返回到 Improving your statistical inferences

4.9

星

590 个评分

•

187 条评论

This course aims to help you to draw better statistical inferences from empirical research. First, we will discuss how to correctly interpret p-values, effect sizes, confidence intervals, Bayes Factors, and likelihood ratios, and how these statistics answer different questions you might be interested in. Then, you will learn how to design experiments where the false positive rate is controlled, and how to decide upon the sample size for your study, for example in order to achieve high statistical power. Subsequently, you will learn how to interpret evidence in the scientific literature given widespread publication bias, for example by learning about p-curve analysis. Finally, we will talk about how to do philosophy of science, theory construction, and cumulative science, including how to perform replication studies, why and how to pre-register your experiment, and how to share your results following Open Science principles.
In practical, hands on assignments, you will learn how to simulate t-tests to learn which p-values you can expect, calculate likelihood ratio's and get an introduction the binomial Bayesian statistics, and learn about the positive predictive value which expresses the probability published research findings are true. We will experience the problems with optional stopping and learn how to prevent these problems by using sequential analyses. You will calculate effect sizes, see how confidence intervals work through simulations, and practice doing a-priori power analyses. Finally, you will learn how to examine whether the null hypothesis is true using equivalence testing and Bayesian statistics, and how to pre-register a study, and share your data on the Open Science Framework.
All videos now have Chinese subtitles. More than 30.000 learners have enrolled so far!
If you enjoyed this course, I can recommend following it up with me new course "Improving Your Statistical Questions"...

Jun 29, 2020

Excellent explanations. Strong examples. Helpful exercises. Highly recommended for anyone who ever has to conduct inferential statistics or read anything that reports a p value or bayes factor.

Mar 02, 2017

Excellent course. The lecturer has written code snippets that let the students visualize the meaning and interrelationship of p-values confidence-intervals power effect-size bayesian-inference.

筛选依据：

创建者 Rossella M

•Mar 25, 2020

Really useful and interesting course!

创建者 JOHN Q

•Jun 04, 2017

Interesting Course. Thanks so much!

创建者 Eleonora N

•Jul 17, 2020

Just great. Very insightful course.

创建者 Farid

•Mar 12, 2017

Exactly what i needed. But now it

创建者 Maureen M

•Mar 21, 2019

The best MOOC in statistis ever!

创建者 Mark K

•Jul 10, 2020

This was an exceptional course!

创建者 Pablo B

•Sep 22, 2017

Enjoyable, useful, necessary.

创建者 Oana S

•Dec 27, 2016

Amazing learning experience

创建者 Maheshwar G

•Jun 06, 2020

This is really impactful.

创建者 Zahra A

•Apr 29, 2017

Extremely useful course!

创建者 Biju S

•Dec 05, 2017

Very interesting course

创建者 Alexander P

•Jul 23, 2017

Phenomenal course!

创建者 Maria A T

•Jun 16, 2017

Excellent course.

创建者 martin j k

•Nov 06, 2017

E

x

c

e

l

l

e

n

t

c

o

u

r

s

e

!

创建者 Sarah W

•Feb 13, 2020

Thanks Lakens

创建者 Nareg K

•Nov 30, 2018

Great course!

创建者 Michiel T

•Jul 24, 2018

Great course!

创建者 Jinhao C

•Jun 24, 2018

A must-take!

创建者 EDILSON S S O J

•Apr 09, 2018

Nice!

创建者 Luis A

•Aug 21, 2017

Dr. Lakens is a very good instructor. He speaks cleary and he is extremaly focused in each subject he's teaching, Unfortunatelly, he keeps making some jargons in somehow he understand frequentist statistics. I'll list some of mistakes:

1. The* p*-value is a probability computed assuming *the null hypothesis is true*, that the test statistic would take a value as extreme or more extreme than that actually observed. When he cite "assuming null effect", he merge "effect size" and "NHSTs". This becomes even worst when we use NHST to analyze variable distributions where, by default, we don't have an "effect", but an "assumption". This is valid for all normality test, such anderson-darling or kolgomorov-smirnoff.

2. Furthermore considering the way he decided to approach to null hypothesis, any statistician knows that a null is always wrong and it is the why we dont accept the null. During all the time, in his videos, he insists to use "accepting the null". When he does that, is like a broken guitar in a symphony. It disturbs the video.

3. The control of type II error always involves some sample-size calculations wether we want to acchieve, at minimium, 80% of power. He simply attached a R script to run and he didnt't mention how we can verify if some study has an effect or not. Point and clicking button, in my opinion, is not adequate when we are in a statistical class where the goal is to improve our inferencial skills.

4. Some of quizzes and evaluations have items where options are not presented in a properly way. The subject of each response vary substantly.

I trully hope this feedback will be read in an academic way, which was the intention.

创建者 Alex G

•Oct 26, 2016

To get this out of the way: The one star deduction is not related to the content of the course, only to the fact that there is occasional imprecise language and some parts of the material have typos and grammatical slip-ups that show that the course has room for some tightening up.

That being said, the selection of topics that are covered is great. You get a small but full package of both knowledge and tools that'll help you to significantly (no pun intended) improve your research. Not only are statistical pitfalls covered and solutions offered, you also learn something about how to approach your research with the right mind-set in order to produce solid empirical knowledge that contributes to a cumulative science.

I was particularly impressed by how the instructor manages to pack lots of important topics and concepts into his 10 or 15 minutes lectures without it becoming overwhelming. The key to this is his ability to maintain focus and his generally clear and concise language. The course material, too, reflects the ability to present just the right amount of information - not too little, not too much.

Overall, the course feels very pragmatic and hands-on. It proves that good and fruitful science is doable and that you can start right now. It makes you *want* to start right now.

创建者 Daniel K

•Jan 15, 2019

Thanks to the creators of this course for putting together an engaging curriculum. One note of criticism is that the assignments for Week 5 required G*power software which as far as I can tell is not available on Linux (I'm running Ubuntu).

The practical examples, specifically the example of the impact of Facebook's A/B testing were particularly interesting. I think this course has improved the tools I have at my disposal for interpreting the language commonly used in academic reporting, and I'm confident the information and tools presented will help in my own research in the coming years.

创建者 Alicia S J

•Nov 11, 2018

Good pacing and ratio of exercises/lecture. I found the assignments very useful and the instructions easy to follow. Comparing my performance on the pre-tests and pop quizzes at the beginning of the course to those at the end clearly demonstrates that the coursework honed my stats intuition, and I'm very grateful! The only critical feedback I have is that occasionally, I found the wording of test/quiz questions to be a bit confusing. Thanks!

创建者 Marija A

•Oct 12, 2018

I find this course very useful, since these are topics that do not stick when you are completely new to statics, but are very useful once you have few years experience in practice. My only remark is that sometimes the multiple choice answers in the quizzes were not clear enough, so a bit confusing.

创建者 Robert C P

•Jan 21, 2018

This course is a great complement to other statistics related courses. Instead of spending time on a bunch of formulas, this class is more about best practices and how to (correctly) apply some of the basic statistical methods.