My day job involves using education data to improve the experience of students.  Specifically I am working on a predictive model that will serve as an early warning system when students begin to have problems in a course.  The idea is that there are a LOT of resources out there that students don’t use because either they are unaware of them or they are so tied up with their own issues that it never occurs to them.  I want to change that by spotting signs of trouble while the student is still taking the class and proactively offering them help.

This conference is an ideal tie-in to this work, and an area where I could easily see myself forming a research agenda as well as a satisfying work life.  We collect so much information; everything we do is tracked electronically and recorded somewhere.  There have to be ways to use this information for something other than selling us another book on Amazon.  I want to work on finding those other uses.  (Frankly this may be the perfect field for me; I have 20 years in IT prior to getting my PhD, almost all of it in databases and data management.  Now I can apply those skills to a real problem.  Very appropriate.)

The goal of the course itself is to provide context and background to the learning analytics field. In many ways it is a small, emerging field, and while there are some things we know there is a lot more we need to know.  The course helps develop people and bring awareness to the field.  Here is the course description:

Capturing and analyzing data has changed how decisions are made and resources are allocated in businesses, journalism, government, and military and intelligence fields. Through better use of data, leaders are able to plan and enact strategies with greater clarity and confidence. Data is a value point that drives increased organizational efficiency and a competitive advantage. Simply, analytics provide new insight and actionable intelligence. Companies such as Microsoft, IBM, Google, and Amazon are investing heavily in technologies and techniques in helping individuals and organizations makes sense of, and unlock the value within, big data.

In education, the use of data and analytics to improve learning is referred to as learning analytics. Analytics have not yet made the impact on education that they have made in other fields. That’s changing. Software companies, researchers, educators, and university leaders are starting to recognize the value of data in improving not only teaching and learning, but the entire education industry.

This course will provide an (generally non-technical) introduction to learning analytics and how they are being deployed in various contexts in the education field. Additionally, the tools and methods, ethics and privacy, and the systemic impact of analytics will be explored, presenting a broad overview of the current state and possible future directions of the field.

I invite any of you who are interested in how we can use data to inform and improve our teaching to join the class.

http://lak12.mooc.ca/cgi-bin/login.cgi?action=Register

I’ll be posting over the next few weeks my thoughts and observations from the class, so you can certainly follow that way as well.  You can also follow me on twitter (@protoscholar) or the conference as a whole by following #lak12

How project management applies to research

Think of yourself as the customer, the project leader as your chair, and the others as your committee and you'll be pretty close...

Your research is a project, or rather a series of projects within a larger program called your research agenda.  The first big project is your dissertation, but nearly every paper, conference presentation and book/chapter after that qualifies.  It therefore pays to learn early some of the basic principles of managing such a project.

Note: I do NOT advise taking formal project management training.  That training is intended for a much more complex environment than most researchers face.  If you end up running a lab somewhere you’ll need to learn those skills (or more likely hire a project manager to do the work for you), but early in your career you will be mostly concerned with yourself and maybe a couple of others.  Full-scale project management is too administration-heavy for that environment.  However there are some principles you can and should borrow.  That’s what I’m going to talk about.

The Triple Constraint

If there is any single project management concept that you should learn, it’s the triple constraint.  It goes like this.

You have three general types of resources at your disposal:

• Time
• Scope
• Money / Costs

You can control the SCOPE of your project.  You do this when you define the research question you are planning to answer.  For this reason, a person introduced me to the concept early on of the LPU: The Least Publishable Unit.  You should choose projects that contain sufficient new ideas to get published, but NOT put all your great ideas into one paper.  When people ask me how I got through my dissertation so quickly I point to this idea.  There were a LOT more questions I could have asked, but I tightly controlled the scope of the project in order to get it done.

Scope ties to TIME.  The bigger the scope, the more time it is likely to take.  And TIME ties to COST, both in terms of additional time at a lower salary to the cost of lost opportunities while you work on the old stuff.

