The academic writing thread

[BassyBill]

My question is how would you collect sufficient data with sufficient variables to a level that you could analyse it such that it would be satistically significant.

If you collect some set of statistics but not enough to tease out any confounding variables you might see a correlation between A and B for which you cannot find any causal relationship because the causation is C->A and C->B and C wasn't captured.

If you have lots of data you are going to have to be fairly clever in terms of the analysis to sift it, though. I know a bit about some forms of analysis, but you'd need a good statistician.

You'd need good design for the statistics capture too, to avoid bias due to the capture, or even over capture of statistics that relies on people filling out forms can lead to either people being turned off the process and just making up responses on question 356 onwards of the 1000 question form, or in a longitudinal study people changing how they fill out forms because they understand what is being asked better, meaning that a change in response might not be due at all to what you might think it is.

Those questions I posted weren't all part of a single study - they were examples from separate small scale investigations I've done in the last couple of years.

For any folks interested (probably not many!), my current research area is as follows (title, followed by the RQs):

"Digital natives becoming teachers - how do they use technology to support their learning and professional development"

What factors influence the range of technology supported learning (TSL) activities that today's trainee teachers engage with in formal and informal learning situations?

What are these students' perspectives on their use of TSL in the context of their Initial Teacher Education (ITE) courses?

To what extent do institutional policies and strategies for TSL reflect the actual practices adopted by students in this category?

 

[BassyBill]

And yes, big congrats to Dr. Midnight!

 

[-Asdfgh-]

Those questions I posted weren't all part of a single study - they were examples from separate small scale investigations I've done in the last couple of years.

Any one of those questions would be hard to answer. I am not sure people realise how hard working out the statistics can be. (By which I means hats off to you if you answer any one of them).

I do hope teachnology is a typo and not a neoligism!

 

[BassyBill]

My knowledge of Q methodology is limited, but I never cared for it. (It's never been especially popular in psychology.) I still maintain a strong preference for theory-driven, rather than data-driven, research. For most purposes I'd rather survey a larger sample with a smaller, conceptually focused set of questions than a small sample with a large, diverse set of items. That said, I'd be happy to talk more about your project and help if I can.

I'm getting soft in my old age. I used to be very much a positivist, but now I find an interpretivist paradigm quite appealing for looking at people's views and perspectives. I'm not looking to convince others or win arguments so much as to shine a light for my own illumination. That said, the appeal of Q to me is the apparent rigour it can offer when interpreting purely qualitative data even from a small number of participants. I'll let you know how it goes.

 

[BassyBill]

Any one of those questions would be hard to answer. I am not sure people realise how hard working out the statistics can be. (By which I means hats off to you if you answer any one of them).

I do hope teachnology is a typo and not a neoligism!

It's a typo, thank God!

Thanks, I'll edit it.

See above about my comments on my rather relaxed approach to data interpretation and stats (and also Q methodology / factor analysis).

 

[Pilgrim]

My PhD is nearly 20 years back, and I work with instructional technology. Most of my writing is opinion or persuasive pieces. I'm usually either training faculty or trying to get them to use tools that move electrons. It drives me nuts when some faculty goof writes on a whiteboard with a worn-out market in a room with 200 students - ignoring a Smartboard which would project an image that all the students could actually read.

No one I have ever met in other professions can equal the tunnel vision of a faculty member teaching the way they were taught in 1960. And they really believe they know what they're doing, even though they've never even gone to an afternoon workshop on teaching techniques.

But I digress. and i believe I will offer further digression...

I also teach an online course that requires all the grad students (usually about 10 per term) to use APA style. I go through every paper with Track Changes on and I correct every comma, every missing quotation mark and citation, sentences that are incomplete or run-on....you get the idea. Why? Because these students are REQUIRED to write at the level of a fully qualified and skilled Master's student. If they can't, they should not earn the degree. My course has an objective of assuring that they are clear on the expectations of their writing. And although a few are very frustrated, it's surprising how many of them thank me for the skills they have gained in 12 weeks. The seven papers they write in those 12 weeks are each a "learning experience."

Do I grade down for failure to follow APA style? You bet, it's a base requirement. Do I grade down for poor sentence structure? Absolutely. Occasionally I get a paper that's so poorly written i can't even follow the arguments. In some cases they don't include an introduction, body and conclusion.

The good news is that about 8 out of 10 grad students write at least coherence and basic competence. One or two out of 10 grad students can't write at the level of a competent high school senior.

