27 Cards in this Set
Front | Back |
---|---|
Keys to a good Hypothesis
|
1. A comparison between variables
2. The relationship between variables is clearly specified & measurable
3. Unit of analysis is clear
4. It is testable
|
Levels of analysis
|
Nominal
Ordinal
Interval
|
Reliability
|
extent to which a measurement tool measures a concept consistently
|
Validity
|
extent to which a measurement tool measures the intended concept with accuracy
|
Variable Name vs Variable Value
|
VN = The unit of analysis > Religion
VV = a value of the named variable > Catholic, Protestant, Buddhist, Muslim, etc
|
Nominal Level Variable
|
communicates recorded differences of the between units of of analysis on the characteristic being measured correspond to numeric codes but codes have no inherent meaning
Eg religion
USE MODE
|
Ordinal-Level Variable
|
Values that can be ranked, able to distinguish relative amount, number and value can be assigned to a variable (e.g. 1 being lowest 4 being highest etc)
Eg strongly oppose - strongly agree or never - sometimes - always
USE MODE OR MEDIAN
|
Interval-Level Variable
|
Communicates exact differences between units of analysis
eg age measured in years 18, 24, & 77
USE MEAN MEDIAN OR MODE
|
Mode
|
The most common value of the variable
|
Median
|
the value of the variable that divides the cases right down the middle- with half of the cases having values below & half having values above the median
|
Mean
|
sum of all the cases individual values on a variable & divide by the number of cases
|
Central Tendency
|
the typical or average value
Mean, Median, or Mode
|
Operationalization
|
Describes explicitly how a concept will be measured
Identifies the procedures that will be used
Describes the metric that will be used
|
The measurement process "operationalization"
|
Concept -> precise definition of concept -> measurement strategy -> Variable
|
Systematic error
|
the measurement tool consistently mismeasures what we want it to measure
|
Random error
|
the measurement tool produces errors that are unpredictable and erratic
|
Confounding/Alternative Variable
|
another variable that may affect the value of the dependent variable
|
Quazi-Experiments
|
Research methods that seek to mimic conditions of a controlled experiment
They often exploit a naturally-occurring process
Researcher doesn't control Independent variable
e.g. political revolutions & their causes
|
Quantitative
|
(large N) controlled comparison, cross-tabulation, correlation analysis, linear regression, etc
broad, general claims. Not necessarily accurate in specific cases.
|
Qualitative
|
(small N) comparative method, case studies, process tracing, etc.
highly accurate claims about specific cases. Not generalizable.
|
Observational Studies
|
Observe the values of the independent and dependent variables in a sample of cases (observations)
Analyze the data to draw an inference about the relationship between the variables
Quantitative or Qualitative
|
Forms Of comparison
|
Cross Sectional
Longitudinal
|
Dichotomous variable
|
a variable that contains two categorizes
|
Sufficient condition
|
a factor whose presence guarantees an outcome will occur
If X, then always Y
|
Necessary condition
|
a factor that must be present for an outcome to occur
if no x, then no y
|
Positive Relationship
|
the lower value of the independent variable is associated with a lower value of the dependent variable
|
Negative Relationship
|
The lower value of the independent variable is associated with a higher value of the dependent variable
|