Week 3 Topic 1

[Wirichada Pan-ngum]

For our discussion, pick one point from this paper ( pls say the page number or item number or both) and describe the point in your own words. You may be right you may be wrong, so let’s discuss in the group.

[Tanaphum Wichaita]

I choose page 340, point 2 : The P value for the null hypothesis is the probability that chance alone produced the observed association; for example, if the P value for the null hypothesis is 0.08, there is an 8 % probability that chance alone produced the association

many people mistakenly think the P-value shows the chance that “random chance” caused the results. For example, if the P-value is 0.08, they assume there’s an 8% chance the result happened randomly But The P-value only shows how well the data fits the assumptions of the test, including the idea that there’s no real effect. It doesn’t prove whether chance caused the result or not.

[Aye Thinzar Oo]

The point that I will select for the discussion will be Point 6:” A null-hypothesis P value greater than 0.05 means that no effect was observed, or that absence of an effect was shown or demonstrated.” from page 341.
That I understand this A null-hypothesis P value greater than 0.05 indicates that the observed data does not provide strong enough evidence to reject the null hypothesis, but it does not imply that there is no effect or association. Instead, it shows that there are many hypotheses, including the null, that could fit the data, and therefore we cannot conclude that there is “no association” or “no evidence” of an effect. So we need to look at the point estimate for the effect size to understand what the data suggests.

[Cing Sian Dal]

I will pick the first misinterpretation: “The P value is the probability that the test hypothesis is true” (Greenland et al., 2016, p. 340). I will elaborate on it with an example of the coin example.

The coin is fair either P(H)=0.5 or P(T)=0.5 which is H0 (null hypothesis)

But if you believe that the coin is biased towards the head, getting more heads, then P(H) ≥ 0.5 which is H1 (alternative hypothesis)

Now, you experimented 100 times and got (observed) 60 heads of 100 times.

You calculate the p-value and get ~ 0.028 (2.8%).

Here, you cannot misinterpret 2.8% as the probability that the coin is fair or the probability of getting more than 60 heads.

Instead, the p-value indicates that there is a 2.8% chance of seeing 60+ heads by random chance under the assumption that the coin is fair.

Because the p-value will randomly vary with each experiment and each flip under the assumption that the coin is fair.

Reference:

Greenland, S., Senn, S. J., Rothman, K. J., Carlin, J. B., Poole, C., Goodman, S. N., & Altman, D. G. (2016). Statistical tests, P values, confidence intervals, and power: a guide to misinterpretations. *European Journal of Epidemiology*, *31*(4), 337–350. https://doi.org/10.1007/s10654-016-0149-3

I might be wrong. Please feel free to correct me.

[Siriluk Dungdawadueng]

I would like to give my view on point 4 from page 341: “A nonsignificant test result (P > 0.05) means that the test hypothesis is true or should be accepted.”

This point addresses the misconception that if a P value is greater than 0.05, it means the hypothesis being tested is true and should be accepted. However, a nonsignificant P value only suggests that the data are not unusual under the test hypothesis. It doesn’t prove the hypothesis is true. There could be other reasons for a large P value, such as a small sample size or other assumptions being incorrect. Therefore, it’s important to consider the context and other factors before concluding that the hypothesis is true.

[Author]

Posts