 

[LiquidMidnight]

Reading Lobster 11s comments above, I was struck by how different our approaches are: I occupy the opposite end of the investigative spectrum, relying heavily on data rather than theory, and qualitative data at that. For me, numbers are great for filing taxes. When it comes to understanding people, culture, history, and place, less so. I suppose that makes sense in my case, as I'm not a scientist; more of an interested outsider, who participates and chronicles. I do analyze, obviously, but case-by-case, with focus placed on lived experiences leading to collective knowledge.

This sort of goes back to the earlier discussion in this thread, but having both a psychology (highly post-positivist) and a sociology (everything from positivism to postmodernism) background, I've spent a ton of time thinking about ontology and epistemology. At the end of the day, I'm a pragmatist. Some questions are better suited to being answered with a regression analysis. Other questions may need a phenomenological or ethnographic approach. People think of the relationship between theory and data as being either deductive or inductive depending on paradigm, but the reality is that it's often abductive. With quantitative research, you still want to lean more on the deductive side to avoid having to run unnecessary statistical tests (that always increase the likelihood of Type I error), but even so, sometimes theories and models need to be refined in light of the data.

And this discussion dovetails into...

Any one of those questions would be hard to answer. I am not sure people realise how hard working out the statistics can be. (By which I means hats off to you if you answer any one of them).

Just going off of how Bill's questions are stated, I'd say that half of them are actually questions that necessitate a mixed-methods or purely qualitative approach. The ones that seem to be dealing with process would probably be well served by a grounded theory method.

I also teach an online course that requires all the grad students (usually about 10 per term) to use APA style. I go through every paper with Track Changes on and I correct every comma, every missing quotation mark and citation, sentences that are incomplete or run-on....you get the idea. Why? Because these students are REQUIRED to write at the level of a fully qualified and skilled Master's student. If they can't, they should not earn the degree. My course has an objective of assuring that they are clear on the expectations of their writing. And although a few are very frustrated, it's surprising how many of them thank me for the skills they have gained in 12 weeks. The seven papers they write in those 12 weeks are each a "learning experience."

I'm an APA nazi, and I actually developed an APA workshop as part of my graduate assistantship duties as a master's student that is still in use by the department today (hopefully with some updates for 6th edition *lol*).

 

[BassyBill]

As someone coming from physical sciences to education research, I've found it difficult to get used to the differing ontological and epistemological stances between the two (in the natural sciences, this isn't usually that much of a consideration). I guess that's why I have a mixed methods/eclectic leaning tendency to research design and also why Q - abductive pretty much by definition - is really occupying a lot of my interest. Purely quantitative number crunching of lots of stats from large samples is not the way to answer many of the questions I'm interested in.

The mathematics involved in factor analysis is amazing, by the way, and well beyond my own capabilities. I'm very much a "dump the data into the software and let that sort all the numbers out" guy. But the interpretation of the stats is where the qualititative approach comes in - hence "mixed methods".

As far as referencing goes, I've been tryng to use RefWorks but for the moment I've given up. The style models within it all seem to have glitches when it comes to generating bibliographies and it takes as long or longer to go through and edit every tiny bit as it does just to generate the list manually. I'm a real nazi in this regard, like LM, and still prefer to go through my work on a word by word basis to check citations and referencing rather than trust something like RefWorks to do it for me.

 

[TUEP]

If I can find my group paper that I led on the modern Russian military and how its tactics and vehicle development are designed for the Russian geography then I will post it.

 

[Pilgrim]

I'm an APA nazi, and I actually developed an APA workshop as part of my graduate assistantship duties as a master's student that is still in use by the department today (hopefully with some updates for 6th edition *lol*).

I'd like to find a good online APA tutorial that I can direct the struggling students to. If you know of one, please share a link.

My best resource so far is the Purdue O.W.L., which is a marvelous style resource, but not a tutorial. I use it a lot myself, because if you're going to throw stones about style, you dang sure better make sure that what you say is absolutely accurate!

 

[-Asdfgh-]

Purely quantitative number crunching of lots of stats from large samples is not the way to answer many of the questions I'm interested in.

To reference Office Space: I'm going to have to go ahead and disagree with you there.

But you could take two approaches: (a) form theory, analyse data, see if it supports it, form potential predictions, test against new data or (b) take data, apply statistically relevant filters (or something that is robust to noisy data in the first place) to take into account noisy data, look for correlations, form theory based on potential causal mechanisms, potential predictions, test against new data.

One thing that occurs to me is that looking at various output variables applying a technique I worked on from something seemingly unrelated which works on multivariate data might yield interesting results. Except that I don't know how I would take into account changing input variables in that scenario. I'd have to think about it but I don't think I have the time to actually develop it. Hmmm.

 

[LiquidMidnight]

I'd like to find a good online APA tutorial that I can direct the struggling students to. If you know of one, please share a link.

