Chevron Left
Volver a Investigación reproducible

Opiniones y comentarios de aprendices correspondientes a Investigación reproducible por parte de Universidad Johns Hopkins

4.6
estrellas
4,113 calificaciones

Acerca del Curso

Este curso se centra en los conceptos y las herramientas que permiten realizar análisis de datos modernos de forma reproducible. La investigación reproducible se basa en la idea de que los análisis de datos y, en general, las afirmaciones científicas, se publican con sus datos y el código del software para que otros puedan verificar los hallazgos y basarse en ellos. La necesidad de reproducibilidad aumenta drásticamente a medida que los análisis de datos se vuelven más complejos, con conjuntos de datos más grandes y cálculos más sofisticados. La reproducibilidad permite que las personas se centren en el contenido real de un análisis de datos, en lugar de en los detalles superficiales que aparecen en un resumen escrito. Además, la reproducibilidad hace que un análisis sea más útil para otros, ya que los datos y el código que en realidad permitieron llevar a cabo el análisis están disponibles. Este curso se centrará en las herramientas de análisis estadístico alfabetizadas que permiten publicar los análisis de datos en un único documento que permite a otros ejecutar fácilmente el mismo análisis para obtener los mismos resultados....

Principales reseñas

AA

12 de feb. de 2016

My favorite course, at least it gives me an argument why scripted statistics is awesome and can be applied to a number of data related activities. Recycling chunks of code has proven useful to me.

RR

19 de ago. de 2020

A very important course that greatly improved my ability to communicate the findings of any sort of data analysis that I perform. This is a critical skill to acquire to "deliver the means."

Filtrar por:

551 - 575 de 580 revisiones para Investigación reproducible

por Lakshay S

13 de jul. de 2019

Pathetic It was . Not at all Interesting !!

por Raviprakash R S

13 de feb. de 2017

Important topic to cover....Nice course

por Shawn L

4 de abr. de 2018

Thought the final project was fun.

por Paul R

13 de mar. de 2019

Along with the principles of "reproducible research", the primary tool introduced in this course is knitr to produce reproducible research papers and Rpubs for publishing papers. I think this specialization covers RMarkdown 3 different times. Assignments were good, at this stage you start to produce proper papers on an analysis topic which is very much needed before hitting the statistics/regression lectures; however this material can be compressed and needs to be combined with the 9th course which covers Rmarkdown/RStudio again.

por Jeffrey R

11 de jul. de 2018

While I have enjoyed and learned a lot from other courses in the specialization I found this course to contain a lot a repeated information and was overly theoretical. I would have been much more helpful to dig deeper into the capabilities that R markdown offers (i.e building templates, syncing with reference managers, etc.) A a scientist I really appreciate your dedication to getting the word out about reproducibility but at the same time I would have liked to have gained more "how-to" knowledge in this course.

por Ricardo M

13 de dic. de 2017

I believe that this module could be split up to other modules on the Data Science Specialization, instead of having a complete module only for this topic.

Additionally, the process of having the classmates review to get an assignment complete should be changed. In some cases, the assignment is submitted before the deadline, but the lack of reviews by others leads to the assignment extend the deadline. The same happens when we finish an assignment and there are no submissions to be reviewed.

por Jordan B

5 de mar. de 2018

Same information repeated almost all the time ... it looks like the video were made independantly of the course and simply uploaded into Coursera as is .. It is ok in general but in this case, it was really painful to watch. Like video 1 (5min) then video 2 (6 min with 3 as a reminder of the video 1). Reminder are fine across courses or even in different weeks of the same course but not in 2 videos in a row. Otherwise content interesting but could have been explained in way less time.

por Anton K

27 de nov. de 2019

The material is shallow. Projects are way too time demanding. Everybody knows that data cleaning is a routine and long process. That is precisely why nobody likes it. And if there's only one way to clean the data - by hand and only after reading a lot of related database documentation - this kills all the fun of studying and makes the overall picture of consepts relations unclear.

por Jaymes P

27 de oct. de 2020

This course was not well organized. It seemed like lessons were just thrown together and covered much the same content. Some were recorded lectures, others filmed in an office, saying virtually the same things with nearly the same slides but on different weeks. Same old story with the instructor--impossible to listen to because of all the so, um, uhhh, so, kinda, so ummm, etc.

por Gonzalo D

5 de mar. de 2016

The course is poorly organised: There is a project on week one that requires knowledges of week two. Some concepts are dictated more than once because it uses videos made for this course + other recorded from a class room.

I think this course should be a 3 weeks project and the price should be the half of it cost.

Though I enjoyed the second project.

por Gianluca M

18 de oct. de 2016

It is not a bad course, but it is very little informative. There is some nice general discussions about data science by the teacher, there is the explanation of the package knitr, and little else.

As part of the data science specialization it is nice. As a stand-alone course, I would definitely not recommend it.

por Julien N

23 de jul. de 2018

Very disappointed by this course (was used to better by JHU)!

Nothing more than a R Markedown tutorial

Not up-to-date (a full video about an deprecated R package).

A section about evidence based analysis that is hard to understand (and of questionnable interest if not to "fill" this rather empty course)

por Roberto M

19 de nov. de 2016

This course seems 'light' in content - too much time is spent reviewing case studies instead of discussing different ways to create documents that enable reproducible research. Perhaps this should be a topic/chapter in another course, and not a standalone course.

por márvin T O

29 de mar. de 2017

Reproducible research with doubt is important but videos and what it is discuss are not appealing and beyond that, what are worthen are the projects. I did not learn so much from the videos but by myself. Though, the forum is very useful.

por Matt E

1 de may. de 2018

This section could have been completed in a two week schedule instead of four. It is not a terribly complex subject. Statistical inference, however, is. It has a lot of content and could easily go for 5 or 6.

por Jackson L

8 de nov. de 2017

This leaves a lot to be desired. I felt the lectures were fragmentary at best and really lacked in depth analysis. A lot of time was spent on the philosophy of analysis rather than practical tools in R.

por Willie C

2 de feb. de 2020

Lecture videos were very repetitive. Course projects were repetitive, too. Important information, but didn't need to be stretched out over a full "four-week" course.

por Abhimanyu B

17 de ene. de 2017

Provides a very summary overview of a very important aspect of data analysis. Expected more!

por Johnny C

3 de abr. de 2018

The course was interesting, but it is bad many of the videos are recorded lectures.

por Pratik P

2 de feb. de 2017

Sholdnt be a different course. It shold be very very concise. Not this long.

por Victor M

8 de dic. de 2017

Last two weeks do not teach anything new

por Cyriana R

1 de jul. de 2017

ok, but the focus is too much on knitr,

por Sindre F

1 de ago. de 2016

Useful for academics.

por Avolyn F

19 de jun. de 2019

I was really passionate about the subject matter, but, although I have experience in R, apparently not enough to complete the assignment. Would have liked a little more warning that this would be needed, I was more interested in the topic of Reproducible Research, which while I agree is easier done via code of some kind, shouldn't be a topic specific to R, should be applicable to Python, SQL, whatever.

Might have time to revisit this, but will probably need to take a few more R classes to even be able to complete, likely won't get around to it, but the first 2 weeks were worth the cost of paying for a certificate, I guess.

por Owen D

8 de sep. de 2021

This course could've been condensed into one video and incorporated somewhere else in the specialization. Doesn't seem like the instructor even took the course that seriously despite emphasizing its importance in the first lecture. Some of the lectures sound like they were recorded while the instructor was drinking with his colleagues at a dinner party.