POL SCI 3 Chapter Notes - Chapter 4: Interrupted Time Series, Crackdown, Measuring Instrument
Document Summary
The connecticut crackdown on speeding: time series data in. Governor in effort to reduce number of automobile fatalities announced suspension measures for drivers convicted of speeding. Were the results claimed for the program in 1956 valid in the light of both formerly and more recently available statistics on highway fatalities. In a true experiment, treatment is applied randomly and without relation to the prior state of the dependent variable. Potential outcomes of its may reduce the plausibility of certain threats to validity. Likelihood of regression is supported by inspection of data largest change is the upswing just prior to the crackdown. Evidence of general instability of the accidental death rate measure. Look at a poll of adjacent and similar states (ny, nj, ri, ms) to provide meaningful comparison. Control data (aggregate and average the comparison states) are much smoother due to the much larger base (canceling out chance deviations: supports hypothesis that crackdown make a difference.