WHAT MAKES FOR A GOOD RESEARCH PROPOSAL: TEACHER RESPONSE TO HOMEWORK #5

Grades on Canvas.

Speaking the language of your audience

Research proposals have to reach an audience, and often an important one. You present a proposal to your Master’s or Ph.D. adviser, your boss, or a potential funder. The goal of any proposal is to sell this decision maker on your research, that: it’s necessary, it’s a good idea, and you are equipped to do it.

This means that jargon is a bad idea, unless you know for certain the audience will understand it, which is always risky. Here’s a link to a Plain Language resource on jargon to help you avoid it: http://www.plainlanguage.gov/howto/wordsuggestions/jargonfree.cfm

Solid precedents

A lot of people who are new to research proposals think that presenting something that is entirely new is a good idea. This is rarely the case. If no one has ever done a research project like yours, chances are you’re going too far out on a limb. Either that, or you haven’t found other research that is like yours.

Precedents are research studies that you’re building on, and they are important. Mine your literature review for birds of a feather: researchers that you want to imitate. Showing that you are aware of other studies like yours also shows that you’re building your study on a solid foundation, rather than creating an entirely new methodology that may or may not work

Solid research design

Along with precedents, you need to make sure that the overall design of your study conforms to best practices. These best practices can be found in your precedents. They are the things that experienced researchers do with studies like yours. Look especially to the ways that more experienced craft research questions, explain their own precedents, and talk about how they used their research methods to collect and analyze data and . These three areas are the ones that new researchers tend to have the most difficulty explaining properly.

Connecting Module #4 to Module #5: Teacher Response to Module #4

Your grades are on Canvas, so check for them there.

I also provide individualized feedback on modules that connect what you individually did to what you need to do in the next module, so be sure to check on that as well.

Now you know two ways of doing research; make them one

Now that you’ve been introduced to both qualitative and quantitative methods, you need to think beyond them. As I’ve mentioned before, researchers who just think of themselves as one of these methods worry me, because I think they’re treating methods as a way to produce findings, rather than as ways to think through complex research problems.

I fell into this trap early on in this class by describing myself as more “qualitative.” As I thought back though, there’s not a single study I’ve done which didn’t involve some quantitative component, whether that was during my data collection, data analysis, or even how I connected one study to the next. Both qualitative and quantitative methods have rich traditions that can be useful in variety of studies. Exploit them both regularly to become the best researcher you can be.

Communicate why you’re doing what you’re doing

In light of the above, you may employ quantitative methods, qualitative methods, or a mix of the two (called “mixed methods” research) in your Research Proposal draft for Module #5. The goal of a proposal is always communication to a decision-making audience, however: you need to translate what you intend to do in a coherent way that is free from jargon and that demonstrates sound research skills.

To do so: think of justifying everything you’re doing for the toughest critic you can think of. That voice in your head that always makes you doubt yourself? Give that voice free reign. Let it critique every procedure, every source, every sentence. Then respond to that critique to make your proposal unassailable.

The feedback you receive from myself and your peers as part of Module #5 will help you do this.

Final steps for module #4: Teacher response to module drafts

The difference between a hypothesis and  a research question

The main thing I’m seeing in the drafts that you need to work on is how to frame a good hypothesis. Quantitative researchers use hypotheses to frame the assumptions that they are trying to assess. So, they typically have a hypothesis as well as a research question. At the same time, however, you don’t need to include a hypothesis in this module. You can if you want to. You do need to be clear, however, what assumptions your study is based on, however.

To use an example from one draft:

Missing in the literature is research that explores the potential and tangible deliverables that technical communicators could provide for community partners via service learning assignments. Administrators and instructors need specific details and information for “the scope of audiences, document types, rhetorical purposes, content, styles and outcomes” to understand why service learning supports superior professional development for students
(Henson & Sutliff, 1998). These deliverables obviously evolve with technology and the demands of the nonprofit sector, therefore the research is inherently ongoing. That said, it is information that is of critical importance for robust preparation of the workforce and to understand the impact of the field of TPC at large (McEachern, 2011). Technical writers have much to offer the greater good, and with the goal of balancing “social conscience with technical learning” in mind, more research will lead to a better understanding of those offerings (Sapp & Crabtree, 2002).

Research Questions
RQ1: What types of technical documents and or deliverables do community organizations utilize to execute their missions?

RQ2: What is the scope of the audience(s) for which a community organization’s technical documents are written?

RQ3: What obstacles prevent community organizations from producing or procuring the technical documents they need to execute their missions?

RQ4: Of the technical documents and deliverables utilized by community organizations, for which ones would they prefer to be trained to produce in-house?

It’s very clear what this researcher wants to study and also the assumptions behind why they are studying this. So, there needs to be a prefatory paragraph somewhere in your study that explains, in laymen’s terms, the so-what behind your study.

