Descriptive Statistics Essay

2. If I introduce someone this morning to the mean, median, and mode of a set of data, I would be introducing them to descriptive statistics. These types of statistics are used to organize and describe the characteristics of a collection of data. The collection is sometimes called a data set or just data. A fine example of this type of data would be the numbers I calculated in question number one. I can describe each group by their average score, their most often score or the score in the middle of list. Another example (that was actually done in my EDF 517) would be to have the students take an anonymous survey, including major, age, political party, etc. From there, a teacher could better understand what he or she is dealing with in the incoming

Statistics that can used be to generalize a population or multiple sets of data is called inferential statistics. This is another type of statistic not dealing with mean, median, and mode. These types of statistics can be looked upon as much more difficult. However the statistics hold a much stronger baring that allow a researcher to infer trends about a larger population based on a study of a sample taken from it. The first thing that comes to mind is polling during elections. A polling company can survey a couple hundred people, asking about the upcoming election and then make a poll representing the entire state. Some accurate, some not so much. That example is probably the simplest in inferential statistics because there is much more that goes into generalizing sets of data. In order to become accurate in inferential statistics it is important to conduct tests such as a t-test or the chi square to determine if further steps can be taken. Those tests calculate the significance in order to know whether the researcher can actually generalize the results to a larger population. A t-test and other tests in inferential statistics can be seen on the normal curve. Other terms seen in inferential stats is hypothesis, ANOVA (analysis of variance), distribution, regression, correlation, along with Z-testing and critical value. There are different types of t-tests and z-tests that are use depending on the type of sample means, whether dependent or independent. It comes down to

When it comes down to errors in stats, alpha and beta help control the errors. From our notes, alpha means the mistake of accepting a hypothesis when the null should have been controlled statistically, typically at the .05 level. Beta means the mistake of choosing the null when your hypothesis is at work (is remaining %). Alpha and beta are easier to spot in data and formula as they are in Greek lettering. Alpha is used for a Type 1 error while Beta is a Type 2 error. Alpha and beta can range from 0 to 1 where 0 means there is no chance of making a Type 1 or Type 2 error and 1 means it is unavoidable. Per more research on beta, the population regression coefficients in problems and research are denoted by beta. Alpha is not calculated, but decided upon. Researchers can use either alpha or beta but throughout research history, alpha has been the favorite. Another definition for alpha is “Acceptable probability for rejecting the null hypothesis while it is true.” It’s a complicated process but alpha and beta serve as the backbone for error in the hypothesis. In the simplest terms, Type 1 error, alpha, is comparable to false positives. This is thinking you have it right when in reality you don’t. Type 2 in simplest terms, beta, is the same as false negatives. Meaning you may think that your experiment had no effect on the variable, but in reality it did. Alpha is considered a more desirable error than beta because at least with alpha, the attempt will be made. Sometimes in