# Discussion session for quantitative/ mixed methods

This is from a handout I developed for a discussion session on quantitative/ mixed methods.

Quantitative / qualitative approaches differences: Very crude representation of commonly thought differences. From Firestone (1987).

 Quantitative Qualitative World view Object facts independent of social world Multiple realities socially defined. Purpose Explain through measurement Understanding through actors' perspectives Approach Experiment: reduce randomness/ noise/ error which get in the way of explanation. Ethnography. Explore definitions. Challenge assumptions Researcher role Detached Immersed Writing Objective: facts speak for themselves Interpretative Conclusions Explain reasons for phenomenon/ change in quantitative terms e.g. x% of variance explained by x,y,z, Qualitative judgement e.g. 'strong leadership required...'

Discussion questions:

Does choice of method stem from our world view or does our world view stem from our choice of methods?

Is the primacy of qualitative methods in education studies a consequence of expediency/ ability or world view?

Does early specialisation in UK education mean social science and humanities reject quantitative approaches due to lack of sufficient mathematics skills?

To what extent is the table above an accurate representation of differences between quantitative and qualitative approaches?

British Academy. 2015. ‘Count Us in: Quantitative Skills for a New Generation’. London: British Academy. .

Firestone, William A. 1987. ‘Meaning in Method: The Rhetoric of Quantitative and Qualitative Research’. Educational Researcher 16 (7): 16–21. doi:10.3102/0013189X016007016.

# Questionnaires in LaTeX using the paperandpencil package

I've been using the paperandpencil (.pdf document) package for creating questionnaires in LaTeX. 1 I thought the package worked very well, but just a couple of notes to myself which others may find useful. I've found it works well, but a couple of things were unclear. The .pdf version of the manual displays but the homepage linked from other sites no longer appears to work. However I have just found a download page on the QDDS website.

1. The file paperandpencil.sty needs to place in the same directory as the .tex file. I had trouble finding it, but found the code over on Github. I pasted this into a text editor, saved it as paperandpencil.sty, and put it in the same directory as my .tex file.
2. The document class needs to be {scrreprt}. This is in the first page documentation, but I managed to miss it.
3. To set page numbers \pagestyle{plain} is required and the \pagenumbering and \setcounter options need to be set.

A minimal code example here:

\documentclass[a4paper, 11pt]{scrreprt}
\usepackage{paperandpencil}
\usepackage[top=2.5cm,bottom=2.5cm,left=2cm,right=2cm]{geometry}
\pagestyle{plain}
\begin{document}
\pagenumbering{arabic}
\setcounter{page}{1}
\section*{Title goes here}
\end{document}


Notes:

1. Produced by the Questionnaire Development Documentation System, based at the University of Duisburg-Essen

# Statistics for Humanities now in PDF format plus some reflections

The Statistics for Humanities book is now available in a much nicer PDF format. The mediawiki built website is still up there though it doesn't look great. The PDF is not perfect by any means but it’s time to move onto other projects now.

The book is published under a non-commercial Sharealike Creative Commons licence. I am thinking of trying an e-book version (I have software which technically supports this)  but I fear that with all the images and mathematics this could end up being a big mess.

The first versions of the book (previously available for public consultation and rejected by the British Academy on the advice of their reviewers) were written in LaTeX. I actually learnt LaTeX for the specific purpose of producing the book. I liked the idea of programmatic control but it all got a bit difficult to control where graphs and images landed. GNU plot which integrates with LaTeX was great for making the graphs and I'll continue to use it in the future if and when the need arises.

The first website version has been written using mediawiki software (the same software behind wikipedia). Wiki markup is easy to learn and plugins mean that LaTeX can be integrated for the mathematical parts. One of the issues with websites is that appearance in browsers can vary, but it is still not as nice looking as wikipedia.  The great advantage of wikis of course is that they are social tools in which people can collaborate to produce a finished product. I floated the idea with a couple of others of making this an open wiki, but was advised I could face problems of vandalism and spam. Although I like running websites it’s not a big part of my 'day job' and I don't want a troublesome hobby.

As I mentioned before the PDF version is available under a creative commons licence which allows people to modify it. This is all very well, but modifying a pdf document is not easy (that is sometimes the actual point), so I do need to think about whether I want to make the source files available. The book was produced using a combination of Serif Page Plus X7 (the full version) and MathType V6.9.

