Pen.eps
Chapter 8: Write

At last, the time is right for you to officially write your thesis/dissertation. If you have completed all the tasks from previous chapters, you will already have a head start on the document you will produce while working through this chapter. For instance, you should already have a strong draft of the description of your research problem and your methodology from the work you did in Chapters 5 and 7. This pre-Write stage work could make up a fifth of your thesis/dissertation draft so hopefully this knowledge will dispel any anxiety you may have about starting to write.

You will use most of the research material you have generated throughout your research journey during this stage so try to do the bulk of your writing in the place where you stored your material. All your binders and electronic files should be readily at hand at all times so your flow is not broken by having to stop and refer to a source. If you have family obligations such as young children or a sick parent, try to find alternate care arrangements on a regular basis during this stage so you can have some uninterrupted time to crank out words. So much of writing time is actually thinking time, so interruptions inhibit productivity in two ways: 1) by preventing your fingers from typing on the keys, and 2) by preventing your brain from processing the thoughts that must be translated into typed text. Hiring caretakers, making deals with a spouse or sibling, or even contracting out home and garden tasks during your writing time will be money well spent.

Topic: Writing Your Abstract

Though you will not be able to write the final version of an abstract until after you have a completed draft, try writing a first draft now. An abstract is essentially a summary of your research problem, methodology, and results that fit in a paragraph or two. In for-pay academic journal databases, abstracts are normally viewable for free and allow a reader to decide whether to buy copies of the articles. Abstracts are also helpful during catalog searches at the library because words from abstract text are used for electronic indexing and searching. Some abstracts do not accurately reflect the contents of the document. This can be frustrating, especially if the reader had to pay for access rather than viewing it as part of a library’s subscription. To prevent this kind of disappointment and preserve your fledgling reputation, write your first abstract draft by stating:

If you cannot complete this exercise yet, do not worry. You will be able to by the time you complete your first thesis/dissertation draft. The attempt will get your mind flowing down the stream of logic needed to write your draft.

Revise and update your keyword list

Hopefully, you have been revising and updating your keyword list as you go. During some stages, you may have decided keywords were not pertinent to your study and placed them in an archive location. During other stages, you may have added keywords to help you discuss your findings. If these alterations are only stored in your head or informal jottings in a journal, take the time to make these alterations in your formal keyword spreadsheets as well. Be sure to include all the sources you have that fit under each new keyword, even if those sources are already listed under another keyword.

Review your question/hypothesis daily

Your research problem is the key to everything you have done on this thesis/dissertation journey. It is important to keep it at the forefront of your mind while you write so every statement you make, every logical progression of thought, and every illustration points toward your research problem. If you have not done so already, write out the official statement of your research problem in the same form (question or hypothesis) that you used in your approved proposal. Post it on your bathroom mirror, above your computer monitor, or even on the backside of your car’s sun visor. Read it and reread it whenever you can to stay focused and primed for writing at all times.

Literature: Stick to What You Have

Although your search through your discipline’s literature is never officially over, the intensity of that search greatly decreases as you progress through the stages of your thesis/dissertation. During the Write stage, it is critical to stay focused on writing and not let your mind trick you into avoidance with one more trip to the library to read. You have already gathered more sources than you need. Extra reading at this stage has the potential to confuse and sidetrack you. Unless your adviser says you must read a particular source before going any farther, resist the urge to read and just get your writing done. However, this prohibition does not mean you have no literature-related tasks right now.

Update your literature review

One literature-related task you have is to update your literature review. Pull out the version you wrote for your proposal. If you did not have to write one for an informal thesis proposal, pull out whatever you did write as an introduction to your proposal. This section/chapter should already be written in past tense because it is describing research already completed. Check over the sources you covered. Are all of them still pertinent to your research? Are there new key and supporting sources that are not covered? Use your revised keyword lists for help in selecting sources to add to this review or to your introduction. Make sure this section/chapter accurately reflects all the background information you needed to conceive, design, conduct, and analyze your research. Diligently updating your literature review now will yield a complete draft of another section/chapter, as well as priming you for writing the sections/chapters you have yet to start by facilitating recall of sources you will want to cite.

