On a Roll this Morning...

I awoke at 3:15 this morning (yes, a Saturday morning) and the research problems were right there in my head.  So I got up to take advantage of that moment of "brilliance."  Now I need to write my null hypotheses.  

Thought the lost-socks-due-to-alien-burglary anecdote was cute (see previous post), I need a little more instruction.  Dr. S explained that a null hypothesis is:  "there is no difference between X and Y."  I found this on ehow.com - How to Write a Null Hypothesis.  

"A null hypothesis is defined as a statement of no change.  That is, a null hypothesis describes an outcome in which there is no statistical difference between the groups you are comparing [Dr. S's no difference between X and Y].  More specifically, if a null hypothesis is fulfilled, your original hypothesis is rejected."

Summarized:

1. Write the hypothesis (i.e., what I believe to be true).  Let's say my belief is the old adage:  You can't teach an old dog new tricks.
2. Turn this into a statement that indicates no difference between experimental groups:  Old dogs and younger dogs can (or can't) learn new tricks.

Missing Socks and Alien Burglary - Null Hypotheses

In the May 14 conversation with Dr. S, he pointed out that my hypotheses weren't written in a consistent form.  He said that I needed "null" hypotheses.  Oddly (or maybe, fortuitously), I found a website called www.null-hypotheses.co.uk by a group called the Journal of Unlikely Science.  In it, the null hypothesis is explained in comical terms:  What is a Null Hypothesis?  The example is:  the loss of my socks is nothing to do with alien burglary.

So two things to do this long Memorial Day weekend:

1. Rewrite my research problems (tightening up the scope of the ones I've written previously), and
2. Write new hypotheses - ensuring that they are null hypotheses.

Round 1 - OK

I had my proposal (version 1) review with Dr. S Wednesday.  Few surprises as I had already figured out a lot of what he commented on.

The take-aways:

1. As I've written before in this blog, my research problems really need some attention.  Dr. S says (and I agree!) that as outlined, the research behind the problems would, in his words, "take about 75 years."  OK, well (I suppose), better to have too many rather than too few.  My research problems are more themes of problems rather than individual ones.  I just need to drill down on one or two, then "discover" the problems from those themes.
2. Subsequently, my hypotheses need to be re-worked, based on the new set of research problems, of course.  Besides, Dr. S pointed out that my first round of them are written in inconsistent form.  I didn't see that, so that was new.

Biggest issue:  I don't have primary, but secondary, data for my study.  What that means is that the data I have has been gathered by others (secondary data).  That's a problem.  I need to gather data myself (primary). But we are both scratching our heads a bit on this one. Dr. S is a qualitative guy and can counsel on surveys, interviews, etc.  But any surveying I could do would be of current or potential customers, and that would be a no-no professionally.

So, with that revelation, Dr. S decided to check in with a man who I think will be my thesis adviser.  He earned his PhD at Northwestern but is a professional in finance in Manhattan not an academician.  He thinks that this man - Dr. H - will be able to guide me better than he can.

In the meantime, Dr. S said to not trouble with another proposal draft until I've spoken to Dr. H.  But he agreed that it wouldn't hurt to continue with the literature review.

Traveling for work this week, but that's what I'll do next week.

Thinking Too Much?

This is the week that I'll be talking with Dr. S about version #1 of my proposal.  In the meantime, I continue to work (re-work, and re-work) my research problems.  At this revision stage (and no telling what Dr. S will say about the research problems), I've written and rewritten the research problems...now on my third major revision.  My conclusions, thus far, are these:

1. Research PROBLEMS arise from the literature review; while
2. Research HYPOTHESES are solved by the analysis.

Now, I have absolutely NO idea if this is correct or not, but it essentially means that the literature review (Chapter 2 of five total chapters) has to be complete before the research problems are "fully cooked."  And the analysis done in Chapters 3 and 4 has to be finished in order to prove the hypotheses.  As Chapter 1 is just a write-up of the proposal outline, there's a circular problem:  I have to do all of the analysis and literature review for the dissertation in order to get my proposal outline approved.  So I'm writing a dissertation before I even know that the thing is approved?  This doesn't make any sense.  Or maybe I'm thinking too much.

Proposal Part 4: Research Objectives/Hypotheses Related to Research Problems

I really love YouTube.  

A simple four-minute video explains how to write research questions and hypotheses from "MsLemmonsClass."  I assume this was written for high school kids (though I never saw anything like this in high school).

I like it.  This really makes the task look a bit less daunting.

So basically, you take each individual research problem and find the independent and dependent variables.

Hypothesis:

The hypothesis is my prediction, my claim.  It is what I think is going to happen based upon my prior information and my prior observations.

Written Format:  IF-THEN-BECAUSE

IF:  If the IV does something (increases, decreases), 

THEN:  then DV is going to (increase, decrease)

BECAUSE:  because ___________.  Why do I think this?  What led me to this claim or hypothesis?

This made writing the hypotheses fairly easy to do.  Got them done in a couple hours.