PagePlusX7 (I paid just under £64 including VAT) is not unlike MS Publisher though I much prefer it. I experimented with other desktop publishing software such as Scribus (free/ open source)*, but felt the learning curve was too steep at the present time. (High level DTPs run into hundreds of pounds so I'm not even going there).

Mathtype 6.9 (£43.20) was a worthwhile investment. I used it to type equations in LaTeX then copy and paste into PagePlus, but it can also be used to 'build' equations like the MS Equation Editor and even has a 'handwriting' feature. It can also be used in hundreds of other software applications. This was not totally plain sailing and for some reason equations using square root signs look a bit strange when posted into PagePlus. (If future editions of PagePlus have LaTeX Math integration that would be delightful).

Despite all the software used to produce the book, no software is necessary to use the book!

*Scribus does have LaTeX integration.

# My Open Educational Resources

I have put many of my teaching resources in the humbox where anyone is welcome to download and reuse them. They are primarily used as handouts in face-to-face contexts, so most would probably need adapting to turn them into materials suitable for purely online context. Most are available under creative commons licenses, but please check first.

My profile on the humbox

Click on "View all resources" to see a full list of my resources.

The PhD research training collection includes resources on the PhD viva, academic writing, employability, academic writing, doing book reviews and ethnography.

The Head of University language departments collection includes scenario planning exercises, for curriculum change, people management and presenting research.

Further resources I have produced for sharing are on other project websites such as SPEAQ (Relating to quality assurance and enhancement) and Getting the Most Out of Feedback which has resources for students and staff on feedback and student evaluation.

There is also my Statistics for Humanities online book.

I have made a list on Diigo of other people's open educational resources on study skills

# British Academy launches leaftet for students on value of quantitative skills

The British Academy has published a booklet of case studies from humanities and social science graduates who use quantitative skills in their everyday work. On the subject of statistics a draft of my Statistics for the humanities book is now with the British Academy for review. I hope to have more news soon.

# Five (possible) barriers to quantitative methods in the social sciences

The news that the Nuffield foundation, the ESRC and HEFCE are to invest £15.5m in centres to develop quantitative skills in the social sciences is very welcome. Nuffield foundation director Sharon Witherspoon’s article, why the social sciences need a skills step-change in the Guardian was published the same day that Society Counts was launched by the British Academy.

My university degrees (BA, MSc, PhD) are all in geography, a discipline which underwent a ‘quantitative revolution’ in the 1960s. At the risk of over simplifying the history of geography the quantitative revolution can trace its routes to Fred K Schaefer’s posthumously published 1953 article “Exceptionalism in Geography: A Methodological Examination in the Annals of the Association of American Geographers”.  Schaefer’s sudden death prior to the article’s publication meant that he was unable to respond to or defend criticism of his ‘scientific approach’ from the likes of Richard Hartshorne, yet there was no shortage of geographers willing and able to build on Schaeffer’s idea. According to the more simplistic narratives of the history of geography, quantitative approaches were gradually edged out during the 1970s as more behaviourist and qualitative approaches took over.  By the time I arrived at university as an undergraduate in 1994 statistics was very much, in Witherspoon’s words, a ‘bolt-on’ module. (Please don’t cite this blog post as an authoritative reference for the history of geography.)

I’ll leave the relative merits of different approaches   to one side, but I’ll share some thoughts about why quantitative approaches are frequently rejected.

1. There is an adage that if you can add, subtract, multiply and divide you can do statistics. If we are talking about the mechanics of undertaking statistical tests then there is a degree of truth in this, but how many beginners’ statistics texts adequately explain the normal distribution, z scores or standard deviations? This stock diagram always appears in some form, but few attempts are made to really demystify it. Who came up with this? What is a standard deviation? Why are 68% of values within one standard deviation of the mean? The beginner is more or less asked to accept this as an article of faith.
2. Similarly critical values. You have to look up your answer on a table of critical values somewhere in the back of the book. Again the beginner is not troubled with any sort of explanation about where these critical values really come from. Why different scores for t, q, u, chi etc.
3. The internet is the enemy of the beginner. This is slightly unfair as there is some good stuff out there, but most statistics resources reinforce the above.
4. Statistics as a ‘bolt-on’ leads encourages surface learning. Statistics is a hurdle that has to be cleared. Technique is emphasised above understanding. Even the best teachers of quantitative social scientists don’t have time or scope to get to grips with true understanding.
5. The pressure to learn statistical analysis software creates an additional barrier in time and learning.

