Talk:Statistical learning in language acquisition

Comments

I have not looked to the content of the article. I will post some comments on format with examples from other article, since I gave these same comments to other editors, but I see that they also apply here. If a see interest from the editor I will give further comments.

On the other hand the article is really specific in its content, and quite similar to some educational projects assignments. I would be interested in learning if the editor is (or was) included in one of such assingments.

Yes, I originally wrote this article as a requirement for a class.

Was the aim of the class assingment to add content to wikipedia or has this been your initiative?--Garrondo (talk) 07:18, 23 April 2013 (UTC)

On sources

While below I have indicated several secondary problems the main problem at the moment with your article is the use of sources

Almost all sources are WP:primary sources, that is, results of single studies. However, Wikipedia is not an academic paper or essay! Wikipedia articles should be based on reliable, published secondary sources (for instance, journal reviews and professional or advanced academic textbooks) and, to a lesser extent, on tertiary sources (such as undergraduate textbooks). WP:MEDRS describes how to identify reliable sources for medical information, which is a good guideline for many psychology articles as well. So please, reconsider your choice of sources and use secondary sources instead!

The tone is not encyclopedic but highly research like. Wikipedia articles should be based on generalities and conclussions, and not so much in describing specific studies. The article's tone is quite: X did this in the year X and X did that in the year Y. It is precisely the type of writting that derives from use of primary sources. However: who are we to decide that these specific articles are the most important ones, and not a random recollection from inserting in google the term we are searching for? Take a look for example at the schizophrenia article a featured article (one of the best articles in wikipedia): you will not find writting saying X performed an investigation in schizophrenics in 1993 and his colleague X in 2002, since it is not the relevant thing for an encyclopedia (and because there are literally thousands of primary articles on the issue). The important thing is the conclussions that experts have already drawn from this articles and written in secondary sources.

Reason for using reviews is because wiki is an encyclopedia, and use of secondary sources is the best way to remain neutral (See WP:neutral, a core policy) in an article. As students or anonymous editors we are not capable of deciding what is truly important for an article. Secondary sources, such as books or reviews are the sources capable of giving this perspective. To decide which primary sources are relevant is most commonly original research (See WP:NOR, a core policy)

There are a lot of primary sources in this article, but that's because this topic is a relatively new topic in the field and there are not a lot of secondary sources about it. However, all the claims that I mentioned that are cited in the primary sources are reiterated in the secondary sources that do exist, and which I've cited. Other Wikipedia articles on psychology-related topics also cite primary sources. As examples, see mirror neuron and object permanence, the latter of which is labeled as a good article.

It is usually a bad idea to compare articles to other articles to make a point against wikipedia rules or ways, unless you compare it to featured articles (WP:FA), since quality is so very variable even among Good articles.

Regarding the non-existance of secondary sources. I am not an expert nor I have the time to check so I would have to take your word for it. Nevertheless it is quite common among newcomers to understimate the quantity of secondary sources for a topic: Have in mind that technical books, (as well as peer-reviewed reviews) are considered great sources, and I am sure that there will be some out there. The more proportion of secondary sources you use in your article the better.

Specific proposal (best way to improve article without huge efforts): Since you say that all the claims that I mentioned that are cited in the primary sources are reiterated in the secondary sources that do exist, and which I've cited what I would do is to use the already included secondary sources as a reference (in addition to the primary ones) for those sentences that at the moment are only referenced with a primary source but are also covered in these secondary sources. I'm sure that at the moment you almost know all the references by heart and it would not take you much effort to do as I recommend. What do you think?

Bests. --Garrondo (talk) 07:34, 23 April 2013 (UTC)

Some definitions in this regard

• First a definition of primary source in science: they are mainly first-hand experiments and investigations. Authors analize data. Hint: if there is statistics or methods section it is most commonly a primary article

• Second a definition of secondary source in science: they are reviews of many first-hand experiments (also meta-analysis). Authors instead of analizing data "analyze" previous works, ellaborate from the previous results and generalize conclussions. Hint: if there is no results section it is probably a review.

• Third a definition of peer-review: peer review is the proccess of quality assurance in scientific journals by which a work (either primary or secondary) is judged and critized so the author has to make ammendments before publication. It is independent of tha article being primary or secondary.

On format

• There should not be spaces between two sequential references (e.g:semantic errors during reading and the impairment of nonword reading.[2] [3] should become semantic errors during reading and the impairment of nonword reading.[2][3])
• There should not be spaces after the full stop and the referece. (e.g: more precise mechanisms used in normal reading. [16] should be more precise mechanisms used in normal reading.[16])
• Refences should be after the sentence and not before the full stop (e.g forcing reading to proceed through the semantic route [2]. should be changed to forcing reading to proceed through the semantic route.[2]), and same occurs with other punctuation marks such as commas (e.g change be a more severe form of phonological dyslexia[8][9][10] to be a more severe form of phonological dyslexia;[8][9][10])

Fixed.