Here’s the thing about the triple constraint.  You can flex one leg of the triangle, but when you do, you need to change the other two accordingly.  The three are still joined at the corners, and something has to give.

A couple of examples:

• My chair announced he was taking a year-long sabbatical in Spain (lucky dog) and that, therefore, I needed to finish before he left.  TIME was constrained.  Therefore I strictly monitored the SCOPE of my project and invested more MONEY by paying people to do certain things for me (like editing and delivering meals) in order for me to get the project done in that time frame.  I kept a tight rein on SCOPE and allowed it to COST more in order to meet the TIME constraint.
• A colleague got to their proposal review and had their committee heap a bunch of additional stuff on to the scope of the project.  (In their defense, she will be writing on this data for a decade and got some great stuff.)  She didn’t have a lot of money to spend, so in her case she cut out everything else in her life to finish the project.  She put in more TIME to make up for the increased SCOPE.

One of the dangers of the triple constraint in academic circles is the COST issue. We tend not to see it clearly.  All too many PhD students take a very long time to finish their dissertations due to massive scopes without any awareness of what that time is costing them.  The tuition cost is easily quantified, but few look at the cost of continuing on as a graduate assistant rather than getting a full-time job. Early career faculty see six years to tenure review as a very long time when, in fact, with academic publishing on a stone-age schedule, it is actually quite short. And this is before we get to what economists call the Opportunity Cost.  What COULD you have been doing/making (now and for the rest of your career) if you were done now?

Scope Creep

One of the biggest pitfalls for academics is Scope Creep.  We ALWAYS have more questions, more ideas, more literature to read, etc.  I got stalled for about a month on my literature review; there were so many interesting things to read, most of which were only peripherally related to  my hypothesis.  Then I almost did it again with methodology books/articles.  It’s in our nature to be curious, but that makes us especially at risk.

You have two protections against scope creep.  First, tightly define your research question for the project at hand and stick to it.  If what you are reading/asking doesn’t apply to that question, set it aside for now.  Second, keep a notebook of ideas to explore and things you want to read.  This is where you place those things not related to your current project but that could lead you in potentially interesting directions in the future.  Plan time between major project to review that material.  Your Nobel Prize may start out in that notebook, so it’s worth going back to regularly.  For more on this topic go back and read my post on Cal Newport’s research framework.

Planning

The first stage of planning is determining which of those constraints represent hard limits.  Think broadly here.  Whether you are heading toward graduation or tenure review, there is always a deadline.  What defines that deadline in terms of graduation might be money (your assistant-ship runs out at the end of year four) or an externally imposed deadline (tenure at year six or your contract doesn’t get renewed).  Regardless, figure out what it is and then back into what your scope needs to be in order to meet your objective.

Once you have that objective defined, you can now lay out how much time and cost will be involved in achieving it, as well as the minimum scope of the projects.  PhDs very rarely do only the minimum, but if you know what that minimum you can be sure to meet it while going above and beyond in your core areas of interest.

I would not advise making up a gantt chart, detailed work breakdown structure or any comparable formal project management tool.  Since much of the work of an academic is solitary or in small groups, using the agile project management ideal of focusing on communication is usually your best bet.  However that is a big huge topic for another day.

If you have questions about managing your research projects and agenda, feel free to leave a comment or ask/follow me on twitter: @protoscholar.

He recently put up a great post that looks at his framework for putting together a coherent research agenda and for supporting that agenda.  My first thought was that I wished I had seen this earlier in my grad student career.  (You know; like before I had graduated.)  My second was that the system could be adopted at ANY point in one’s career, although it might take some time to really get everything into it.

What I like the best about this conceptual system is that it uses a mission statement (which includes the primary area of research) in order to coordinate and direct the work, ensuring that there is a thread of coherency to the work.  My work is a primary example of what happens when you lack that coherency;  I have publications in history of ed, higher ed, charter schools, and looked at a national data set for my dissertation.  I can force them all together if I have to, but it takes some explaining.

