RESEARCH PROPOSAL PROJECT

-Forests, Fall, 2018

This is a multi-stage project, and is the central assignment for this class. The end-product  will be a paper, but it's not the usual sort of 'term-paper' where you summarize what you can learn about a topic from existing literature.  Rather, it's a proposal and design for a piece of primary research addressing  a question of your own formulation through original observation, measurement, experimentation.  You will get there in several stages -- the same steps that might be followed by a research scientist at any stage of their career.  You'll spend the most time developing a viable research question/hypothesis; that is, in fact, the hardest (but most interesting) part of science, but you'll also develop conceptual context and background (a literature review), a design for gathering data, and a discussion of how you'd interpret results. 

The product of each step should be submitted as a shared google doc (or, if you prefer, a shared document in some other word-processor format).  Some steps might involve graphical elements -- sketches of curves or diagrams of methods or graphical models of relationships.  This is good. If you have trouble figuring out how to incorporate these in a file for digital submission, it's fine to submit a paper version.  FOR ALL ASSIGNMENTS, include your name in the filename(s) of the submitted work (e.g. john smith - proposal step one)

STEP ONE:  Five Questions: DUE 17 SEPTEMBER.

    The first task is to develop some interesting and workable questions.  This is not as easy as it seems.  Here are some guidelines:

• Begin from your own direct observations of ecological systems -- any ecological system is okay, really (forests, meadows, streams, lawns...).  We'll do some of this in class/lab, but you should spend more time on your own  simply exploring with an inquisitive eye.  Make this explicit: "I saw such-and-such phenomenon/pattern, and it led me to ask..."
• Ecological questions can concern the distribution, abundance, diversity, or behavior of organisms, their population dynamics, or their relationships to the physical environment or interactions with one another. 
• It is better to begin with simple, restricted questions.  You might be motivated by 'big picture’ questions ("What determines whether a stream has brook trout in it?"), but research has to be framed around answerable questions; find ways in which your general interests can be manifested in particular instances, exemplary scenarios ("Are small streams more likely to have trout if they run through forest?").
• Keep questions simple, but not trivial. They should be about patterns, not about particular instances; answering them should allow you to make predictions. It’s not very interesting to ask “Why did that squirrel run up that maple tree yesterday morning?” because, even if you could answer it, the answer wouldn’t speak to more general patterns very clearly. On the other hand, “Are squirrels more likely to run up maple trees than beech trees?", or “Are squirrels less likely to run away if they’re eating something really good?,” or "Are there properties of tree 'structure' that make squirrels more likely to run up them?" are questions whose answers might tell you something more general about squirrels.  
• Questions need to be answerable; another reason to avoid questions that are too broad. 
• BUT ALSO, avoid questions where the answer is likely to be easily looked up ('How tall do maple trees get?), or where the answer is a trivial count or measurement ('How many maple trees are there in the woods behind Jennings?'); BUT each of these could be turned into interesting research questions ('Is the height of maple trees affected by properties of their environment? By squirrels running up them?').

    Frame five possible research questions based on your observations in the field (it is fine to use things we’ve seen or discussed in labs, on or off campus, or in your own independent observations). 

    WHAT YOU SHOULD HAND IN for each question is 
    a) A sentence or two about the observational context that led to the question, and
    b) A concise statement of a question springing from these observations (one or two sentences).

    You do NOT need to explain how you'd address these questions (measurements, sampling scheme, etc.); that will come later.  BUT you should be able to at least imagine that the question is answerable through well-designed experiments and observations.  Can you generate potential answers to these questions (alternative hypotheses)?

STEP TWO: Three Questions Refined: DUE 11 OCTOBER.  

    Refine your research question possibilities to THREE questions, each with at least one or two briefly stated hypotheses/predictions (i.e., plausible answers).  Use the same guidelines as above to help think about what makes a good question.  You can refine/modify questions from your first set, or add new ones or both.
    Your possible answers/hypotheses might be folded directly into your questions (“Do maple seedlings do better at higher soil pH?”; the possibilities are 'yes' or 'no' -- BUT, you should probably say something about WHY you think they might!), or they might be stated separately (“What environmental factors influence maple seedling growth and survival under the canopy? H1: they do better at high soil pH; H2: decaying cigarette butts enhance seedling growth conditions...” Note that these hypotheses don't fully explain why -- the mechanism behind your expectation.  That's okay; take things a step at a time!). Even though you may start with a specific 'guess' about what’s going on, it is always important to think about plausible alternative answers/hypotheses.  Add to the criteria in Step One for what makes a good (i.e., interesting) question/hypothesis: 
       a) A good hypothesis is consistent with the data/observations you already have
       b) A good hypothesis is testable -- you can think of observations/experiments that would produce one kind of result if the hypothesis were true and a different result if it were not true.
       c) A good hypothesis is generalizable -- it leads to expectations of what you should see/find in addition to the initial pattern that caused you to think of it...   
    For example, “There are around 20 woodpecker nests on campus" is not a very interesting hypothesis, because, even though it could meet the first two criteria, it misses on the third.  But if you changed it to, "Is the density of nesting woodpeckers on campus related to noise levels? [or some other factor...]" you could imagine a range of  implications and follow-up questions. 

