Hypothesis:
Making a prediction about a causal relationship
Causal Mechanism
Theory about why the IV caused DV
Factors to consider w/ causal mechanism
1) temporally, IV comes 1st
2) must have reason why IV caused DV
3) endogeniety
endogeniety
relationship will often work in both ways.
“Chicken or egg” problem
Negative correlation/Inverse correlation
as X increases, Y decreases
Positive/Direct Correlation
As X Increases, Y Increases
Pearson’s r
scale between -1 and 1 that measures how strong or weak a relationship between 2 variables is
Data combo for CROSSTABS
IV = NOMINAL/ORDINAL
DV = NOMINAL/ORDINAL
Data combos for MEANS COMP or T-TEST
IV = NOMINAL/ORDINAL
DV = INTERVAL
Data combos for REGRESSION ANALYSIS
IV = INTERVAL/DUMMY
DV = INTERVAL
Regression analysis (beta)
Measure of magnitude of change the IV has on DV
Magnitude
how many X to change Y?
Significance (p-value)
Estimating random error to determine if we can trust our results
What is the gold standard with significance?
P-value = 0.05 or less
Association (R^2)
How much explanatory power does our IV have for change to DV?
Goal of social scientists?
to make observation of a larger population
Population parameter
characteristic of the population of interest.
Sample statistic
Estimate of the population parameter based on the sample drawn from the population.
Considerations when sampling
1) homogeneity of pop.
2) size of pop.
3) resources for research
What to ask yourself when sampling?
Can it be generalized to
1) GREATER POPULATION?
2) SIMILAR POPULATIONS?
3) OTHER SUBGROUPS IN THE GREATER POPULATION?
Describing correlation
1) relationship
2) significance
3) association
Clustered Bar Chart
comparing 2 variables
Simple Bar Chart
Analyzing change across IV for 1 variable
