T-tests are the most common statistical data analysis procedures used for testing hypothesis in statistics (Fagerland, 2012, pp.12-78). We have several types of t-tests, but the most commonly used t-test is the two sample t-test also called the student’s t-test. The two sample t-test normally tests whether two independent populations have some different mean values. In our case, we’ll compare the real mean weight of a bag of chocolate chip cookies and the mean weight claimed by the industry.
Our research question:
Is there any difference between the real mean weight of a bag of chocolate chip cookies and the mean weight claimed by the industry?
The null hypothesis
The real mean weight of a bag of chocolate chip cookies and the mean weight claimed by the industry are the same.
The alternative hypothesis
There is a difference between the real mean weight of a bag of chocolate chip cookies and the mean weight claimed by the industry.
To answer our research question, we shall first set a significance level of 0.05. We shall then compute the probability value (p-value) based on the real mean weight of a bag of chocolate chip cookies. The p-value represents the chances of obtaining different parameters as compared to those specified in the null hypothesis. We shall finally compare the p-value and the significance level, and if the p-value is lower, we shall reject the hypothesis. T-tests are very helpful in life as they can be used to remove doubts in estimations, like in this case t-test is used to compare the real mean and the claimed mean.
References
Fagerland, M. W. (2012). t-tests, non-parametric tests, and large studies—a paradox of statistical practice? BMC Medical Research Methodology, 12-78.
Jeffrey N. Rouder, P. L. (2009, April). Bayesian t tests for accepting and rejecting the null hypothesis. Psychonomic Bulletin & Review, 16(2), 225–237.
Ruud Wetzels, D. M. (2011). Statistical Evidence in Experimental Psychology. Perspectives on Psychological Science, 6(3).