In: Statistics and Probability
I am doing a research paper in statistics on how college degrees effect quality of life. I came up with the below data. What statistical test would I use to compare the variables? How would I set up the data in SPSS? I am using the null hypothesis that there is no significant change in quality of life with degrees and the alternative hypothesis is that there is significant change in quallife with college degrees.
Level | Total | Poor | Fair | Good | Excellent |
Highschool | 719 | 22 | 128 | 366 | 203 |
College | 477 | 7 | 40 | 187 | 243 |
The data here is on the level of the degree i.e. high school or college and the effect of the same on the quality of life. Since the goal is to check if there is any relation between the two variables, we can perform a goodness of fit chi-square test on this data.
The given data is:
Level | Total | Poor | Fair | Good | Excellent | TOTAL |
Highschool | 719 | 22 | 128 | 366 | 203 | 1438 |
College | 477 | 7 | 40 | 187 | 243 | 954 |
TOTAL | 1196 | 29 | 168 | 553 | 446 | 2392 |
The alternate hypothesis is that there is a relation between the two variables and the null hypothesis is that there is no relation between the two variables.
If there is a relation between the two variables, we can get an expected value for each of the cells by multiplying the total of the row and the total of the column and then dividing this value by the grand total. Doing so, we get an expected table as:
Level | Total | Poor | Fair | Good | Excellent | TOTAL |
Highschool | 719 | 17.4339 | 100.9967 | 332.4473 | 268.1221 | 1438 |
College | 477 | 11.5661 | 67.0033 | 220.5527 | 177.8779 | 954 |
TOTAL | 1196 | 29 | 168 | 553 | 446 | 2392 |
Next we compute the ratio:
Doing so, we get:
Level | Total | Poor | Fair | Good | Excellent |
Highschool | 0 | 1.1959 | 7.2198 | 3.3864 | 15.817 |
College | 0 | 1.8026 | 10.8827 | 5.1044 | 23.8416 |
The chi-square statistic is given by the formula:
Summing the values, we get:
The number of rows in the table is n1 = 2 and the number of columns in the table are n2 = 5. Hence, the degrees of freedom become: df = (n1 - 1) + (n2 - 1) = (2 - 1) + (5 - 1) = 1 + 4 = 5.
Let the significance level be 0.01. Using a chi-square table, we get a critical value of 15.086.
Since the chi-square statistic is greater than the critical value, we can reject the null hypothesis and conclude that there is a significant relation between the two variables.