    WHAT TO HAND IN:  For each of the three questions, a paragraph or two, including objectives of step one (refined) PLUS laying out of hypotheses and predictions.  A page or two total, probably.

STEP THREE: Two Questions with basic Research Design.  DUE 1 NOVEMBER

Choose two of the questions from the previous step, and FRAME A BASIC RESEARCH DESIGN FOR EACH. Develop a paragraph or two about how you would address your question (test your hypotheses);  Focus on what predictions emerge from your hypothesis, and the kinds of data/information you'd need to test your predictions.   It might look something like, 'If my hypothesis is correct, I would expect seedlings surrounded by cigarette butts to grow faster or be less likely to die.  I could either observe seedlings near doors where there are a lot of butts and compare them to seedlings that aren't surrounded by butts, or I could experimentally place butts around some seedlings.  If I'm right, I'd expect....'  (Expanded a bit, of course.)  The next stage will focus on developing this part of the proposal and doing some background research, so don’t worry too much about getting it all ‘right’ or in full detail yet.

    WHAT TO HAND IN:  A paragraph or two for each of the questions, in addition to material from Step Two (refined) should be enough.  You might want to start doing some literature research as you are thinking about this stage; it will be called for in the next anyway.

STEP FOUR: Background Research, Review, and Annotated Bibliography DUE 19 NOVEMBER .      

Choose ONE of your (now) TWO questions, and do some background research in the primary research literature (It's possible to design and execute interesting research without having any familiarity with what's already been done -- but this isn't the way good science usually gets done).  We will talk about how to do this, but you can start any time.   Three things to think about (we'll address each in class):
    A. Finding relevant papers: Useful papers may not be about the exact, specific question you're interested in but bear  on similar situations with different species, or similar processes in different circumstances, or on methodological approaches applied in other scenarios.  It won't always be possible to find papers that are precisely and generally parallel to what you're proposing.  If you wanted to study the effect of bird-dispersed seed on distribution of wild apples, you might look at papers on bird behavior, on apple seed germination, on measurement of seed dispersal in general, etc.  Depending on your project, you might end up with very few closely relevant papers, or you might have lots and end up picking out only the few most relevant/helpful.
    B. How you read research papers: You don't read this stuff like a novel!  You will not be able to follow many of these papers in full detail, but it's usually not necessary to do so. What is important is: first, understanding the basic research question and overall approach to it; second, following the logic of the study as it concerns your project; and, finally, thinking about the assumptions, unanswered questions, follow-up needs (ALL research papers have these).
    C. How your reading integrates into your research plans: Your research design is likely to evolve as you read, either because you get new insights into questions/hypotheses, OR because you realize your question is (partly?) asnwered, and you  need to revise it to address some 'next' question or facet.  This is fine.

    WHAT TO HAND IN: An annotated bibliography that includes, for each paper you've found useful: 1)  full citation (use format like those in the papers you're reading), 2) a  brief statement of the primary research question, approach, or findings, as they bear on your proposal, and 3) a brief explanation of how it bears on/contributes to your study.  

    Some entries might be very brief (e.g., 'explained a method of following seed movement by coating seeds with fluorescent powder...'); others a paragraph or two.  There's no 'required' number, but shoot for somwhere between 6 and infinity.  FINALLY, note that this is a 'working' bibliography; you'll continue reading and exploring the literature in preparing the final product.
  
FINAL PRODUCT: A Research Proposal DUE 10 DECEMBER.

The final product of this project (and the 'biggest' piece of work for this class) will be a fully developed research proposal -- a paper that lays out the background and design and expectations for a study that would address your research question, test your hypotheses...  In structure, the proposal will look very much like the research reports you have read in your literature research and reviews and for class.  It should have:
    - an Introduction that lays out conceptual background and defines your question/hypothesis.  Where did the idea come from, and what are the most important/interesting connections to larger body of understanding and concept?  You might subdivide this section and have an expllicit 'Literature Review' sub-section.  You should wind up this section with a clear, succinct statement of the hypotheses and predictions shaping your research design.
    - a Methods section that explains exactly how you would address your question, test your hypotheses/predictions through further observation, measurement, comparison, experimentation.  It is particularly important that your logical development in these two sections make clear how the Methods you propose are connected to the questions/hypotheses.  Explain how you'd digest/summarize your results.
    - a Discussion section that considers how you would interpret possible results of your Methods.  "If my results look this way, it would mean ....  If they look that way it would, on the other hand, mean...."  You might wish to use graphics here (or elsewhere) to show patterns that you would be expecting or not expecting; these might even take the form of 'mock-up' results.  If you use graphics, refer to them clearly in the text.  You should also consider, here, what further research problems/questions might be suggested should you see particular results.
    - a References Cited section: Throughout the proposal, cite published information as appropriate to build background and context, recognize prior work, support methods, etc..  The publications cited must then be listed fully in this section.  Use a format for citation and reference that you've seen in one of the primary research literatures sources you've used.