The second thing I like about this is that he doesn’t try to suggest that you can go away for months and come back with some type of genius product.  The entire process is iterative and in manageable chunks.  He talks about learning new things in his field (although at a rate of 1 item per week, that must be fairly tightly defined) on a weekly basis, and using that to brainstorm new ideas.  This is something I could easily adopt myself, although my field (possibly due to the presence of numerous think tanks and such) puts out far more than 1 new item per week.

The new ideas need to be timeboxed into testable chunks, and small enough to take under a month, but big enough to be something that could turn into a talk or something on which he can get feedback.  This might be challenging for some social scientists, but could well be possible if you are mining an existing data set.  (All the more reason to ensure that when you take the time to collect data, you go beyond just collecting the minimum.)  Notice that this step isn’t required to be publishable on it’s own; the idea is to use these short projects to create the building blocks of something bigger and to vet the pieces.  I would imagine that some don’t work out and get tossed; if so, limiting the time spent on them is an incredibly important component of long-term success.

Finally the new items are used to support grant applications and papers.  By the time this point is reached, you’ve gotten some feedback on the ideas and have a selection of pieces that can be combined into something bigger.

If I were a full-time academic I would be working to implement this already.  Because I’m not, I have to be a bit more patient about how I implement something like this.  My research agenda is only partially my own.  My day job informs many of  my research questions and, realistically, expects answers to them that are not publishable due to the proprietary data I have access to.

Nonetheless, the approach (if not the speed) is something I am going to try to work in to  my own approach to research.  There is no reason I can’t apply a process like this at any speed, and this approach would be exceptional for a student working on their dissertation.  Yes, they may need to read more than one new thing per week, but the idea of regularly brainstorming small, testable ideas is one that has merit for any researcher.

I don’t think I could have been more wrong.  I need to learn the following:

• Data Mining - finding those nuggets of information in a large data set that can be used to predict student behavior and therefore design interventions for them.
• Predictive Modeling and Analytics - seems like my job is going to be all about statistical crystal balls.  While this often is considered to include data mining, it’s sort of a matter of data mining telling us what is going on, while predictive analytics suggests how to use that information.
• Markov Chain modeling - suggested for predicting enrollment and student behavior more accurately
• R - extremely comprehensive and growing statistical programming language.  Harder than SPSS, but far more complete as well.
• Bayesian analysis - A different statistical approach from what is called the frequentist approach of traditional hypothesis testing.

But here’s my question:  How do I go about learning all those things?  None of them were covered in my program.  (I’m pretty sure that I would have needed a PhD in statistics to see these things covered.)  I’ve taken a short program on R that helped a little, but it really is a programming language rather than a menu driven tool, and that is going to take some work before I feel like it is as easy to work with as SPSS; work I often don’t have time for when I just need to get something done.  The other areas (the first 3 specifically) are all things I need for my job and have to learn.

Now, obviously, there are books.  Lots and lots of books.  But how do I figure out not just which ones are the most accurate but also the easiest for self-teaching?  I know what some of the best are (such as Gelman‘s book on Bayesian Data Analysis) but don’t have that warm fuzzy feeling that says I know enough to learn it that way.

MIT Open Courseware has a Data Mining course from 2003, but it’s only partial and (more importantly) doesn’t include solutions for tests or homework.  How can I tell if I understood itt?  And what has changed in the field in the last 8 years that might impact my work?

Other options include industry-oriented training programs.  There is one 2-day program on predictive analytics, but it isn’t offered until October in SF.  It’s $1400, and I may talk to the folks at work about paying for it, but when you add travel it starts to get pretty expensive.  Another is in August in Minneapolis for $1500 and is really only part of their full week (which would actually be $3500 for the whole thing).

And then there are academic programs.  I found a Master’s (Northwestern) and two certificate programs (Stanford at $3900 per course and UCSD at a more modest $695 per course).  But is that falling back on what I know?  After all, it’s safe to say that anyone who gets a PhD is “good at school”.  Regardless, the Master’s repeats too much of what I already know, so that’s out.

