Waching daily Jan 30 2019

Welcome to our first video!

We are the Canadian Society for Epidemiology and Biostatistics at McMaster University.

We are an interdisciplinary, student-run organization that aims to provide an avenue for undergraduate

and graduate students to explore their interests in Epidemiology and Biostatistics.

We get it.

Sometimes basic concepts in statistics can be daunting or a bit confusing.

It is important to have a strong understanding of these before you begin to learn more complex

concepts.

That's why we will have a short series of videos that will help you understand the fundamental

concepts of Epidemiology and Biostats

When thinking about statistics the first thing many people think about is a p-value.

P-values are used across several disciplines.

Well, let's get right into it!

So what is a p-value?

A p-value is a probability value that provides a level of significance within a statistical

hypothesis test that represents the probability of an event occurring.

The p-value provides an indication of how likely the event will occur by chance alone.

In essence, we use a P-value as an approach to hypothesis testing.

We use the calculated probability to determine whether there is evidence to reject the null

hypothesis.

Once you calculate a p-value you can draw conclusions about your study from your calculated

p-value.

A Smaller p-value indicates strong evidence in favour of the alternate hypothesis.

This also suggests your sample.

A low P value suggests that your sample provides enough evidence that you can reject the null

hypothesis for the entire population.

If the p-value is bigger than 0.10, then there is no evidence against the null hypothesis

and the data appears to be consistent with the null hypothesis.

If the p-value is bigger than 0.05 but less than 0.1, then there is weak evidence against

the null hypothesis.

And if the p-value is between 0.01 and 0.05, then there is moderate evidence against the

null hypothesis.

And if the p-value is between 0.001 and 0.01, then we find strong evidence and finally,

if the p-value is less than 0.001, then we find very strong evidence against the null

hypothesis and in favour of the alternative hypothesis.

Here's a quick example using a p-value.

Let's say a vaccine study produced a P value of 0.04.

Now the question might be how you would interpret such a number and whether it is significant.

This P value indicates that if the vaccine had no effect on your health and immunity,

you'd obtain the observed difference of about 4% to be due to random sampling error.

In other words, given that our p-value is less than 0.05 we are mostly certain that

we have enough evidence that the null hypothesis which is, this vaccine not having an effect

on immunity, could be rejected.

A common misunderstanding of p-values is that the calculated probability is suggestive of

the importance or size of the observed effect.

It is imperative researchers only interpret the p-value in terms of rejecting the null

hypothesis.

In some cases, a statistically significant result can be due to the sample size.

If a study has a larger sample size, it may increase the likelihood of statistical significance.

Therefore, If you have a statistically significant p-value that does not necessarily mean the

findings of your research has any clinical significance.

It is common practice for researchers to report findings in terms of confidence intervals.

Confidence intervals are the most current method used to evaluate the significance of

a study.

The confidence level describes the uncertainty associated with a sampling method.

As shown in the example, a confidence interval has a lower limit and an upper limit.

In this example, we have a 95% confidence level that the interval would indeed include

the parameter and 5% that it would be an outlier.

This is important since the confidence interval gives you a bigger picture and a better idea

o whether the study results are significant.

Therefore, you could use this as a way of comparing different studies and their significance.

Thanks for listening to this short video on p-values.

If you have any questions please email us.

Stay tuned for future videos!