Term
|
Definition
| : information acquired by means of observation or experimentation |
|
|
Term
|
Definition
|
|
Term
|
Definition
| the type or levels of political actors to which the hypothesis is thought to apply |
|
|
Term
|
Definition
| a body of statements that systemize knowledge and explain phenomenon |
|
|
Term
|
Definition
| derived from theory, makes a specific prediction that can be empirically verified. It predicts that 1+ causal factors will produce a single effect. |
|
|
Term
| components of a hypothesis (4) |
|
Definition
Unit of analysis
Independent variable
Dependent variable
Direction of influence |
|
|
Term
| Your hypothesis should not: (2) |
|
Definition
Give a value judgment
Be immediately verifiable (e.g., be a fact) |
|
|
Term
| Your hypothesis should: (6) |
|
Definition
Be an empirical statement
Be testable
Have an explicit comparison
State a direction of influence
State a unit of analysis
Have clear concepts |
|
|
Term
|
Definition
| explanatory/treatment/control |
|
|
Term
|
Definition
|
|
Term
|
Definition
| the validity of generalized inferences in scientific studies, usually based on experiments as experimental validity—extent to which the results of a study can be generalized to other situations and to other people |
|
|
Term
|
Definition
| the ability to show that variate of the independent variable actually causes the dependent variable to change |
|
|
Term
| 5 criterion for showing causality |
|
Definition
#1: temporal ordering: the IV must occur before the DV
#2: correlation: two variables are correlated when changes in the variable occur together with changes in the other
#3: causal mechanism: you have to be able to tell a plausible story that connects the IV to the DV (often includes an intervening variable” that gets us from IV to the DV
#4: rule out confounds: we can say that X and Y are confounded when there is a third variable that influences both X and Y
#5: correct unit of analysis: ecological fallacy: the use of data at an aggregate level of analysis to draw inferences about individuals |
|
|
Term
|
Definition
| : gets us from the IV to the DV |
|
|
Term
|
Definition
| causes changes in the DV, is correlated w/the IV, and is “casually prior” to that independent variable—chronologically or logically, it comes first |
|
|
Term
|
Definition
| : the IV must occur before the DV |
|
|
Term
|
Definition
| : split cases into 2+ groups of 30+ through a process that’s truly random—using a random # generator is key |
|
|
Term
|
Definition
|
|
Term
|
Definition
|
|
Term
|
Definition
#1: random assignment
#2: pretest: to check how the random assignment process worked, measure the value that the DV takes on for each case before any treatment has been applied. Each group should average about the same values of DV.
#3: apply the treatment
#4: post-test |
|
|
Term
|
Definition
| nature assigns cases to control and treatment groups in some nearly random manner |
|
|
Term
|
Definition
| : researcher assigns cases to control and treatment groups randomly in a real-world setting |
|
|
Term
|
Definition
| : scientists merely observe 2+ groups of cases that have been treated differently; may often be referred to as an “observational study” |
|
|
Term
|
Definition
: refers to the degree to which two measures of constructs that theoretically should be related are related: established if two similar constructs correspond with one another
Ex) participants rate how well they sleep and are also video recorded; convergent validity means that the two scores are similar |
|
|
Term
|
Definition
: indicates that the results obtained by an instrument do not correlate to strongly with measurements of a similar but distinct trait
Important because it is common to come up with an operational definition for a construct that actually measures more than one construct (e.g. trying to measure for anger and accidentally also measuring for depression)
Ex) sleep quality and sleep quantity |
|
|
Term
|
Definition
| : complete recipe for going out into the world and measuring a variable |
|
|
Term
| Cross Sectional Research Design |
|
Definition
| : make comparisons across cases (ex states) |
|
|
Term
| Time Series Research Design |
|
Definition
| : make comparisons over time |
|
|
Term
|
Definition
: true value + bias (validity) + random error (reliability)
The most worrisome measurement issues must be correlated with the independent variable in order to lead to false positives; most measurement problems will bias your research in the direction of not finding a significant result |
|
|
Term
|
Definition
| : degree of correspondence between measures |
|
|
Term
|
Definition
| : the extent to which an experimental test or any measuring procedure yields the same results on repeated trials |
|
|
Term
|
Definition
| : something that happens during the course of the experiment that externally threatens the test |
|
|
Term
|
Definition
| control and experimental group interact |
|
|
Term
| Regression to the Mean Threat: |
|
Definition
| if a variable is extreme on its first measurement, it will tend to be closer to the average on a second measurement and—if it is extreme on a second measurement, I will tend to have been closer to the average on the first measurement |
|
|
Term
|
Definition
| people act differently by dint of being tested |
|
|
Term
|
Definition
|
|
Term
|
Definition
| one chosen by a method involving an unpredictable component |
|
|
Term
|
Definition
| setting other things constant |
|
|
Term
|
Definition
| a graphical representation showing a visual impression of the distribution of data—it’s an estimate of the probability distribution of a continuous variable |
|
|
Term
|
Definition
| : visual representation of chronological data |
|
|
Term
|
Definition
| : bottom and top of the box are always the 25th and 75th percentile (the lower and upper quartiles) and the band near the middle of the box is the 50th percentile; the ends of the whiskers can represent several possible alternative values—upper and lower extremes |
|
|
Term
|
Definition
| : function of both the bin-width and the proportion (proportion/bin-width) |
|
|
Term
|
Definition
| : how often something occurs |
|
|
Term
|
Definition
| : subset of the population which is used to gain information about te whole population |
|
|
Term
|
Definition
| entire group of cases about which you want information |
|
|
Term
|
Definition
| number describing a sample; statistics vary from sample to sample |
|
|
Term
|
Definition
| number describing a population, usually a mystery |
|
|
Term
|
Definition
you can put cases into a category, but can’t specify an order or relationship between the categories (ex. Religion: Catholic, Protestant, Mormon Jewish)
Dummy variables: variables that only take on two values, usu coded as 0 and 1 |
|
|
Term
|
Definition
you can put cases into different categories and order the categories (ex. Strength of religious belief)
Differences between numbers don’t mean anything: values between numbers don’t correspond to actual value |
|
|
Term
|
Definition
| not only can you order the categories of the variable, you can specify the difference (ex. Calendar years and temperature) |
|
|
Term
|
Definition
| interval measure but 0= absence of the measured thing (ex. Income) |
|
|
Term
|
Definition
|
|
Term
|
Definition
| : 50th percentile (if you’re in-between, go for the average) |
|
|
Term
|
Definition
| most commonly occurring number (if there’s two most-commonly occurring numbers, they’re both modes) |
|
|
Term
|
Definition
| how spread out the data is (highest number – lowest number) |
|
|
Term
|
Definition
| from your 25th percentile to your 75th percentile; cuts off outliers (find the median, find the medians of either side of the median, subtract the two from each other) |
|
|
Term
|
Definition
| shows how much variation or dispersion exits from the average (low standard deviation indicates that the data points tend to be very close to the mean) |
|
|
Term
| Measures of Central Tendency: |
|
Definition
| any of a number of ways of specifying a central value—average of set measurements. A measure of central tendency is a measure that tells us where the middle of a bunch of data lies. The three most common measures of central tendency are the mean, median, and the mode. |
|
|
Term
|
Definition
| describing the spread of the data or its variation around a central value--range |
|
|
Term
|
Definition
| hump w/ tail on the right; means that the mean is greater than the median, represents extreme values that are pulling the values to the right |
|
|
Term
|
Definition
| hump w/ tail on the left, means that the median is greater than the mean |
|
|
Term
|
Definition
| how the sample deviates from the population |
|
|