Another issue is that some of these technical course have prereqs of calculus and matrix algebra.  Now, I took business calculus as an undergrad, but I would be an idiot to think I remember any of it.  I’ve seen matrix algebra a couple of times, but don’t truly get it.  More importantly, these are things needed to understand the theory behind the techniques, not to perform the techniques.  The academic in me thinks they are important, but the professional who needs to show results is less convinced.

And here’s the real rub of the whole thing.  I feel as though a PhD should have prepared me to be able to learn anything, but in reality I only feel that confidence in relation to non-technical subjects.  I taught myself psychology in a summer, and while there is still more to learn I feel relatively conversant in it.

These types of highly technical topics, however, strike me as much harder to teach one’s self.  Teaching myself from a book seems to require a better grounding in the prereqs than I really have.

And then there is the ego-factor.  I feel a bit awkward asking for my boss to pay to send me a 2 day course in predictive analytics when I’ve already been dabbling in it and have all this statistics background.  It’s like I shouldn’t need them to do that.

So I’ll pose this to those of you who read this far.  How would you go about learning a new highly technical skill/area?

Conference proposals (and resulting presentations) serve a number of purposes.

• They allow a researcher to get their work out to an audience, in some ways more quickly than through journals
• They offer a great networking opportunity at which you can meet others interested in the same topic
• They get a new researcher exposure and name recognition with the established leaders in the field
• You can get some feedback on an early draft of a paper and refine it before submitting to a journal

My current thinking is that my summer would be better spent writing up my work for a journal, not crafting a proposal.  However one thing on my to-do list is to research missing data methodologies and apply them to my data in order to allow me to draw some conclusions on the really different questions I was trying to answer.  (The conclusion in my dissertation was that I can’t answer them because once the missing data is removed the sample is no longer representative.)A friend suggested that I could get a lot of mileage out of submitting the proposal with a discussion of the intended change for missing data techniques written up.  It would both be interesting and different, as well as ensuring that I actually did the work.  I can write a good proposal.  (Heck, the very 1st one I did my 1st year of graduate school was accepted and upgraded from a poster to a presentation.)  But would I be better off just charging forward?

I haven’t decided yet.  I have to admit that conference proposals are good ego boosts, in that so far most of mine have been accepted.  It feels good to get that positive reinforcement back, and far more quickly than a journal would get back to me.  But is it a waste of time when I have a pile of other presentations under my belt?  Can I do both without delaying either?

I am now deep in to the “hurry up and wait” phase of my dissertation. At the beginning of January my chair finally signed off on my proposal, and if I want to graduate in May I needed to get him a draft of the results and discussion chapters by the end of January.  I did, working nearly every minute I wasn’t at my day job, and sent it to him literally at 10pm on the 31st.

Now I wait until he gets around to reading it, putting in comments, and sending it back to me.  There isn’t a lot of sense in continuing to work on anything, since he may come back with a direction that will make any work done irrelevant.  I’ve arranged for an editor/formatter, verified all the key dates (to the editor by March 15th, apply to graduate and turn in document for format review March 31st, defend around April 15th) and tried (several times) to get my chair to at least acknowledge that he received the document and give me a tentative date.  I need to know so that I can arrange time off for that last final push to get the dumb thing done and back to him for review at least one more time before the editor.

Now, I believe my results chapter is pretty good.  Or rather it is every bit as dull and repetitive as it needs to be, thorough about explaining what I did and why, and covers all the things I found.  There is an infinite amount of additional digging I could be doing, but that starts to get out of control fast.

Unfortunately I didn’t find what I had hoped was there, which made the discussion a bit more difficult.  I tried, but expect to get a lot of comments back on that and some on the results.  But that’s why I worry that he might suggest a direction change, in which case this is going to take a lot of time.

In the meantime, I took care of all that girly maintenance stuff – eyebrow wax, facial, pedicure, haircut.  And a little shopping.  It was enjoyable, but one weekend was enough.  If Chair hasn’t gotten back to me before next weekend I’m going to start getting really antsy….

