Questions:
1. Definition of rare event rule and its interpretation
2. P-value
3. Conclusion that p value tells
4. Author’s conclusion based on the probability
5. Comments on the findings by the author
Answers:
1. Definition of rare event rule and its interpretation
Rare event rule is a type of probability rule that is used when the chance of occurring an event is very rare. Many events occur rarely. Under certain specified assumptions, the probability of that particular event is very low; i.e. extremely rare (Abelson, 2012). This type of event is called rare event rule. If any probability comes out to be a rare event rule, it can be concluded that the assumptions are not correct. An example of this rare event rule is the tossing of a coin is fair as the probability of getting head and tail though is 0.5, but that is a rare event.
2. P-value
P –value is defined as a statistical function of the “test statistic” of a particular statistical model, which measures the degree of extremeness of observations. P value is used in hypothesis testing when the model is assumed true. P value is defined as the probability of obtaining the final value, which is equal to, or extreme to the observed value (Corder & Foreman, 2014).
3. Conclusion that p value tells
P value gives the probability value. This also gives an indication of accepting or rejecting the null hypothesis. If the p-value is less than 0.05 and then the null hypothesis is rejected as there is a strong evidence to support the rejection of null hypothesis. If the p value is greater than 0.05, then the null hypothesis is accepted and there is a strong evidence to support the acceptance of null hypothesis.
4. Author’s conclusion based on the probability
According to the study of de Winter, (2015), it was claimed that statistical significant results are published more than the non-significant results. It was also not clear if this was more prevalent in the “social science” subjects than the physical science subjects. The author researched the abstracts of various papers that were published during the time interval of 1990 and 2013 (Hinton, 2014). The author framed the two hypotheses as “significant difference” versus “no significant difference”. In this case he found no significance difference in his study as the p value was less than 0.05. The author also had another example: to test the set of hypothesis that p< 0.05 is more significant than p> 0.05. In this case, the author found out that p< 0.05 was more significant that the alternative hypothesis which states that both have equal significant. Thus, the author concluded that p < 0.05 was more prevalent during the time period of 1990 to 2013 as the resulted p value of this hypothesis was greater than 0.05
5. Comments on the findings by the author
The findings of the author were “not unusual”. This is because there should not be any difference between the publications of statistically significant results depending on the type of subjects. Publication of results should be irrespective of the type of subject it is (Kanda, 2013). Also, “p< 0.05” was found to be more significant in the second example of hypothesis testing and this is relevant because “p< 0.05” gives a better result than “p> 0.05”.
References
Abelson, R. P. (2012). Statistics as principled argument. Psychology Press.
Corder, G. W., & Foreman, D. I. (2014). Nonparametric statistics: a step-by-step approach. John Wiley & Sons.
de Winter JC, Dodou D. (2015) A surge of p-values between 0.041 and 0.049 in recent decades (but negative results are increasing rapidly too) PeerJ 3:e733
Hinton, P. R. (2014). Statistics explained. Routledge.
Kanda, Y. (2013). Investigation of the freely available easy-to-use software ‘EZR’for medical statistics. Bone marrow transplantation, 48(3), 452-458.