Review the theses/dissertations you previously selected

Another literature-related task you have in the Write stage is to review the model theses/dissertations you selected in Chapter 2. This does not mean you must read these documents in detail again as you did when developing a thesis/dissertation style guide. Instead, read them quickly for:

You may want to make an outline of your own thesis/dissertation that mirrors the organization of your model documents while reviewing them. This will ensure you do not miss a section/chapter in your first draft and will likely add proper diction to your headers and main points. This outline does not have to be a detailed, five-level affair. But, setting up your section/chapter titles with subheads and a few bullet points underneath will give you a head start while your brain is already in an organization/focus/diction mode.

Assistance: Consult With Caution

Try to work as independently as you can during this stage. The writing of your thesis/dissertation demonstrates that you are qualified for the degree you seek and that you deserve to be included in academia. The worst thing you could do right now is ask for so much help that you have no words of your own recorded in your final document. This inclusion of others’ words passed off as your own is sometimes done purposefully, but you would not be reading this book if you were planning on tricking/paying someone else to write your thesis/dissertation for you. Instead, the thing you want to guard against is an unintentional, subconscious parroting of discussions you have about your research with other graduate students or committee members. Even in the world of academic diction, there is such a thing as voice. Your academic voice will differ from another person’s, however slightly. Because your committee will probably know most of the academics you know, it would not be a good idea to slip into someone else’s voice and be accused of not doing your own writing, or worse, your own thinking.

Adviser and committee

Although you should work independently, do not work in isolation. Make regular appointments with your adviser to look over your first draft as it takes shape. This will not only show your adviser you are making progress, but it will also give him or her a chance to steer you back on course if you begin to veer off track. Writing off topic, making faulty conclusions, or even writing in a nonacademic voice are all rookie errors that are easily fixed when caught early. But, if your adviser begins talking about new questions, remember that now is not the time to explore anything new. Gently propose that you add these new questions to “future study suggestions,” and come up with a plausible reason why this new avenue is outside the scope of the original — and long-time approved — research problem.

When your adviser approves, consider distributing an early — second or third — draft of your thesis/dissertation to your committee members. This way, they can begin to orient themselves to your research well before your defense. They may have questions or concerns above and beyond those of your adviser that will need to be addressed before you distribute your defense version. Taking care of these initial revisions now in an informal setting will decrease the number of final revisions you will need to make post-defense. It will also cut down on the number of challenges you will have to face verbally during your defense.

Extra help for ESL students

If English is your second language and you have not yet joined a thesis/dissertation critique group, now would be an excellent time to join one. These peer critiques by graduate students at the same stage as yourself will be invaluable tools for polishing your thesis/dissertation prior to sharing it with your adviser. Keep in mind that as part of a critique group, you will be expected to give constructive feedback on others’ papers. Do not worry if you are not strong in highly technical areas of grammar. Someone in the group will be. Just make sure you contribute to the group critiques somehow. You might be adept at catching areas of inconsistency or flagging places with confusing sentence structure. Do not assume you are failing to understand a passage because English is not your native tongue. Flag it and ask. Even if the passage is technically correct, the other graduate student may choose to revise so his or her exhausted committee members will not misinterpret it.

Spending time conducting group critiques will not only increase the proficiency of your written work, but it will also increase your proficiency with English as a whole. During these discussions, it is important to make sure all your remarks are directed at the work and not the writer. Rather than taking criticism of your work as a personal insult, do whatever it takes to receive the criticism as valuable assistance. The goal is to make written work better, not create an interpersonal hierarchy.

If your university does not have a critique group, it may have a thesis/dissertation support group where you could find someone with strong writing skills who would be willing to critique your writing for a per-page fee. You might also be able to find writing assistance from an on-campus writing center or an international student center. Critiques of academic papers take a great deal of time to conduct, so break the job into sections. This will give you critiqued and revised sections to present to your adviser at regular intervals. It will also give you a chance to switch the person critiquing your manuscript if the first person does not work out.