So I’m trying to run the logistic regressions today.  First I put each of the most promising set of variables in individually.  When you do a logistic regression in SPSS it’s base model is to look at the distribution of the outcome variable.  If more than 50% have the outcome variable, guess all have it.  If less, guess none have it.  For me this makes a blind guess around 70% correct.

The ONLY version of the model that adds anything substantive (around 10% better prediction) uses the two variables that you get latest in the process – like at the same time you get the outcome variable.  Specifically you can make a better guess as to whether someone will need placement into remedial math if you know whether they needed to be placed in remedial reading or writing.  Unfortunately that’s a bit too late to do anything about the situation.

All of the other potentially predictive variables, including high school GPA, math test scores at two different points, math self-efficacy, or demographic stuff, fails to add more than 0.1% to the predictiveness of the model.  Worse, some of these actually reduce it by up to 0.7%.

I don’t know what I’m doing wrong here….I mean, I didn’t expect a nobel prize for this thing, but I expected to find something.  People do it all the time.  I have to assume I’m doing something wrong…..I just don’t know what…

Books about the Graduate School Process

The graduate school process is a thing onto itself.  From application to oral exams to thesis/dissertation defense, the process is like nothing else in the world.  Here are some books that can help.

Demystifying Dissertation Writing: A Streamlined Process from Choice of Topic to Final Text Unlike Destination Dissertation below, this book provides excellent techniques for some of the very mechanical tasks you will be doing.  What is the most expedient way to take notes on the literature?  How do you organize all that material?  Destination Dissertation also contains these kinds of tips, but frankly I find the ones in Demystifying Dissertation Writing more usable.   Check out my review as well for more ideas of what is in this book.

Playing the Game: The Streetsmart Guide to Graduate School This book is one of the more realistic books I’ve come across with respect to the actual process of graduate school.  It’s also irreverent and funny.  It starts with asking an important question; do you REALLY want to do this and why?  If yes, it then plots the shortest possible path from application to graduation.  It further helps you clarify where you should cut corners and where you shouldn’t.  I’m pretty sure most faculty don’t want us thinking about the process this way, but frankly given the crappy academic job market this type of fore-thought is a leg up on finding a job after graduation.  That alone makes it worth the money.

Destination Dissertation: A Traveler’s Guide to a Done Dissertation While this book is focused strictly on the end process of completing your dissertation (which is far and away the hardest part), it contains a remarkably down to earth approach.  Specifically, it lays out the steps you need to go through and estimates of how long each should take (in working hours).  This may sound like no big deal, but the hardest part of project management is always figuring out what all the steps to the plan are.  These guys have done it, AND given you an idea of the level of effort required.  Whether you can work on it every day all day, only on weekends (like me) or just a couple of nights per week after the kids are in bed, this book provides a structured plan and an idea of how much further you have to go.

Note that I have left out one of the most commonly cited books in this area, Writing Your Dissertation in Fifteen Minutes a Day: A Guide to Starting, Revising, and Finishing Your Doctoral Thesis. This is for two reasons; First, it is somewhat out of date in terms of technology and process.  Second, there is one principle piece of information you get out of it, and I can sum it up for you right here:

Write a little every day, even if what you write sucks or you do nothing more than type “I don’t know what to write” over and over.  Write often, don’t wait for bit chunks of time.  The more you do it, the better you will get.

I’m certainly not telling you NOT to read this book, as it’s considered a classic for a reason.  But If you take that one piece of advice and use the techniques and processes from the other books it will get you just as far.

Books about the Academic Process

Being an academic requires changing your approach and integrating new ways of writing and communicating with other scholars.  These books can help with that process.

The Craft of Research, Third Edition (Chicago Guides to Writing, Editing, and Publishing) Also a classic, this books goal is to help you move into clear, concise research writing.  It will help while in school and for years to come as you work on becoming published and later teaching others how to do the same.  It isn’t short, but it is comprehensive and can be used as a reference book when you get stuck.  Think of it as the old 5 paragraph essay format you learned in middle school all grown up.

