POL S 15 Lecture Notes - Lecture 15: Level Of Measurement, Statistical Significance, Scatter Plot
Document Summary
Hypothesis: decline in piracy has caused climate change (rise in global temperatures) Testable hypothesis: the lower the number of pirates in a year from 1800 to. 2000, the higher the global average temperature in degrees f. Scatterplot of number of pirates and global temperature, overlaid with regression line (two interval scale variables) There is statistical correlation but that does not equal causation in this case. The mechanism isn"t defined we have to know how the correlation is made possible. Values of variables "just happen" to go together, by coincidence. Don"t know the mechanism, or mechanism is inaccurate. Values of variables coincide because changes in the independent variable cause changes in the dependent variable. Difficult to identify the "causal effect" of one variable ("treatment") on the other variable ("outcome") Identification requires certain assumptions and careful empirical work. Meaningful and statistically significant association/relationship (covariation) exists between the values of the variables (covariation) exists between the values of the variables.