Data: Communicating Your Findings

After all the data collection and analysis you just completed, you know what your data is and what it says. You are the one spending the most time with it, so own the fact that you are the expert in this study. Even if you had a data consultant run portions of the analysis for you, you have been the one looking for the meaning behind the numbers and trends. Keeping your expert status in the forefront of your mind will alleviate some of the anxiety that accompanies the writing stage. Nothing halts writing progress like anxiety, and for many people, nothing produces anxiety like math-related tasks. Remember that you have already done all the collecting and analyzing; all that is left to do now is the communicating.

Key findings

During the Analysis stage, you made an outline of your findings. Retrieve that outline now and review it. Refreshing your memory on the findings you need to present before writing the bulk of your thesis/dissertation will facilitate composition. It also gives you one last chance to run questions past your adviser before launching. You do not want to get entire chapters written only to find out you will have to start from scratch because of a faulty premise or inaccurate understanding of analysis output.

This outline of your findings is not the outline of your thesis/dissertation. You may be required to write multiple chapters, presenting findings from one outline section in one chapter and another outline section in another chapter. However, this outline may be re-used during your defense for the “Results and Conclusions” section of your presentation, depending on how your department expects defense presentations to be organized.

Computer-generated graphs and tables

Hopefully, you and your adviser have come to an agreement about the visual presentation of your data and analyses. If not, finalize your sketches before moving on. Take out your filed sketches and captions, and review the variables, factors, and potential output they are expected to convey. Computerized versions of your visuals should be done in such a way that they can be used both as figures in your thesis/dissertation document and as slides for your defense presentation. Generating graphs with spreadsheet software will produce output that can be easily embedded in word-processing software from the same company (for instance, Excel into Word, both from Microsoft). Saving MYSTAT graphs as JPEG files, which can be read by numerous software applications, will allow you to import or simply copy-paste graphs into word-processing documents and digital slideshows. Building your tables in a word-processing application will allow you to easily embed them in other word-processing documents of the same format along with their sister slideshow applications.

Software

Chances are you have already built tables within a word-processing program, so the following review will be brief. Pull out a hand-written table that you need to computerize, and open a new word-processing document. You will copy this table into your thesis/dissertation manuscript a little later. But, tables have a funny way of messing up the formatting of surrounding text so finalizing the layout of a table in a separate document — with the option of deleting and starting over, if need be — is a much less frustrating way to construct one.

1. In Microsoft Word and OpenOffice Writer, select Table > Insert > Table.

2. In the dialog box that pops up, choose how many rows and columns you need. Be sure to add a row and/or a column to house any headers.

3. Enter your headers and data in the appropriate cells so it looks like your sketched table.

4. Find a way with bold, borders, and/or shading options to set your header rows/columns apart visually.

5. Save this file with a name that indicates the figure number assigned to it and a keyword or two; for instance, “Figure6-Table-LeavesPerStem.”

If you used MYSTAT in Chapter 7 to generate your graphs, check that you have a JPEG version of all your graphs before moving on. If you do not, take a moment to open up your output files in MYSTAT, and do a “Save as” procedure to make the conversion — the bottom dropdown box allows you to select a new file type. If JPEG is not listed as an option, look in the left-hand column for the particular graph you want to convert. Double-click on its name, and it should bring up a new tab with just the graph. You should be able to save this stand-alone version of the graph as a JPEG.

If you used a spreadsheet, as suggested in Chapter 7, to generate your graphs, you should already have embeddable graphs ready for your thesis/dissertation. Simply click on the graph’s tab within the spreadsheet, and then:

1. Right-click on the graph so the entire image is selected.

2. Select Edit > Copy in the spreadsheet program.

3. Place your cursor in the appropriate place within your thesis/dissertation word-processing file.

4. Select Edit > Paste in the word-processing program.

You should now have a graph image embedded in your thesis/dissertation waiting for a figure label and a caption.