“They Say / I Say”: The Moves That Matter in Academic Writing (Second Edition) This book is all about the process of entering the conversation.  Scholars use specific phrases that vary only slightly between fields, but in essence every paper amounts to a recitation of what other scholars have said, either an acknowledgment of why they are right and important to your argument or a polite disagreement, and then presenting your own argument.  This book gives you a leg up on integrating these types of arguments into your writing smoothly and respectfully.  Additionally this is a short, easy read.  Given some of the stuff you will read in graduate school that will be a welcome change.

The Chicago Guide to Writing about Numbers (Chicago Guides to Writing, Editing, and Publishing) Even if you plan on writing a qualitative dissertation, you will still be talking about other people’s numbers and this book will help you learn to do so clearly and concisely.  This is not as easy as it sounds, as anyone who has seen a page full of numbers or a junk chart knows.  This book isn’t about artistic infographics; it’s about clarity in research papers, which are going to be your entire world for a while.

Books about the Writing Process

We all like to think we are decent writers, but we all need some help.  Here are some of the best:

The Elements of Style (4th Edition) Strunk and White’s classic book is as relevant today as ever.  What really amazed me is that it is short, concise, and easy to find things in.  Read it through if you can (or at least skim it) then keep it close at hand as you write.

Eats, Shoots & Leaves: The Zero Tolerance Approach to Punctuation Who thought punctuation could be funny?  This book never ceases to make me laugh while FINALLY explaining where I should and shouldn’t use a comma.  (I still use too many in casual writing.)  This book is useful as much because it is an easy and pleasant read as the fact that you will learn something from it.

Economical Writing, Second Edition While written for economists, this book is a quick, easy read that makes a strong case for getting rid of the weasel words, writing in active tenses and being concise.  People outside economics may not recognize the names she throws around or understand what the quotes she uses means, but you will quickly see how you can improve your writing while making it clearer to your reader.  Probably one of the most valuable skills out there.

Your Field’s Style Manual

Every field has a chosen style manual.  Own it.  Social sciences generally use APA, Humanities generally use MLA, History generally uses Chicago, etc.  Whatever it is have a copy close at hand.  (Since you only need one, this counts as just one book.)

Publication Manual of the American Psychological Association, Sixth Edition
MLA Style Manual and Guide to Scholarly Publishing, 3rd Edition
The Chicago Manual of Style, 16th Edition

If you are short on time I would suggest going for the first book in each section.  Those are the books I turn to again and again.  They aren’t necessarily the shortest or the most amusing.  They are, however, the must reads on this list (along with your style manual) and books that will serve you well throughout your career.

