All 7 entries tagged Web Resources

Open Access?

The first time I had come across the 'open access' publishing/repository concept was during a meeting of the 'Researchers of Tomorrow' cohort at the British Library, some time in 2009. Now (at time of writing), the entire UK system of academic publishing is about to change towards open access as the default publishing option, and it's not exactly a unanimously supported decision. This blog post focuses on the implications of open access for PhD research and is based mostly on my own experience.

The web page for Researchers of Tomorrow reports, as some of the study's key findings, that

Open access [...] appear[s] to be a source of confusion for Generation Y* doctoral students, rather than encouraging innovation and collaborative research.

And that

Access to relevant resources is a major constraint for doctoral students’ progress. Authentication access and licence limitations to subscription-based resources, such as e-journals, are particularly problematic.

*In this case, born between 1982 and 1994.

The majority of us, the study participants, all PhD students, had not really heard of open access until that 2009 meeting. We were not sure what it meant for quality control, peer review, prestige and impact/REF-related aspects of the publishing process. We had the opportunity to discuss open access with publishers, RCUK representatives and other stakeholders, but in my opinion some confusion still remained. There was also an awareness of open access publishing being stigmatised, with some participants reporting that their supervisors would not approve of such a research output over publishing in high-impact non-open access journals.

However, as we met up and discussed our various concerns both formally and informally, what emerged was just how wide the disparity was between journal subscriptions and access to research resources at different universities. People reported not having access to the right journals, or only having access after a certain embargo period, having to wait weeks for items from inter-library loans, paying for a paper based on the abstract and then finding that it was not useful, having to travel to London from all over the UK to use the British Library, and asking colleagues to get articles from sources to which their own institution was not subscribed. This chapter of the Researchers of Tomorrow report discusses these findings in detail.

That was back in 2009. I’m not sure how much attitudes to open access have changed, but there are different concerns being raised and different open access models being proposed: pay-to-publish (‘gold’) and open after an embargo period (‘green’). Among the main concerns about the move to open access publishing is the source of funding that would otherwise come from journal subscription fees. The BSA (British Sociological Association) receives just under half of its income from journals (Network BSA magazine, Spring 2013). But the costs/funding changes associated with the proposed open access policies in the UK and the implications for the kind of research which gets published and by whom are not entirely clear.

However, as PhD fees are broadly similar, and as RCUK studentships do not differ in their value by institution, I think that it is deeply unfair that some students have worse academic resources than others, based purely on their institution’s journal subscriptions. From the PhD perspective, a move towards more good quality research articles being freely available is a good move. What kind of open access publishing model is the ‘best’ kind of model remains to be seen.

Related websites

Warwick Researcher Life Blog: http://blogs.warwick.ac.uk/researcherlife/entry/open_access_information/

Warwick - Open Access: What's in it for you?: http://www2.warwick.ac.uk/services/library/researchexchange/topics/gd0006/

Nature - Pros and Cons of Open Access: http://www.nature.com/nature/focus/accessdebate/34.html

Taylor & Francis Open Access Survey - Initial Findings: http://www.tandf.co.uk/journals/press/OpenAccess-Mar2013.pdf

: 26 Mar 2013 10:25 |  Tags: Open Access PhD Plans Web Resources |  Comments (0) |  Close comments |  Report a problem

A little note on CIs

Writing about web page http://www.scotland.gov.uk/Topics/Statistics/Browse/Health/scottish-health-survey/ConfidenceIntervals

Confidence Intervals are pretty mega, but sometimes it can be difficult to create a graphical representation showing a point estimate (e.g. the mean) and the corresponding confidence interval.

Fortunately, this Scottish Health Survey web page has a nice summary explaining what CIs are, and a further explanation as to how to graph them in Excel.

For CIs in Stata, the following might be useful:

http://www.biostat.jhsph.edu/courses/bio622/misc/graphci_methods_2009_revised.pdf

But you may need to prepare the data in Stata before running the syntax.

: 18 Feb 2013 16:27 |  Tags: Statistics Web Resources |  Comments (0) |  Close comments |  Report a problem

Stata syntax surprise

Here are two bits of syntax that have truly facilitated my life.

When faced with a load of survey variables to clean or recode or regress, the long-winded way is to do them all individually.

e.g.

regress var1 iv1 iv2 iv3

regress var2 iv1 iv2 iv3

regress var3 iv1 iv2 iv3

Needless to say, it stops being fun after about var5.

So the solution is to set up a little macro. Say there are 7 variables we want to regress using the same regression model.

forval x=1/7 {

regress var`x' iv1 iv2 iv3

}

It's so beautiful!

But wait! What if your variable doesn't have a number? What if it's vara varb varc vard etc?