Finalized captions

Review your model theses/dissertations to see if their captions are placed just below their associated graph/table or on a facing page. If placed on a facing page, the formatting required is fairly simple. Merely type the text on a separate page in your word-processing document, and hit the enter key enough times above and below the text to center it vertically on the page. Begin the caption with “Figure [x],” where [x] is the number belonging to the visual. Check your models to see if the font for “Figure [x]” should be rendered in bold, italics, or both. The text of your caption, however, will likely be in standard font and single spaced.

If your models show the caption directly underneath a graph:

1. Open the word-processing document you would like to embed the graph in. Place the cursor in the location you would like the graph to appear.

2. In Microsoft Word or OpenOffice Writer, select Insert > Picture > From File, and navigate to the JPEG version of your graph.

3. Click on the graph file, and select “Insert” or “Open.” The graph should now appear as an image in your document.

4. To add a caption, right-click on the picture, and then select “Caption.” Fill out the dialog box with the caption text and location.

5. Click “OK.”

6. Your figure label and caption should now appear below your graph. If you do not like the way it is formatted, you may use the font formatting options just as if it were regular text.

If your models show text directly below tables, you may simply copy and paste the table into the appropriate place in your word-processing document and then type the text below the table. Be sure to format the figure label and caption the same way you formatted the graph’s labels and captions.

Consistency is the main consideration here. If you have something formatted slightly out of step with preferences, it may not cause much bother if the formatting is consistent throughout the document. Even if you make a major faux pas and your adviser insists on corrections, with consistent formatting you will automatically know to make the corrections on every figure rather than having to study each one to figure out if it needs adjustment or not.

Privacy considerations

Review the privacy considerations you outlined in Chapter 4. If you have study subjects or locations that need protection from public attention, now is the time to make those alterations in your text. Before making these alterations, be sure you are totally finished with your analysis and know what your conclusions are. If you start using code names before you finish with your analysis, you may get yourself or your adviser confused during discussions or even mix up data sets. Your defense will be difficult enough without the gut-wrenching realization that you have mixed up your data and your conclusions might be flawed.

Keep a spreadsheet or paper list that records the real name for every fictitious one. Keep a printed copy of this record with your data so if your adviser or a committee member asks you a question about a particular data set, you can quickly find which data set to search for the answer. This list should also accompany you to your defense for the same reason.

Anonymity

Depending on your research design and study subjects, you may need to make your subjects anonymous before sharing data about them. Setting up fictitious names is one way, for example, in case-study research. Be sure the fictitious names do not sound too much like the real names so people will not be able to guess the real identities. Get a baby-naming book from the library or visit BabyNamesSM (www.babynames.com) if you are stuck for names that do not rhyme with the original and have a different syllable count. Be sure you have replaced every instance of the original name in your final document by doing a find-search:

If you need to make entire groups anonymous, consider creating group numbers or group letters. Be sure these group symbols will not be easily deciphered, such as using a teacher’s initials as the code for his or her class.

Protection of research site

If your research site is private property or able to be damaged by too much public notice, consider describing the site in such a way as to protect it from interference. For instance, if you have been studying an endangered plant in a large state park, your research site should be described as “[name] State Park” rather than “in the meadow near the fork between [x] trail and [y] trail of [name] State Park.” Otherwise, you may get many local college and high school students going out to “see” it, trampling it to death before it can set seed for the next year’s population. Similarly, if you obtained permission to collect data on private property, chances are the owners will not take kindly to a flurry of new requests or flat-out trespassing. Describe these kinds of sites with enough detail to show why they were selected for your data collection but not with enough detail to allow hordes of admirers to ruin it for future researchers.

Written Work: Unifying Your Writing

You have outlines, lists, data, analyses, graphs, tables, and sources. Now, you get to collate all these items into a unified whole that contributes one new piece of information to your discipline. Intimidated? Hopefully after all the preparations you have made, you are not intimidated but rather looking forward to the challenge instead.