Rebuttals and thoughts welcome.

# British Academy publishes position statement on quantitative skills

From the British Academy 'Society Counts' webpage.

The British Academy has launched a Position Statement on the issue of a quantitative skills deficit in the humanities and social sciences. Well-rounded graduates equipped with core quantitative skills are vital if the UK is to retain its status as a world leader in research and higher education, rebuild its economy and create a modern participating citizenry. Quantitative methods facilitate ‘blue skies’ research, and without them, effective, evidence-based policy-making would be unthinkable. Yet, the UK currently displays weak quantitative ability within its humanities and social sciences.

The online book for Statistics for Humanities I am working on is funded under the Languages and Quantitative Skills programme.

# Update on the "Statistics for humanities" website and the failings of the Internet

Firstly…

For the past few months I have been working on a British Academy funded online book to introduce humanities students to statistics. The website is under development and is not public at present. If any readers are interested in providing feedback, please get in touch (j.canning[at]soton.ac.uk) and I can give you an access password in the next few weeks.

Why is this website/ book/ resource is needed?

There are thousands of introductory statistics texts on the market, and I’ve only looked at a small number of them.  In my view a majority of them go too far too fast. For some disciplines this may be appropriate, but introducing the normal distribution in Chapter 1 is frightening to students who have not studied mathematics since the age of 16, and many humanities students are in this situation.  Just to give an example I have the Second edition of Statistics in Geography by David Ebdon on my desk.*  I bought it when I was a geography undergraduate in the mid-1990s, by which time the text was almost 20 years old. I actually think it’s a good book on many levels and I frequently refer to it, but the first chapter introduces data types, probability theory, the normal distribution, hypothesis and significance. As a geographer without an A-level in Maths I found all this a bit much. In the sense of getting good marks I did well in statistics at as an undergraduate, but I can’t claim I really understood what I was doing. For non-mathematicians, especially those in the humanities and social sciences, statistics is very much a ‘hurdle’ to be overcome. Surface learning is the order of the day. With this book I take slower approach whilst hoping to make statistics seem interesting and relevant, but using humanities type examples.

The Internet

We have become so used to the idea that “everything” is available on the World Wide Web that we take it for granted that anything we want to know is out there online somewhere.  Searching for anything to do with statistics leads to seemingly random pages put up to support undergraduate-level statistics courses. Some of these are very useful of course, but on the whole these relate back to a face-to-face course of which we have no knowledge. Some of these websites are among the oldest pages on the World Wide Web. In many cases this is not a problem, but there is no shortage of webpages with references to pre-‘Windows’ versions of Minitab . Wikipedia is useful for many things but statistics really isn’t one of them, as discussion of the statistical tests is highly theory bound.  On the plus side there are any good videos elsewhere. As I’ve mentioned before Daniel Judge’s youtube videos are particularly excellent.

Two annoyances (or surprises)

1. Surprisingly, although the World Wide Web has been with us for nearly 20 years, displaying mathematical notation online is still a problematic area. I have managed to resolve it to my satisfaction and made this the subject of my last post.
2. A second surprise (annoyance) lies in in my attempt to find critical values tables in a useful online format. Every statistics book contains them  and they are available online in various formats—I’ve seen some in tables on webpages, scans of tables from books, pdf etc. etc. I have yet to find the tables I need in one place.  It strikes me as surprising that Neave’s Statistics Tables: For Mathematicians, Engineers, Economists and the Behavioural and Management Sciences is not available as a website. Copyright warnings are printed on the amazon preview, but I’m not sure the tables themselves are under copyright. Copyright and critical values tables are not something I expected to have to think about. If anyone could point me in the right direction about this I would be very grateful.

*This 1985 second edition is still in print. Not sure what today's undergraduates would make of the 17 computer programs written in BASIC.

# Pupil attitudes to languages survey now online.

I'm please to report that the findings of our survey of 1000+ school pupils and their teachers is now online. It also contains case studies of the status of languages within selected schools throught the UK.

John Canning, Angela Gallagher-Brett, Fabio Tartarini and Heather McGuinness (2010) Routes into Languages: Report on teacher and pupil attitude surveys (Southampton, Routes into Languages). Available from the Routes website.