1. Set deadlines for yourself - even artificial ones will keep you moving forward.  These deadlines should be realistic but not lazy.  For example, I gave myself a set deadline for sending copies to my chair and then worked to hit that and send him SOMETHING.  It’s way too easy to see time as infinite and never actually finish what you’ve started if you don’t give yourself interim milestones.
2. Outline – organize your thoughts as outlines or bullet points.  Then write within those sections when you are ready.  This helps you to never face a blank page and always have an idea of where you are going.  You can move things around later if you need to, but that starting point is invaluable.
3. Keep it focused – One of the hardest things for me to get my hands around was staying tightly focused.  It’s really easy to wander off into side topics and marginally relevant information.  Remember that your goal is to finish what you are doing, and that means staying focused on that end result.  Particularly in the literature review, keep asking yourself “how does this relate to my research questions”.  If the answer is that it doesn’t directly relate, it’s time to redirect yourself.
4. Keep it succinct – While most are, dissertations/thesis don’t HAVE to be painfully long.  Don’t mistake length for clarity.  Clarity comes from short, focused phrases, using the active voice and avoiding excess descriptive or weasel-words.  I recommend Economical Writing.  While targeted for economists, it can help any academic writer avoid the traps that lead to lengthy, confusing documents.  More important, it’s short thereby modeling the behavior.
5. Find out what the required format is and use it from day one - Every university has a format they want dissertations to be submitted in.  Look that up BEFORE you start writing, create yourself a template in your chosen word processing software, and then use the template.  The result will be a LOT less work later.
6. Make use of appendices – The goal of the main text is readability.  Huge tables of numbers hurt that goal.  Use an appendix.  My dissertation already has two; the first will contain the list of fields I am using from the data set, including their definitions and possible values.  It may also include comprehensive descriptive statistics for those fields.   The second will include the detailed statistical tables of my results.  In line with the text will be just the key statistical elements needed to make my argument.   This will make the final document both easier to read AND easier to convert to a series of journal articles when I’m done.
7. Create a writing schedule that works with YOUR life – One of the most common pieces of advice you hear is “write every day”.  That is the entire point behind the Academic Writing Club is to do just that, and its great advice.  But it may or may NOT work for you.  If not, find a schedule that DOES work and stick to it, whatever that schedule is.  If you can only write for 8 hours a day saturday and sunday and not think about it during the week, then do that.  Just make sure you are consistent and include any limitations in your deadlines (point 1).
8. Read a few other dissertations from your university – Focus on recent ones on related topics and ones that your advisor has been chair on.  This will help you get a strong feel for the style and quality you are shooting for.  This is important, because most of what we read (peer-reviewed journal articles) seem far beyond what we are capable of right now.  But no ones dissertation is at that level.  Those are the best work of experienced academics; the dissertation has been rightly called “the worst piece of research you will ever do”.  No one expects it to be at the level of papers in the top journals.  You will find writing much easier if you have a better idea of what the real target is than if you keep comparing yourself to the pros.  Your dissertation’s job is to show basic competence and promise to eventually reach the pro level, not that you are there now.
9. Write from the very beginning – One mistake I made was thinking I needed to read first, then write.  There is some reading that needs to be done first to ensure that you have the background, but at some point you need to start organizing and writing as well.  Start with writing your outline; model your document on the studies you are reading and use their organization as a way to begin yours.  You will almost certainly end up moving things around, but you would be amazed at how helpful getting something started is for moving the project forward and helping you  understand where the gaps in your knowledge are.
10. When in doubt, just start writing anything – Write about how you don’t know what to write, then write about all the thoughts swirling around in your head.  Write about your fears and worries.  Get them out.  You may find that the process helps you organize your thoughts, figure out the gaps and help you move forward.

One more thing related to but not specifically about writing:  Work WITH your advisor – It is reasonable to ask your advisor to review your work and give you feedback/direction.  It is not reasonable to either pester them every day or to behave as though you are begging for their time/attention.  Mentoring you is part of their job; holding your hand/doing the work for you or ignoring you is not.  Find a balance.  For example, when I send off a chapter/chapters to my advisor I ask for an estimate from him of when he can respond.  I don’t expect it on my time frame, but I do want to know his so I can plan my work accordingly.  The first time he got back to me on my preferred time frame and I went back to work. The second time he was unable to do that.  KNOWING when I would hear from him let me decide how to allocate my time better and make forward progress.  Communicate what you would like (for him to review by X date) and what you need (to know when the review will be done regardless of whether it is X date or later) and be respectful of his time.  By acting like a professional, you will encourage him to treat you that way as well.

<crickets chirping>

A year later we heard that the journal had gone through a couple of “changes” and wanted to know if we wanted our paper back or were willing to wait*.  We said we’d wait.

<more crickets>

Friday we FINALLY got the paper back – revise and resubmit.

The number one comment by both reviewers?  The literature review was out of date.

^&(*^%$%$#^&^&()()%$&*)*&&^%#@&*(

Well, DUH!  Of course it’s out of date.  YOU PEOPLE SAT ON IT FOR YEARS!!!

*sigh*

So now, instead of working on my literature review, I’m working on the lit review for this paper instead.

It is always better to have published papers in the field, so I know that this is worth the effort in the long run, but if this takes too long I may never graduate.  Again.

* A while back I did a paper with someone in the law school and discovered that they basically shop their papers around to everywhere, all at once, and take “the best” [read: prestigious] offer.  I am so jealous of that system.