There is another beautiful loop.

foreach x of new `c(alpha)'{

dis "x'"

regress var`x' iv1 iv2 iv3

}

This loop will cycle through the aphabet. If you have variables from a to k, it will stop at k.

And these two little loops have revolutionised my syntax.

They can be used with anything!

My next mission: to find out how they can be combined.

---

Internet credits for the alpha loop: http://www.poliscijobrumors.com/topic.php?id=24913

Other useful stata sites and things:

http://www.cpc.unc.edu/research/tools/data_analysis/statatutorial (love the syntax walkthroughs)

http://www.ats.ucla.edu/stat/stata/ (love the annotated outputs)

Publication-ready tables: try tabout outreg/outreg2 and estout

http://www.ianwatson.com.au/stata/tabout_tutorial.pdf

http://econpapers.repec.org/software/bocbocode/s456416.htm

http://repec.org/bocode/e/estout/

: 08 Feb 2013 20:23 |  Tags: Stata Web Resources |  Comments (0) |  Close comments |  Report a problem

Interview tips

I was just Internetting away and came across this guide for junior researchers interviewing 'elites' (aka VIPs), which has some interesting interview tips. Very useful generally, particularly about the author's own experiences and what he learned from them. Thank you, William S. Harvey.

: 09 Jul 2012 18:40 |  Tags: Web Resources |  Comments (0) |  Close comments |  Report a problem

Online lectures and courses – free!

Writing about web page http://www.openculture.com/freeonlinecourses

When I heard that Harvard and MIT created a joint venture to offer free distance learning courses worldwide I had a quick look on the Internet. The JV is in its early stages: currently there is (was?) just one course in Circuits and Electronics, but others will come.*

Yet, without wanting to rain on the 'new' e-distance learning parade or anything, there are already lots of freely available resources, including full courses, from many top universities, conveniently listed here: http://www.openculture.com/freeonlinecourses

Just saying.

I suppose the lure of the Harvard and MITs edX, and similar partnerships, is their high-profile set-up, offering free courses which will give those who are successful a 'certificate', backed by leading higher education institutions. According to this NYTimes article, the 'certificate' is apparently not convertible to any official (educational?) credit, but I wonder whether this might change.

Although it is clearly highlighted that these free e-courses are not equivalent to a bachelor's degree, could they provide some kind of alternative when higher education is increasingly less affordable? People from low-income countries have responded to this development positively. But another question is whether these courses will remain free of charge, or whether some kind of tiered system will evolve, where the basic material is provided free, but to get some actual educational credit one would have to pay.

But anyway. Learning things is always good, so on the whole it looks like these university endorsed e-courses are a good move. Not sure about using natural-language software to assess essays, particularly humanities essays. Also not sure about crowd-sourcing, or peer-assessment if it is the only way of grading work. But in general, I am interested to see what happens. And if I can, I shall definitely look into some of these online resources myself.

*Although the MITx image shows happy Apple users, the Circuits and Electronics course apparently has only partial non-Flash support, so the girl with the iPad may not be getting to fully experience the course content.

An inexhaustive list of free online course databases

http://www.openculture.com/freeonlinecourses- courses of all kinds for many levels

http://www.onlinecourses.com - HE courses, has ability to create course lists and track progress

http://futurelearn.com- UK thing, yet to offer stuff, watching this space

https://www.coursera.org

https://www.udacity.com

Generally, The Internet.

: 27 Jun 2012 18:24 |  Tags: Higher Education Web Resources |  Comments (0) |  Close comments |  Report a problem

Multiple Correspondence Analysis

Writing about web page http://factominer.free.fr/

I went to a particularly interesting ERSC workshop about a month ago, which sparked off my interest in using multiple correspondence analysis as a way of linking qualitative and quantitative data.

From this brief inrtroduction, I have now started looking into Bourdieu and Bourdieu-type applications of MCA to career development / skills areas of research.

One really good resource I found through a book on MCA by Brigitte Le Roux and Henri Rouanet is this FactoMineR environment for R. The website contains useful walkthroughs and examples how MCA (and PCA and CA) can be used.

I will be updating on what I actually thought of using MCA and in what way it can be useful.

: 27 Feb 2012 13:29 |  Tags: PhD Plans Web Resources |  Comments (0) |  Close comments |  Report a problem

ReStore and ESDS

Just found ReStore, a 'sustainable repository of online research methods resources'. Among many useful things, under Stata resouces, there is an ESDS distributed pdf guide to using Stata to analysing the Labour Force Survey. Very useful.

Another useful guide for LFS analysis is this: http://www.esds.ac.uk/government/resources/analysis/

: 18 Oct 2011 12:26 |  Tags: Lfs Web Resources |  Comments (0) |  Close comments |  Report a problem