and IV.
|
|
Regression
|
A procedure that derives statistics to evaluate the relationship between two interval variables.
|
|
Least-squares line
|
the single best fitting lines that explains the relationship of the variables (where the R squared is the highest)
|
|
Slope
|
Measured by coefficients, the angle at which the line rises or falls, the effect of the IV on the DV and it can tell us if the relationship is positive or negative (highly dependent on the units of mesurement)
|
|
Pearsons's R
|
Measures how closely points are around the regression line (fit)
|
|
R-squared
|
The percent in the variance of the DV explained by the IV (closer to 1 is best)
|
|
F-test
|
How many standard deviations the entire regression is away from the mean (same as T-stat = slope/standard error)
|
|
Describe the principle behind regressions
|
The key idea is that there is a line that best describes the relationship between the IV and DV
|
|
Explain the function of Pearson’s r
|
It tells us the strength of association and is a measure of how good a predictor one variable is of another
|
|
List ways to calculate substantive significance
|
-Coefficients (slope) for each IV
-R squared (fit) for overall regression |
|
List ways to calculate statistical significance
|
- P-Value for each IV
- T-stat for each IV - F-stat for overall regression - Sig. F for overall regression |
|
Coefficients for regression essay
|
Found not in brackets, 1 unit increase in x correlates with a ________ increase in Y
|
|
R squared for regression essay
|
Found at bottom of graph, % variance in DV explained by IV's, above .3 or closer to 1 for substantive
|
|
P-value for regression essay
|
Found at bottom indicated by stars, less than .05 to determine relationship
|
|
T-stat for regression essay
|
Found by dividing coeffient/standard error (found in brakets), ought to be more than absolute value of 2
|
|
F-stat test for regression essay
|
Found by dividing coeffient/standard error (found in brakets) for whole regression, more than 2
|
|
Significant F for regression essay
|
Found by stars, less than .05 to determine relationship of whole regression
|
|
Three standard deviation numbers
|
68, 95, 99.7
|
|
Method problems for regression essay
|
Multicollinearity, Amount of Cases, Endogeneity, Spurious Relationship, Auto-Correlation
|
|
Spurious
|
A relationship where the independent variable really does not affect the dependent.
|
|
Intervening Variable
|
When the independent variable goes in between the other two variables =b where a=b=c
|
|
Beta Weights
|
Also called standardized regression coefficients, they are standardized results of multiple regressen analyses. They show effect of each indpenednt variable on the dependent variable, controlling for all other independent variables, and they use the standard deviation of the variables to remove the effects of the particular units in which the variables are measured.
|
|
Reciprocal
|
In causation, for X to influence Y while Y influences X (causal relationships should only be one direction). Ex. ideology and partiy indentification.
|
|
What is the most important principle to remember when dealing with control variables?
|
not finished
|
|
What are the criteria for inferring causality? Briefly explain each.
|
not finished
|
|
How should you interpret regression coefficients, or b's? Compare this to the interpretation of beta, or beta weights.
|
not finished
|
|
Difference between independent causation, spurious correlation, intervening variable, and complete causation.
|
look at pictures on page 180 of monroe
|
|
Causal Mechanisms
|
The intervening connection between indepedent and dependent variables.
|
|
Process Tracing
|
Using time and sequence to test the causal story. (tracing sequence or history, and chart the events that happen between X and Y.)
|
|
Causal Effect
|
The change in the expecetd value of the outcome variable brought about by a specified change in the vaule of an independent variable.
|
|
Why can a causal mechanism not be directly observed?
|
"They are a component of the of a causal process that intervenes between agetns with causal capacities and outcomes." Or in laymen terms it is because they are like the plus sign between two numbers, without the numbers the plus sign cannot be observed to do anything.
|
|
Which is more important: the causal mechanism or the causal effect? Why?
|
They are equally necessary components of explaining causal theories. This is because they illustrate different but important connections between the independent and dependent variable.
|
|
What are the advantages of process tracing in theory development and testing?
|
Process tracing is the empircal tracing of the causal mechanisms. Theory development and testing: identify paths to an outcome, point out excluded variables, check for spuriousness, and permit causal inference on the basis of a few or even a single case.
|
|
Goodluck on the test
|
I'm sure you'll do awesome...this is kind of a cheesy thing to do, so sue me. -Nathan
|
|
Knock Knock!
Who's there? Wendy. Wendy who? |
Wendy wind blows de cradle will rock (www.knock-knock-joke.com 2006).
|
|
Cross-cutting cleveages
|
Ask Allison about it...
|
|
quantitative research
|
uses numbers and statistical methods, based on numerical measurements of specific aspects of phenomena, seeks general discriptions, seeks measurements and analyses that are easily replicable by other researchers
|
|
qualitative research
|
focuses on one or a small number of cases, use intensive interviews or in depth analysis of history, discursive method, concerned with rounded or comprehensive account of some event.
|
|
falsifiable
|
capable of generating as many observable implications as possible, allows more tests of the theory with more data and a greater variety of data.
|
|
parsimony
|
a simple assumption about the nature of the world
|