Introduction
One of the senior managers of Company W is having a complicated time in understanding the null and alternative hypothesis, particularly when it comes to the sale of snack food products. For this reason, more clarification is needed regarding the null and alternative hypothesis testing of snack food products. In order to clarify a hypothesis test to the senior manager a null and alternative hypothesis related to Company W’s new snack food formula will be formulated. Also, a Type I and Type II error will be explained, as well as the ramifications for businesses who commit these types of errors.
Hypothesis Testing
Hypothesis testing is utilizing gathered information to establish the possibility that a particular theory is true. The procedure for testing a hypothesis is usually a four step process; with step one being the formulation of the null hypothesis (H0), and the alternative hypothesis (H1). The next step would be to recognize a test marker that may be exploited on to evaluate the accuracy of the null hypothesis. Step three would be to calculate the p-value, which is the amount of true facts the researcher has against the null hypothesis. The final step in the process of hypothesis testing is to balance the p-value to a satisfactory magnitude (Wolfram Research, 2011).
Type I and Type II Errors
There are two kinds of errors that could be completed by a researcher when testing a hypothesis, which are error Type I and error Type II. A Type I error is declining the null hypothesis when it is actually correct. This kind of error takes place by a researcher when the rejection of a null hypothesis is carried out, but in reality the value really does come from the statistically sampled group, and the null hypothesis actually should have been retained. A Type II error is allowing the null hypothesis when it is really false. This type of error takes place when a researcher retains the null hypothesis when in truth they should have declined it (“Multiple Hypothesis Testing,” n.d.).
Formulating Null and Alternative Hypothesis
When the null and alternative hypotheses are formulated the researcher has to keep in mind that they are nearly opposites of one another. For example, suppose Company W wanted to see if the new formula in their snack food tasted the same as the old one to consumers. A null and alternative hypothesis statement would be formulated, such as
HO: The new snack food formula does not taste the same as the old one.
H1: The new snack food formula does taste the same as the old one.
A level of significance would be selected from these statements, a test marker would be identified, a decision rule would be formulated, and in order for Company W to arrive at a decision on whether the new formula for their snack food tastes the same as the old or not, a sample would need to be taken from their consumers (Wolfram Research, 2011).
Conclusion
A researcher should always check to ensure that the test they are designing is suitable for what they want answers too. When performing hypothesis testing a researcher must keep in mind that one fact about testing a hypothesis is that it might be declined but the hypothesis can never really be completely agreed to until all feasible facts are assessed. A Type I error being committed by a business could be costly by having to pay legal allegations plus brand trustworthiness will be lost, while a Type II error could be costly to a business by losing research funding for investigational studies. For these reasons, a business should always ensure that no errors are committed in the process of performing research, or the research could become way too costly (QuickMBA, 2010).
References
Multiple hypothesis testing and false discovery rates. (n.d.). Retrieved from http://www.stat.berkeley.edu/users/hhuang/STAT141/Lecture-FDR.pdf
QuickMBA. (2010). Marketing research. Retrieved from http://www.quickmba.com/marketing/research/
Wolfram Research. (2011). Hypothesis testing. Retrieved from http://mathworld.wolfram.com/HypothesisTesting.html