Descriptive statistics and one-sample t-test summary

The four questions look like arithmetic and only the first one is. The scores given are 95, 92, 90, 90, 83, 83, 83, 74, 60 and 50, and the shape of that distribution is what makes question two answerable. The values cluster in the eighties and nineties with a tail down to fifty, which pulls the mean below the median and gives the set a negative skew. So the honest answer to which method is best for this data set is that the median resists the two low scores while the mean does not, and that the mode is not informative here beyond noting the repetition at 83. So compute all six statistics first, then look carefully at the relationship between the mean and the median before writing anything at all, because that relationship is the actual evidence for your answer rather than a general statement that the median handles outliers better than the mean does.

The one-sample t-test compares the sample mean against a hypothesised population mean of 70, so state the null hypothesis as the population mean equalling 70 and the alternative as it differing, then report the t statistic, the degrees of freedom, which are nine for ten observations, and the p value. Interpreting the result is where marks are lost: a p value close to the conventional threshold demands care, and with a sample of ten the test has low power, so failing to reject is not evidence that the population mean is 70. Say that explicitly. The final question about when the information would be useful should stay concrete and connected to the data, covering things like judging a class against a passing standard, deciding whether a low score is unusual for this group, or comparing cohorts, rather than a general statement about the value of statistics in decision making.