Complexity and length

After reviewing the model theses/dissertations during all your preparations, you should have a good idea of the expected complexity and length for your own document. Disciplines and departments vary, which is why your best guide is successful work from your own department. To get a feel for the difference between the two types of documents (thesis versus dissertation), some general guidelines follow.

Thesis

A thesis is normally 30 to 50 pages plus figures, front matter, and back matter. Ultimately, though, the fewest number of words needed to communicate the reason for and results of the research should determine the length. Content should demonstrate an understanding of the topic and the entire research process. It should be written in clear, concise, and logical language so the author can show his or her ability to engage in interdisciplinary discussions.

Dissertation

A dissertation is normally around 200 pages of material, but this number of pages can vary widely, depending on the subject of the paper. Like the thesis, however, the length truly is determined by effective communication of the reason for and results of the research. Brevity, even at this stage of scholarly writing, is desirable. Content should demonstrate an area of expertise within the topic and mastery of the entire research process. Since a dissertation will cover multiple facets of the research problem, organization and focus are the key to writing in a clear and concise way so readers — your adviser, committee, and the academic community as a whole — can follow your logic and concur with your conclusions. The document should set the stage for the author to engage in interdisciplinary discussions, as well as informative discussions with the public.

Style guidelines

As with your written proposal, you will want to use your department’s preferred style while writing your thesis/dissertation. This style is not meant to override your natural writing voice, but it is meant to infuse your voice with an academic tone. It is also meant to allow readers to easily access your content because they will already understand the “rules” of the document and can focus all their energy on the content rather than the delivery. As a reminder, these rules involve:

Review the model you built in Chapter 2 from previous theses/dissertations. This is your primary guide. A secondary guide will be whichever professional style manual your department prefers. The tertiary guide for your thesis/dissertation will be your adviser’s eye, so try not to depend too heavily on it.

Contents

When preparing to write the bulk of your thesis/dissertation, leave the possibility open for writing it nonsequentially. Some sections will come easier to you than others, especially those you drafted in the Proposal stage and revised in the Analysis stage. Take a bit of time now to make a rough outline of your thesis/dissertation using the headers, listed below, as your first level.

Your university may use different words for the sections listed above. If so, use the ones your adviser will prefer. Also, the order of these sections may vary depending on your department’s preference. Construct your outline accordingly. The important thing is to draft a functional outline so you can skip around from section to section, working on whichever section your mind is ready to tackle.

Once your outline is complete, it is time to write. Make sure to use proper citation techniques in every section/chapter because attempts to fill in missing citations later will place you in danger of unintentional plagiarism. Choose the composition medium that works best for you, be it paper or electronic. If you compose in a word-processing program, then revisions will be easy to make. If you compose on paper, composition may take place anywhere without the heat and whir of a hard drive, though you will have to be diligent about making copies and computerizing the text as soon as possible. The writing stage may take you weeks, or it may take you months. But, as long as the end result is a successfully defended thesis/dissertation within the university time limits, your rate of composition does not matter.

Case Study: Writing Right

Linda Morales

Assistant Professor of Computer Sciences

University of Houston at Clear Lake

Biography:

Linda Morales is an assistant professor of computer science at the University of Houston at Clear Lake. She received her doctorate in computer science from the University of Texas at Dallas. Her research interests include: the design and analysis of algorithms, security for networks and multicast groups, information security education, and ethics in computing and information security. She teaches courses in algorithms, data structures, and security. She has been involved in several funded research projects.

Dissertation title and synopsis:

Pancake Networks and Pancake Problems

This work used low-dilation embeddings to compare similarities
between star and pancake networks and between binary hypercubes and pancake networks. In addition, a quadratic lower bound for the topswaps function is exhibited. This provides a nontrivial lower bound for a problem posed by J. H. Conway, D. E. Knuth, M. Gardner, and others. We describe an infinite family of permutations, each taking a linear number of steps for the topswops process to terminate and a chaining process that creates from them an infinite family of permutations, taking a quadratic number of steps to reach a fixed point with the identity permutation.

