As set forth in the two correlation studies below, there is a statistically significant positive correlation between so-called Blue States and abortion rates. By contrast, there is a statistically significant negative correlation between Red States and abortion rates. Stated plainly, women in more Democratic (liberal) states are far more likely to have abortions than women in more Republican (conservative) states.
A statistically significant correlation is a prerequisite of causation, but not necessarily dispositive of it.
1. There is no correlation between Blue States and abortion rates.
2. There is no correlation between Red States and abortion rates.
1. There is a correlation between Blue States and abortion rates.
2. There is a correlation between Red States and abortion rates.
Statistical Correlation Measure
The statistical correlation measure used herein is the Pearson product-moment correlation coefficient (denoted by r) which measures the linear correlation between two variables X and Y, giving a value between -1 and +1. The closer the r statistic is to +1, the greater the positive correlation between two variables. The closer the r statistic is to -1, the greater the negative correlation between two variables.
For purposes of this study, an r statistic of +1 would indicate a perfect positive correlation between strongly partisan Republican or Democratic states and abortion rates. Alternatively, an r statistic of -1 would indicate a perfect negative correlation between strongly partisan Republican or Democratic states and abortion rates.
An r statistic of 0 indicates no statistical correlation between partisanship and abortion rates.
X variables are the Cook Partisan Voting Index (CPVI) measurements of how strongly a U.S. state leans toward one political party compared to the nation as a whole. Y variables are abortion rates per 1,000 women (aged 15-44) in each U.S. state.
Critical Value Range
-.28 to .28 (49 degrees of freedom at an alpha of .05; two-tailed test)*
*Explanation: In order to reject either null hypothesis and accept the corresponding alternative hypothesis, the r statistic for X and Y must be .28. In such a case, there is at least a 95% probability that the correlation did not simply occur by mere chance. Any r statistic between -.28 and .28 indicates no statistically significant correlation between X and Y.
Blue States r = .78 (p-value=.000000)
Red States r = -.78 (p-value=.000000)
The alternative hypotheses are accepted because the respective r statistics of -.78 and .78 are substantially outside of the critical value range of -.28 to .28.
Women in more Democratic (liberal) states are far more likely to have abortions than women in more Republican (conservative) states.
The p-values for these studies are .000000; therefore, there is less than a 1 in 1,000,000 chance that these correlations occurred circumstantially rather than because of a true relationship.
To read more from this author CLICK HERE.
“Cook Partisan Voting Index by State,” The Cook Political Report
“Abortion Incidence and Services in the United States in 2005,” Guttmacher Institute