Introduction to nonparametric tests

Loading...

En provenance du cours de University of Cape Town

Understanding Clinical Research: Behind the Statistics

728 notes

University of Cape Town

Understanding Clinical Research: Behind the Statistics

728 notes

If you’ve ever skipped over`the results section of a medical paper because terms like “confidence interval” or “p-value” go over your head, then you’re in the right place. You may be a clinical practitioner reading research articles to keep up-to-date with developments in your field or a medical student wondering how to approach your own research. Greater confidence in understanding statistical analysis and the results can benefit both working professionals and those undertaking research themselves. If you are simply interested in properly understanding the published literature or if you are embarking on conducting your own research, this course is your first step. It offers an easy entry into interpreting common statistical concepts without getting into nitty-gritty mathematical formulae. To be able to interpret and understand these concepts is the best way to start your journey into the world of clinical literature. That’s where this course comes in - so let’s get started! The course is free to enroll and take. You will be offered the option of purchasing a certificate of completion which you become eligible for, if you successfully complete the course requirements. This can be an excellent way of staying motivated! Financial Aid is also available.

À partir de la leçon

Which test should you use?

The most common statistical test that you might come across in the literature is the t-test. There are, in actual fact, a few t-tests, but the one most are familiar with, is of course, Student’s t-test and its ubiquitous p-value. Not everyone, though, knows that the name Student was actually a pseudonym, used by William Gosset (1876 - 1937). Parametric tests have very strict assumptions that must be met before their use is justified. In this lesson we take a closer look at these tests, but perhaps more importantly, their strict assumptions. Once you know these, you will be able to identify when these tests are used inappropriately.

Introduction to nonparametric tests3:18

Checking for normality5:07

Thinking nonparametrically2:27

Comparing paired observations: Signs2:42

Ordering values: Ranking2:33

Paired comparisons: Sign ranks2:22

Summation of ranks: Rank sums6:04

Comparing two populations: Mann-Whitney-U test4:07

More nonparametric tests5:09

Case study 313:40

Rencontrer les enseignants

Juan H Klopper
Dr
Department of Surgery

In this section, I want to discuss Nonparametric tests.

Now we've seen a lot of tests up until now,

we've seen all the various t-tests for two groups.

We've seen the anova for more than two groups, linear regression, all of these

tests very common in the literature, but they are what we call parametric tests.

And we need to use a parametric test, as all those t-tests,

you need the population parameter to be normally distributed.

Now, think about it this way, for instance.

Imagine I have two groups of patients, group A, group B.

One has got early appendicitis, the other group has got perforated appendicitis.

And I take sample data for white cell count from each of those groups.

Now remember, I've got my sample data for white cell count.

The population in this group refers to everyone out there with

early acute appendicitis.

I have a sample of that population in my parameter,

or my statistic here that we're talking about is my white cell count.

That's my variable.

On the other hand, I have perforated appendicitis.

My sample data white cell count, but there's a population out there,

millions with perforated appendicitis.

Now, what I want to do with my t-test is,

to see if my two sample sets are from different populations.

How likely is is it that those are two different populations.

And I do those analysis from my t-tests for instance.

But that would posit affect that my white cell count data for

the population out there with early appendicitis.

That white cell count has a normal distribution in that population.

And my white cell count in the population out

there with perfect appendicitis has a normal distribution as well.

If that is not so.

If it is not from a normally distributed distribution pattern

if I can use that word, then I cannot use a normal parametric test like a t-test.

Then I have to move on to use a nonparametric t-test.

And I think very often we do find in the literature that the t-test was used and

if we had access to the data,

we might look at the fact that it wasn't that appropriate to use a parametric test.

Now there's one other place to use a nonparametric test of course, and

that is when we're dealing with categorical data,

with the proviso that it is an ordinal type of categorical data.

You've got to be able to order it in some way,

because ordering plays a big role in the use of nonparametric tests.

Remember the ordinal data I have to put it in some order.

If someone chose they agreed or strongly disagreed, there is some order to that.

If someone gave a five star rating versus a three star rating.

If someone chose nine on a pain scale versus a five on the pain scale,

there is some order to that.

I can use, all my nonparametric tests for that kind of analysis as well.

Explorer notre catalogue

Rejoignez-nous gratuitement et obtenez des recommendations, des mises à jour et des offres personnalisées.

Coursera propose un accès universel à la meilleure formation au monde, en partenariat avec des universités et des organisations du plus haut niveau, pour proposer des cours en ligne.

© 2018 Coursera Inc. Tous droits réservés.

Télécharger dans l'App Store

Disponible sur Google Play