Regarding refs in the middle of sentence: this is usually avoided in wikipedia per Wikipedia:Citing_sources#How_to_place_an_inline_citation_using_ref_tags. Instead of including them in the middle of a sentence (or at the very beginning in your case) try to move them after the full stop or the next punctuation mark if it only applies to part of a sentence. For example:

In an artificial grammar learning study with adult participants, Saffran, Newport, and Aslin[4] found that participants were able to locate word boundaries based only on transitional probabilities, suggesting that adults are capable of using statistical regularities in a language-learning task

should be

In an artificial grammar learning study with adult participants, Saffran, Newport, and Aslin found that participants were able to locate word boundaries based only on transitional probabilities, suggesting that adults are capable of using statistical regularities in a language-learning task.[4]

if the reference specifically states the last part of the sentence or

In an artificial grammar learning study with adult participants, Saffran, Newport, and Aslin found that participants were able to locate word boundaries based only on transitional probabilities,[4] suggesting that adults are capable of using statistical regularities in a language-learning task.

if it does not.

Bests.--Garrondo (talk) 18:42, 18 April 2013 (UTC)

• Internal links to technical or important concepts are very interesting and the article at the moment is probably underlinked. I would recommend the addition of many more internal links. I would say as a hint that any concept that 14 years old may not know should be linked. On the other hand: concepts should be only linked in the first appearance of the article.

• Automatic citations: You might not know that you can automatically create references from scientific journals by inserting the pubmed number (pmid) or digital object identifier (doi): go to the editing toolbar, click cite, click templates, click cite journal and insert either of the two identifiers in its appropiate place, and voila!: you have your citation. It has the advantage that it reduces errors (although it is better checking since sometimes it makes some mistakes) and also gives a link to the article abstract direcly when you go with the mouse over the inline citation or at the reference at the end of an article. For example in the case of Jones-1985- article Deep dyslexia, imageability, and ease of predication just by inserting its doi (10.1016/0093-934X(85)90094-X), obtained from either pubmed (pmid:3971130 see [1] or the publishing house abstract to the article (See [2]) I get (in this case using doi):. You might notice that gives an error in the year that you can easily fix and have:. I am not sure if I have explained myself adequately but these kind of things make editing much easier, so if you did not really understood my explanation please ask.

All the references now have DOIs.

• The lead should be a summary of the full article. Right now it does not cover the info from all sections. It would be great if you could do a brief summary of these sections to enhance the lead. See: WP:lead

Information from all sections is covered, though briefly, in the lead.

• Citations are not needed in the lead.

See: WP:lead

Both of these issues are fixed.

Regarding the lead: it might be interesting to enhance the third paragraph. At the moment a whole paragraph is used for the initial experiment and a only small paragraph for the remaining of the article. I would give some more details on phonological and syntactical acquisition (maybe in 2 separated paragraphs?). --Garrondo (talk) 07:40, 23 April 2013 (UTC)

• The article appears to be linked in very few articles (actually in none). Maybe you could think of other articles that could point to this article and insert a link in them either in-text (preferred option), in the "see also" section, or at the beginning of specific sections (for this one there is an specific template that has to be used). The more links the article has the more probable it is to attract traffic, which will lead to further improvements in the future.

--Garrondo (talk) 15:16, 16 April 2013 (UTC)

There are links to this article in other articles. See language acquisition and Jenny Saffran for examples. (There are a lot of others as well; those are just two examples.)

As I told you today the sentence saying that there were no links pointing to this article was a mystake, however I have now taken a closer look at the "what links here" and the article is not very well connected: specifically it is linked from the following articles

Language acquisition ;Machine learning ;First language ‎ ;Speech perception;Phonological development;Jenny Saffran ‎;Elissa L. Newport ‎ ;Richard N. Aslin;

Maybe you know other articles which would be sensible to add links towards this page. As I told you this is quite important in order that the article is not lost in a dark corner of the wiki :-).--Garrondo (talk) 14:53, 18 April 2013 (UTC)

Great job with all your fixes. I could only take a quick look today, so I will try to give some further advice if I have time later. Regarding links I think it was a mistake from me (probably I hit the "what links here" in the talk page instead of the main article).Bests.--Garrondo (talk) 07:35, 18 April 2013 (UTC)

I have crossed comments that I feel are fully fixed. I will try to explain the not yet crossed as I have time. Bests.

A (minor) comment on wiki-etiquette: Whenever you insert a comment in the middle of other people comments is very important to sign, along a correct indentation (the latter you have no problems), so anybody can follow the "conversation". In your case I would have signed every time you said "fixed" or any other comment. Bests.--Garrondo (talk) 14:53, 18 April 2013 (UTC)

Too much detail on primary sources

I have eliminated important quantities of detail on the description of primary articles methods. Wikipedia is not the place to detail how every primary article on an issue was performed, but to summarise their conclussions. --Garrondo (talk) 12:20, 19 April 2013 (UTC)

Lead

I have made several tweaks to the lead and asked for clarification of some concepts that should be included in lead. See also comments above on lead. Bests.--Garrondo (talk) 08:01, 23 April 2013 (UTC)

Good work|

I think this is an impressive quality article, especially given how few edits there have been and how few worked on it. Thanks especially User:InnocuousPilcrow for such good writing and User:Garrondo for patient reviewing and tweaking. I hope the educational assignment gave you a really high mark. I'd recommend the phrase "transitional probabilities" in the first sentence be replaced with something more accessible to a lay audience, but I don't know enough about this subject to rephrase it. MartinPoulter (talk) 18:49, 28 June 2013 (UTC)