User talk:Deng9578/sandbox

Latest comment: 7 years ago by MichelleWiest in topic Instructor Review

Instructor Review

edit

This is reading much better! I suggest reading this article out loud to each other to catch grammer mistakes or awkward phrasing. Other suggestions-

Try reading the first paragraph to someone who is not a statistician - perhaps one of the 251 students who comes into the SAC - and see if they understand it.
Be careful about statements like "The null hypothesis is always a part of an experiment." As statisticians, we think this should be the case, but it isn't always. Consdier softening statements like these to account for reality: "The null hypothesis play an important role in designed experiments."
Consider making a graphic illutrating the null and alternative hypotheses. This could simply be two normal distributions with different means, labeled strategically to illustrate the 'null distribution' versus the 'true distribution.' I think the illustration could also help with the references made to the type I and type II error. Take a look at an introductory statistics text book for ideas for the figure.
Third section label shoudl probably "Obtaining" rather than "Obtain."
Fourth section starts with Bayesian inference out of nowhere. Consider what the topic sentence of this paragraph should be.
keep it up!MichelleWiest (talk) 17:04, 7 March 2017 (UTC)Reply

The introduction in first paragraph gives background with the basics of what the null distribution is. This is a strong part of this article since Wikipedia is written for the public who often does not have any knowledge on the topic. I thought that it is important that the authors present how null distribution is used under different contexts. I know that this is just a draft in the sandbox but I would suggest using the same font as other Wikipedia articles. I would also suggest to link words used in the article to other Wikipedia articles. Lukasmar (talk) 01:04, 23 February 2017 (UTC)Reply

Peer Review from Michael

edit

This draft of the article covers useful ground and seems pretty comprehensive. I have just a few suggestions. It would be helpful to get a clear sense of the main idea in the lead paragraph if "null distribution" were defined in the first sentence. Something like: "In hypothesis testing, a null distribution is the distribution of a test statistic associated with the null hypothesis." Here is a URL with a similar definition: http://www.conceptstew.co.uk/pages/s4t_glossary_N.html. The lead might also explicitly mention the topics dealt with in the third and fourth paragraphs. The example presented in the second paragraph might be further developed by giving a concrete instance with specific values. It would also improve clarity to explicitly state the null and alternate hypotheses and describe what the corresponding distributions are. There is a difference in tone and style between the first two paragraphs and the final two paragraphs. This may be because they were written by different authors, so the differences might be evened out if each author edited the others' paragraphs. The last two paragraphs have a lot of densely presented material and might benefit from splitting into additional paragraphs for further development. The overall structure of the article is sensible and the point of view is neutral and does not show any particular bias. Consider making an outline of the topics to discover where adding transitional sentences or subheadings might help keep the structure clear to the reader. There are many statements in the article that need citations. The three citations included so far seem to be from reliable sources, but the citation formats should be double checked, including page numbers for the material you are referencing. Nice job so far! Mw011235 (talk) 05:31, 23 February 2017 (UTC)Reply

Peer-Review from Xinming

edit

Hi guys, I’ve just reviewed your article on Null distribution. Overall, it’s a good article, and I have a better understanding of Null distribution now. But there is still room for you to make it better. Here are the problems that I found:

  • 1, An outline of the article. It will be easier for readers to know clearly what information is given for different paragraphs or different parts of the article if an outline is given.
  • 2, Definition of Null distribution in the first paragraph is a little long. Maybe you can be summarized or paraphrase a little bit.
  • 3, For the 1st and 2nd paragraph, there are not sources listed.

Hope this help.--XinmingLin (talk) 19:51, 23 February 2017 (UTC)Reply