My best resource so far is the Purdue O.W.L., which is a marvelous style resource, but not a tutorial. I use it a lot myself, because if you're going to throw stones about style, you dang sure better make sure that what you say is absolutely accurate!

I definitely will, but I've found OWL to be the best online resource so far, only bested by the book itself.

 

[LiquidMidnight]

As far as referencing goes, I've been tryng to use RefWorks but for the moment I've given up. The style models within it all seem to have glitches when it comes to generating bibliographies and it takes as long or longer to go through and edit every tiny bit as it does just to generate the list manually. I'm a real nazi in this regard, like LM, and still prefer to go through my work on a word by word basis to check citations and referencing rather than trust something like RefWorks to do it for me.

Yeah, I've yet to find any type of satisfactory citation program or feature. Even End Note, while a great reference management system, is rather lacking in producing stylistically correct citations.

 

[BassyBill]

To reference Office Space: I'm going to have to go ahead and disagree with you there.

And off we go - PARADIGM WAR!!

Just kidding. Thanks for the interesting contributions to the thread so far from everyone.

 

[BassyBill]

Yeah, I've yet to find any type of satisfactory citation program or feature. Even End Note, while a great reference management system, is rather lacking in producing stylistically correct citations.

Well, I'm very glad it's not just me.

 

[DiabolusInMusic]

I definitely will, but I've found OWL to be the best online resource so far, only bested by the book itself.

I would second this. Mind you, I prefer Chicago style.

I don't write in the same league as you guys, I am an undergrad so my papers are not important. I just finished writing one piece about the White Paper being the impetus for the Canadian Indigenous social movement. I am now working on a paper about the same thing but only applying some framing theory on the situation. I had two major papers for two different classes, only a fool picks two different topics!

Oh, and congrats on the whole Doctor thing. Just don't be one of those pretentious a-holes that demands Dr. must be used before their name (unless the situation calls for it )

 

[Lobster11]

In my earlier post I introduced myself as being firmly on the positivist/deductivist/quantitative of the "paradigm war," but this is not to say that I believe this approach to be superior in any general sense. I readily acknowledge that different kinds of questions, and different stages in a research program, call for different philosophical approaches and methodologies. Some obvious examples are the differences between applied and basic research and between exploratory research and theory testing. My personal bias owes to the fact that I work in an area of basic science that focuses on theory-testing. Sometimes the song calls for playing fingerstyle; sometimes it calls for slapping, other times with a pick. (Am I allowed to say that in Off-Topic?)

 

[PJ Muzikmansky]

As someone who's had no formal education past high-school, but would love to be able to write well, I find the company on Talkbass to be both intimidating and humbling at times.

If a genie was to appear out of a bottle and grant me three wishes, they'd be:

1) To bring my father (1938-92) back from the dead;
2) To write like Christopher Hitchens
3) To play bass like Anthony Jackson.

Gentlemen, I salute you.

 

[Lobster11]

On the topic of writing per se, has anyone seen Steve Pinker's new book Sense of Style? I've only read the first chapter so far, but that chapter alone is worth the price of admission....

 

[Pilgrim]

I would second this. Mind you, I prefer Chicago style.

I don't write in the same league as you guys, I am an undergrad so my papers are not important. I just finished writing one piece about the White Paper being the impetus for the Canadian Indigenous social movement. I am now working on a paper about the same thing but only applying some framing theory on the situation. I had two major papers for two different classes, only a fool picks two different topics!

Oh, and congrats on the whole Doctor thing. Just don't be one of those pretentious a-holes that demands Dr. must be used before their name (unless the situation calls for it )

On the contrary - it sounds like you have your teeth into worthy topics and are taking a smart approach with them. What you're writing about sounds interesting, and not trivial.

I go by Al, and half my colleagues don't even know my full first name. So did my dad, a PhD who taught Engineering for 34 years at my alma mater. In Colorado where I live, people go easy on the "Doctor"honorific unless they're in a somewhat formal situation. In Texas where I earned my PhD, everyone seemed compelled to add "Doctor" and address me by my last name after I finished. Just a difference in the formality of the culture in two different parts of the US.

About writing - I love first person, active voice, clarity in writing and economical use of language. Third person and passive voice academic writing bothers me and I tend to avoid it. As fa as I'm concerned that style only belongs in a formal research report that must be dispassionate and impartial.

Want to know who I would most like to be able to write like? Answer: Mickey Spillane, with Damon Runyon close behind. Spillane's writing is so earthy and distinctive that you can read one or two paragraphs in any book he wrote and you know who the author was. His protagonist Mike Hammer is a classic figure in American popular literature.

Sample of Spillane (and boy, is this NOT an academic book...)