Numbers 1.4 Hypothesis Testing

The sad reality in education is there is a performance gap between ethnicities. The reasons behind this are debatable, as are the solutions, but what cannot be debated are the facts. Comparing three ethnic groups in reading display some of the disparity that is occurring in US public schools. The mean score for white, eight grade students in reading is 18 points higher than black students and 26 points higher than hispanic students. To be certain that the data is statistically significant, we compare the descriptive stats between the three groups and ascertain if the hypothesis is true or false. For the three groups, there are the following hypotheses:

White/Black Test

Ho There is no statistically significant difference between white and black student test scores

H1 There is a difference between white and black student reading scores

White/Hispanic Test

Ho There is no statistically significant difference between white and hispanic student test scores

H1 There is a difference between white and hispanic student reading scores

Black/Hispanic Test

Ho There is no statistically significant difference between hispanic and black student test scores

H1 There is a difference between hispanic and black student reading scores

Comparing the scores for the White/Black test, the mean score white students is 274 compared to 256 for black students. Using a T-Test to compare the scores and determine the statistical significance, the p-value shows we should reject the null hypothesis. The difference between scores is significant.

Carrying on to the second testing group, the average score for hispanic students is 249. The disparity of these two groups is even larger than the first. The p-value score tells us to reject the null hypothesis and believe that the difference in reading scores is significant.

For the comparison of the black and hispanic students, the difference in average test scores is seven points, 256 to 249. Even though the disparity is less than the first two test cases, the difference is still significant.