Numbers 1.6

Researchers and data analysts test data points to ascertain statistical significance in order to prove or disprove a hypothesis. T-Tests are a way to evaluate large data sets and determine the significance of the numbers. In 2011, the NAEP recorded the reading scores and categorized them based on student reported race. Classifying the reading scores generates two hypotheses:

H1    There is a statically significant difference between the NAEP mean scores of Black and Hispanic students on the 4th grade reading assessment.

H0: There is no difference between the mean scores.

To prove one of these hypothesis, this requires evaluating the reported scores by way of the T-Test. This can be accomplished through excel, or an online tool. Using an online tool is helpful for time, efficiency, and accuracy. Students should learn to use Excel to do the data analysis for the sake of growing their skill set, and can use the online tools as a resource to verify accuracy. Below are screenshots of the T-Test results from an online tool and Excel:

Raw Data with T-Test of Reading Scores

Online T-Test of Reading Score Data

The data points for the T-Test in Excel match the output from the online tool. This helps verify the accuracy of the work done in Excel.

In this particular data analysis, H0 is proven to be true. There is a 3 point difference between the mean reading score for Hispanic and Black students. Additionally, the P-value for the scoring is .012 further confirming the lack of statistical difference between the two reporting groups.