REPLY 7-1 DeCr (100 words and 1 reference)

We have evidence to suggest that they are, we must assume that they are not. This is the motivation behind the hypothesis for the Chi-Square Test of Independence:

With this scenario I would use a .05 level of significance as it is common and provides a 5% risk of determining a difference exists when in actuality there is not one

According to null hypothesis, all flavors should have the same preferences so expect

In the population, we know that 40% of people prefer vanilla, 50% chocolate, 5% strawberry, and 5% mint

The hypotheses are:

·
H0: There is no preference among flavors

·
Ha: There is no preference among flavor

n=80

REPLY 7-2 KiTe (100 words and 1 reference)

Byju (n.d.) tells us that the “Parametric test is a kind of hypothesis test in statistics that gives generalizations for generating records regarding the mean of the primary/original population.” On the other hand, a nonparametric test does not need population distribution, and it is not founded based on an underlying hypothesis despite it being a hypothesis test. The difference in the median is the focus of a nonparametric test, and all test variables are considered nominal and ordinal data. Essentially, parametric tests use statistical distribution in data, while nonparametric tests do not rely on any distribution (Byju, n.d.). Unlike parametric tests, no assumptions are made while measuring the central tendency with the median. Nonparametric tests also include the Mann-Whitney and Kruskal-Wallis components. Because a parametric test needs only a small sample size, it can be generally ruled as the more powerful of the two. An excellent example of a parametric test would be the one in discussion question one (ice cream flavors), as it involves a chi-square test. An instance in which nonparametric testing could be used is to study how the amount of sleep individuals get affects their health levels (propensity to become ill, weakened immune system, aches, pains, etc.).

REPLY 7-2 ChRo (100 words and 1 reference)

A parametric test is a type of hypothesis test that generalizes the data involving the mean of the population data. This is based on the t-statistic that is given beforehand and is used throughout the parametric test. The t-statistic must be known before a parametric test can be conducted. A non-parametric test is a test that does not require a population distribution and is based on differences on the median. Unlike parametric tests there is no need to have the t-statistic before conducting this type of test. The more powerful of the two tests is the non-parametric since it is more flexible and has less requirements to use. It can be utilized in more situations compared to a parametric test making it something that gets more use. An example of a situation that would use need a parametric test is testing the difference between the spread of a virus between two different countries. An example of a non-parametric test is in a scenario where students are given a test and their scores are recorded and then are given a month break and then are retested. Their scores are then compared to what they got the first time to see if there is a difference. The non-parametric test would be used to see if giving that month to prepare made a difference in the scores.