First-person account:

I encountered considerable personal difficulties that delayed my work by several months. The difficulties had to be resolved before I could finish my research and write the dissertation. The problems were such that it was not possible to devote time or mental focus to my work until the personal difficulties were resolved.

Once I was able to write the dissertation, I became aware of the challenges of communicating difficult concepts through the written word. This is not an easy skill to learn. Mastery of this skill requires practice, an intimate technical understanding of the topic, and an understanding of how to break the topic down into subtopics that can be grasped by readers. These skills are essential for writing articles in peer-reviewed journals.

One tip I have for graduate students is this: Before starting to write, clearly identify the main points you must communicate to the readers of the dissertation. Keep these main points in constant focus during the writing process.

Each main point will probably have to be broken down into subtopics in order to explain concepts clearly, and again, the writer should focus on clearly communicating the subtopics to the reader. The selection of topics and subtopics should be periodically reassessed to determine if they are appropriate and if the words are effectively communicating the concepts.

Finally, keep your writing style clean, focused, and simple. Keep your sentences short — no long or elaborate sentence constructions. And don’t force the readers to have to look up lots of unusual words. Stick to common vocabulary, except when the nature of the topic requires specialized vocabulary. Particularly for scientific papers, you want your readers to focus on the scientific concepts in the paper and not make them have to parse difficult sentence structures or scratch their heads wondering what the subjects and predicates of your sentences are.

Organization: Stick With The Routine

It is often tempting while writing to pull out stacks of reference material and leave loose sheets out in piles for quick and easy access. Resist this urge whenever possible to keep the risk of losing or ruining your papers at a minimum. Putting away the excess papers has another advantage: Physical movement can often jar new thoughts loose, so you may find these filing pauses make you more productive than had you stayed glued to your chair staring at your monitor.

Scheduling regular appearances

Because most of this writing will take place at your home, you may feel less inclined to commute to campus. But, regular appearances, though time consuming, will help you successfully navigate the writing stage. These appearances may include:

If you are writing your thesis/dissertation as part of a low-residency graduate program, these opportunities for visibility may not apply. You can, however, check in with your adviser on a regular basis and apprise him or her of your progress. The goal for this visibility is to allay any concerns your adviser may have about your ability to successfully complete your degree or the opposite and also to remind your adviser you still exist and will need his or her input soon for revisions and defense.

Tracking multiple drafts

Once you finish your whole document, or possibly distinct sections of it, you will be required to submit your work to your adviser for review. He or she will suggest revisions. Some of these revisions will be substantial; others will be minor. Make sure to view all revision requests as part of your training to be a successful participant in your discipline. In order to avoid confusion between the multiple drafts you will generate during this process, find a way to distinguish one draft from another. Whether with different colors of paper or different page footers, make sure the distinction is readily apparent so that as soon as your adviser (or committee member) pulls out the draft, you can tell which draft it is. If you do not do this, you may end up in arguments or circuitous discussions because of literally not being on the same page.

You might also consider tracking which drafts have been distributed to each member of your committee. This can be done electronically in a spreadsheet or simply on a piece of paper stored in your binder system. Number the drafts sequentially in the order they are presented to your adviser. This means your adviser will have every draft distributed to him or her, while other committee members may only see every third draft. Track both paper and electronic drafts you distribute. This way, you will have a record of the draft, person, and date in case of any disagreements or confusion during committee meetings.

Maintenance

Before moving on to the next section, make sure your organization and disaster-recovery systems are up to date. As an extra precaution, make sure you have paper and electronic versions of your current drafts stored in multiple locations. Mail paper drafts to a grandparent. E-mail PDF drafts to a cousin. Load word-processing drafts onto a memory card or flash drive, and store them in a fireproof safe. Do whatever you have to do to ensure that, barring a nationwide nuclear attack, you will always have access to a current draft of your document.

Summary of Tasks

Phase 1

Phase 2

Table of Contents