reply to post;

Although the video above mentions that "degrees of freedom is a tricky concept", it is possible to understand it to a certain extent. The video defines the degree of freedom as being "whenever we calculate something, like a mean, standard deviation, or test statistic, the degree of freedom is how many independent pieces of information you needed in order to make that calculation." Therefore, we have to keep in mind the whole picture or all aspects of what we are calculating, even though we are not referring to the "sample size". In a coin toss as brought out in the video, there are 2 possible outcomes. Also, since there are only 2 possible outcomes, we know that if it isn't heads it has to be tails, so therefore it will be 1 degree of freedom. That means that only one piece of information is necessary to determine the other outcomes. The number of outcomes needed to know what something is lets us know the number of degrees of freedom. For example, I say my favorite color is one of the three: yellow, green, or purple. In order for someone to know my favorite color, I would have to tell them 2 of the 3 colors that are not my favorite, that would allow them to find out my favorite color. That means that 2 pieces of information are needed in order to determine my favorite color, meaning there are 2 degrees of freedom. In my opinion based off of what I know, the smaller the degrees of freedom are facilitates solving an equation. I like the "One sample T-test, because it is easy to understand the calculation, it is n-1, so it means the sample size minus 1. Once you do that calculation, you can figure out how many degrees of freedom there are. In the case of the One sample t-test, only one piece of data is needed , which is n and that represents the sample size. My question is why are there so many ways to calculate the degrees of freedom, and which is the best one to use.