According to the U.S. Department of Labor, unemployment rates have increased to 7.9 percent. The unemployment rate can be defined as the number of people actively looking for a job as to the percentage of the labor force. Although statistics show an increase has occurred, there has also been an increase in higher education jobs. Higher education growth has been steady, stable and greater than overall U.S. jobs. The market share of higher education jobs compared to all U.S. jobs continued to increase and a trend has persisted for several years. Individuals who complete educational programs possess the qualities employers are seeking. These strong qualities include the ...view middle of the document...
The bias in our study is the age factor determining unemployment rates and educational level from 25 years and older.
Testing methodologies will include:
1) Correlation utilizing covariance and the correlation efficient measures to identify relationships between variables.
2) Regression (i.e. test predictability of unemployment rates based on data to variables)
3) Analysis of categorical data (level of education)
4) Prove hypothesis is Type II error
The data consists of two variables; unemployment rate is scaled and referred to as continuous variable, and level of education as categorical or ordinal variable.
In order to analyze our data to prove our hypothesis, we need to know if our data has a normal distribution. Above, the results are illustrated via the K-S test. Our sample size of 520 is stated under the degrees of freedom. The variables indicate they are both not normally distributed. The significant value should be more than .05 if it was normally distributed. In our case, the data has a significant value of .000, which is less than .05 indicating a deviation from normality.
The P-P plot graph is another way to illustrate our data not being normally distributed. They are not nearly symmetrical, but look positively skewed. The data values are deviating away from the diagonal.
Lastly, the histogram shows us it is clearly not normal. There are two peaks indicative of two modes, guilty of bimodal. In a normally distributed data, the mean equals the median equals the mode. This data shows no equality with the mean, median or mode. Therefore, a non-parametric test is performed to measure the relationship between unemployment rates by education level. The significant level of .000 indicates a highly significant relationship.
One non-parametric correlation test is the Kendall’s tau. This test is a better estimate of the correlation in the population. The value of correlation coefficient is closer to zero than the Spearman correlation (it has increased from -.779 to -.635). Though differences in the correlation coefficients exist, the significant value of .000 is still a highly significant relationship.
Another non-parametric test is Spearman’s correlation coefficient. In the table of Spearman’s rho, the significant value for this correlation is less than .05; therefore, there is a significant relationship between unemployment rates and the level of education. In addition, the relationship is showed as a negative: as level of education increases, unemployment rate decreases. As a result, our hypothesis is supported.
Regression analysis involves the identification of the relationship between a dependent variable and one or more independent variables. With regards to the study objective, we sort to statically analyze the relationship between our variables (the level of education and...