#HBC | Acrostic
♖♘♗♔♕♗♘♖
- Joined
- Jan 31, 2010
- Messages
- 2,452
Summation: Null hypothesis as a mathematical theorem is inapplicable due to (a) the inability to fulfill the sample size "n" due to testing only for the existence of a monotheistic deity, (b) the fact that the mean is either unpredictable or skewed "X", (c) The fact that the population size "u" again does not fulfill the condition of the hypothesis, (d) The fact that Z cannot be calculated therefore making it impossible to find the cutoff value "c", (e) The fact that an appropriate alpha value cannot be assigned due to inability to fulfill "n" "sigma" and "u" (f) The fact that an appropriate beta value cannot be assigned due to similar reasons as previously stated.Believing in God is similar to believing that invisible unicorns roam the Earth controlling gravity or that George Washington was a furry. There's no evidence to support either proposition that God is real or he isn't. That's the problem with unfalsifiable propositions. The reason it's rational to take the null hypothesis in the face of unfalsifiable hypotheses is so you don't give more weight to one of two equally valid, yet contradictory notions. For example, you can't believe that the uni-color pencil I'm using is blue and red at the same time.
Conclusion: Inappropriate application of a statistical theorem via abusing the inherent logistics of mathematics in order to substantiate your point.
Proof (as far as my tolerance could bear):
The following is sourced from Statistics Principles and Methods 5th Edition written by Richard A. Johnson and Gouri K. Bhattacharyya (Pg. 308-320).
(a) The null hypothesis is used in order to test a hypothesis concerning a population mean "u." In order for the null hypothesis to be used, the sample size that is being polled "n" will be large (n>30 for a rule of thumb).Johnson & Bhatta said:The Steps for Testing Hypothesis [Re: Null Hypothesis]
1. Formula the null hyothesis H0 and the alternative hypothesis H1.
2. Test criterion: State the test statistic and the form of the rejection region.
3. With a specified "α," determine the rejection region.
4. Calculate the test statistic from the data.
5. Draw a conclusion: State whether or not H0 is rejected at the specified "α" and interpret the conclusion in the context of the problem. Also, it is a good statistical practice to calculate the P-value and strengthen the conclusion.
Exodus 20 : 1-3 "And -od spoke all these words: "I am the Lord your -od who brought you out of Egypt, out of the land of slavery. You shall have no other gods before me."
Determining whether or not -od exists isn't suited under the primary conditions of the null hypothesis. You aren't dealing with a population of deities in a given experimental sample but are specifically testing a single monotheistic deity and its possible existence.
(b) Assuming that the research hypothesis that you want to prove is that -od does not exist, also known as the alternative hypothesis or H1... then the null hypothesis that nullifies the research hypothesis would be the null hypothesis or H0 that would stipulate that -od does exist.
The question now comes down to the value criterion. What standard does the null hypothesis use in order to prove its statistical theorem among a specific population?
As stated before, this is the sample mean, X, calculated from the measurements of n = 30+ randomly selected members within the same population (again not possible under given population sample). Based on this sample mean we will outline an example of how to determine the critical value as follows:
Reject H0 [-od exists] if X < c
Retain H0 [-od exists] if X > c
The decision rule that proves that [-od does not exist] is the same as the rejection of H0 which is: X < c. This is known as the critical region. It is important not to confuse X which is a randomly selected sample mean, with "u" which is the population mean.
(c) The possibility arises that a false acceptance of the claim could occur. Assuming that a low probability occurs (again how you would know this value in this given scenario I don't know) alpha is attributed a value of 5% or .05 of "u" which is the population mean.
In order to determine the "c" cutoff we must calculate Z. Z=X-"u"/"sigma"/"sq root of n"
At this point it comes down to the point that I can't assign a numerical value that fulfills the statistical requirements of the null hypothesis. None of the values can be filled in order to possibly prove mathematically that either conditions exist. It is simply foolish to use a test based on a mean population to assess the existence of a divine entity. Honestly. Unless you're talking about a null hypothesis that isn't substantiated by mathematics. Which is silly because calling such an opinion by the same name as an objective statistical test is bigotry.