Pearson Correlation Test

The present assignment applied Pearson correlation test to evaluate the direction and strength of the association between 25 pairs of data. The determination of the association between the variables provided useful insights regarding the association between different variables and whether the independent variable had a notable impact on the dependent variable. The data was derived from the Centers for Medicaid and Medicare Services (CMS) – a database containing the national healthcare spending data. After considering different information on the website, only one dataset was selected namely the Medicare drug spending for the year 2011 to 2015 (CMS, 2016).

Statistical Analyses

The Pearson correlational method was employed in the present study. The outcomes derived from the statistical tests were presented in the following sections. The data tables and the correlational analyses were depicted in the MS Excel file.

Correlation Test: Pair 1

The correlation test indicated that there was a definite association between the claimant count and total spending (dependent variable) in the year 2011. The correlation was significant because of p < 0.05.

Correlation Test: Pair 2

The dependent variable was cumulative annual spending because it varied depending on the number of beneficiaries. The correlation test affirmed that the correlation between the two variables was positive but non-significant (p > 0.05)

Correlation Test: Pair 3

The results indicated that there was a negative association between the unit count and the average cost per count However the association was not significant.

Correlation Test: Pair 4

The association between cost per unit and beneficiary cost share was positive but not significant.

Correlation Test: Pair 5

The correlation outcomes between the claimant count and total spending indicated that an increase in claims would result in an accompanying increase in the spending (dependent variable). The relationship between the two variables was significant; p < 0.05.

Correlation Test: Pair 6

The correlational tests illustrated that there was a definite association between the beneficiary count in 2012 and the total spending in the same year. Nonetheless, the correlation was insignificant.

Correlation Test: Pair 7

The dataset affirmed that there was a definite association between unit costs and the mean costs per unit. The association was negative. Therefore, an increase in the unit cost was associated with a decline in average price per unit.

Correlation Test: Pair 8

The correlation test indicated that there was a definite association between the mean beneficiary score and total spending. The association was however insignificant.

Correlation Test: Pair 9

The statistical results indicate that there was a positive association between the healthcare claims and total spending and the association was significant.

Correlation Test: Pair 10

The coefficient of determination value illustrated that there was a definite association between beneficiary count and total spending. However, the association was not significant; p > 0.05.

Correlation Test: Pair 11

The outcomes from the Pearson test illustrated that there was a negative association between average cost per unit and the unit costs in 2013. Given that p > 0.05, the association between the dependent and independent variable was insignificant.

Correlation Test: Pair 12

The association between total spending and claim count was positive, and the relationship was significantly based on the p-value.

Correlation Test: Pair 13

An analysis of the relationship between the total spending per user (dependent variable) and the whole count indicated that there was a negative association between beneficiary count and the total count per user and the relationship was not significant.

Conclusion

The Pearson correlation method helped to determine the association between the dependent and independent variables. The outcomes derived from the tests indicated that there were both positive and negative associations between the variables, significant and non-significant p-values. Significance values greater than 0.05 affirmed that the relationship was robust.

References

CMS (2016). 2015 Medicare Drug Spending Data. Retrieved from https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Information-on-Prescription-Drugs/2015MedicareData.html