Question

anyone know anything about chi square need help medal given

Answer 1

Chi-square tests are nonparametric tests that examine nominal categories as opposed to numerical values. Consider a situation in which you may want to transform numerical scores into categories. Provide a specific example of a situation in which categories are more informative than the actual values.

Take an example of a hypothesis from a parametric test that you worked on earlier in this course, such as an analysis of variance (ANOVA) or a t-test. Explain what changes would be required so that you could analyze the hypothesis using a chi-square test. For instance, rather than looking at test scores as a range from 0 to 100, you could change the variable to low, medium, or high. What advantages and disadvantages do you see in using this approach? Which is the better option for this hypothesis—parametric approach or nonparametric approach?

Answer 2

This sounds like an exam question, but here are a few general rules for dealing with continuous vs categorical data.

1. Sometimes we're more interested in a "cut score" than the actual score. For example, if a student needs to score at least a 70 on a test to pass a class, then you might be more interested in the percent of students passing the class than the actual mean score across all students. This is like transforming a range of 0-100 into a "pass/fail" category.

2. Like all nonparametric tests, the chi^2 requires fewer assumptions about the distribution of your data, so you're less likely to violate some important "rule" of a parametric test.

3. Also, as a rule, when you throw out data you lose precision and statistical power. I say "throw out data" in my example because you're reducing a very nuanced score in the 0-100 range into basically a two category 0 or 1 (fail or pass) data point. A statistical test that compares, say, the pass rate for two classes may not show a statistically significant differences, whereas a parametric test of the means for those two classes would show a statistically significant difference.