You are here

Group Exercise: Correlations

Select two numeric variables from ANES 2016 and compute the correlation coefficient. Interpret the results, as appropriate.

Select a third numeric variable that you think might mediate the relationship between the first two variables. Compute the partial correlation coefficient. Interpret the results, as appropriate.

What have you learned from the correlation and partial correlation?

Comments

> cor.test(V162099,V162100,method='pearson')

Pearson's product-moment correlation

data: V162099 and V162100
t = 16.496, df = 3574, p-value < 2.2e-16
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.2352650 0.2961865
sample estimates:
cor
0.2659913

There is a moderate/positive relationship in favorability toward poor people and favorability toward big businesses.

SAMIR FATHALLA

> cor.test(V162100,V162109,method='pearson')

Pearson's product-moment correlation

data: V162100 and V162109
t = 14.718, df = 3546, p-value < 2.2e-16
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.2086857 0.2707137
sample estimates:
cor
0.2399446

the favorability of the tea party moderately affects the favorability toward big businesses.

I believe that favorability towards the green party will effect favorability towards big business
> cor.test(V161089,V162100,method='pearson')

Pearson's product-moment correlation

data: V161089 and V162100
t = -2.0477, df = 2529, p-value = 0.04069
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.079520277 -0.001725347
sample estimates:
cor
-0.04068447
we reject the null hypothesis and the favorability towards green party and its association with the favorability towards big business is negative and moderate.

> cor.test(V162099,V162100,method='pearson')

Pearson's product-moment correlation

data: V162099 and V162100
t = 16.496, df = 3574, p-value < 2.2e-16
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.2352650 0.2961865
sample estimates:
cor
0.2659913

There is a positive and moderate relationship in favorability toward Poor People and favorability toward Big Businesses.

Jake, Anna, Brandon

> cor.test(V161086,V162100,method='pearson')

Pearson's product-moment correlation

data: V161086 and V162100
t = -0.86955, df = 3584, p-value = 0.3846
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.04723253 0.01821722
sample estimates:
cor
-0.01452321

- IV - > Democratic Party, Reject the null hypothesis. The correlation test shows a negative and weak relationship between the favorbility between Democratic party association and big business. Sociologically speaking, this could be due to structural racist and sexist issues.

> ANES2016<-read.csv("http://www.shortell.nyc/online/files/anes_timeseries_2016.csv")
> attach(ANES2016)
> cor.test(V162099,V162100,method = 'pearson')

Pearson's product-moment correlation

data: V162099 and V162100
t = 16.496, df = 3574, p-value < 2.2e-16
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.2352650 0.2961865
sample estimates:
cor
0.2659913
there is moderate and positive in favorability toward poor people and big businesses.

cor.test(V161090,V162100,method = 'pearson')

Pearson's product-moment correlation

data: V161090 and V162100
t = -4.6712, df = 2912, p-value = 3.129e-06
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
-0.12216972 -0.05008751
sample estimates:
cor
-0.08624148
I reject the null hypotheses there is no relationship between democrats and big business. Also it correlation is strong and negative.