We need to know:

  • Briefly, what is the state of knowledge leading up to this? What do we know and what don’t we know?
  • Why is the thing you are trying to assess, the central variable you’ve identified, important for researchers to understand more about?
  • Why will your methods help you collect the type of data that will help you assess this variable?

In qualitative research, you can often adopt a more grounded approach that uses fewer assumptions about your target population. In quantitative research, you have to make certain assumptions in the form of a hypothesis or educated guess based on past research. Again, this has more to do with what a quantitative instrument can detect. You have to be more careful as a quantitative researcher regarding what you are detecting and how.

If it helps you, just come up with a hypothesis, or testable statement on your research.

In the above example, for instance, you could say:

Hypothesis 1: Non-profits experience numerous obstacles when attempting to document their internal processes.

Hypothesis 2: These obstacles prevent non-profits from engaging in effective technical communication practice.

These are testable because they can be either true or false. The obstacles either exist or they don’t. They either prevent effective practice or they don/t The researcher has a hunch the obstacles exist and that these obstacles prevent effective technical communication. The researcher may find the obstacles don’t exist, however, or that the non-profits are overcoming them, which means there’s a reason why this research is being conducted: to see if the hypotheses are true or not.

Whether you use a full-blown hypothesis or not, it’s useful to think on these terms when dealing with quantitative research designs: what assumptions are you trying to test out? How will you know if those assumptions are correct?

Connecting Module #3 to Module #4: Teacher Response to Module #3

Your grades are on Canvas, so check for them there.

I also provide individualized feedback on modules that connect what you individually did to what you need to do in the next module, so be sure to check on that as well.

Ask yourself the “so what?” question

Whenever you create a research instrument, it should arise naturally from questions you are trying to answer. It should provide a natural means of answering those questions, in other words.

This isn’t always what happens, though. It’s easy to think of methods as solutions to research questions. Have a question? Through a survey or interview at it? But why that method? How does that method help you collect data that will be useful to answering your questions?

Answering that last question starts with asking the broader question of why people should care about what you’re researching. This is the “so what” question that every researcher should ask themselves when beginning a new inquiry. And the answer should never be: because you think it’s important. It has to be because you think it’s important to other people.

Don’t think in (quantitative or qualitative) silos

Another trap is to fall into the “I’m this kind of researcher” mindset. I just did that in my post about Homework #4. It’s easy to do.

The danger in doing so is that you start thinking like a robot. Again: you’re putting the method before the inquiry. Just because you’re unfamiliar with a method, doesn’t mean it isn’t exactly you need for a current inquiry. Similarly: you may be the best ethnographer in the world, but maybe ethnography isn’t going to be useful in the current inquiry.

Answering questions, questions that provide solutions to problems, should always be the fundamental aim of every researcher. This is especially true when those questions are hard to answer and when their attendant problems seem unsolvable. That’s where good research comes from: the “wicked problems” that can never be definitely solved.

The Qualities of Good Quantitative Research: Teacher Response to Homework #4

Grades on Canvas

As I mentioned before, I’m mostly a qualitative researcher. I have reviewed numerous quantitative studies and can tell you some of the pitfalls of this discipline from that perspective, however.

The “So What” Factor

It is much easier, in my view, to justify a quantitative research design. As long as your literature review is solid, a hypothesis will naturally arise. As long as this hypothesis isn’t too outlandish, it is probably testable via a survey.

When reviewing a lot of quantitative research, however, I’m left thinking “so what?” I’m left wondering what was really tested. I don’t believe quantitative research can tell us much about human communication, for instance, and I certainly don’t think it can predict how humans will behave in future situations.

In quantitative research, statistics help researchers to prove cause and effect relationships, as opposed to things being up to chance. Descriptive statistics allow researchers to describe data, such as when using mean, median, and mode to describe frequency distributions. Inferential statistics allow a researcher to conclude that relationships exist among variables by using chi-square, t-tests, and f-tests. Quantitative research studies must have validity and reliability. Validity in quantitative research refers to whether or not the experiment actually measures what it claims to be able to measure, and experimental studies need to
have both internal and external validity. Internal validity means that any change in the dependent variable is due to the independent variable, while external validity means that the study’s results can be generalized to outside situations beyond the experimental setting. Reliability refers to whether or not the experiment accurately measures a single element of human ability.

Quantitative research’s main limitation is that it takes place in a highly structured and isolated setting – “unnatural conditions” – and while this allows researchers to draw strong conclusions, this setting makes for a situation that is not always accurate to the way people actually act in the “real world.” – Margaret (section 601)

Research Should Solve Problems

My main critique of quantitative research, however, is a critique of all research: I firmly believe all research should help solve real-world problems. Academics are frequently guilty of doing research for research’s sake. This has a lot to do with the mechanism by which our research is published: peer review. There is no real assessment mechanism for why research was conducted, or for whether or not it had any outcomes whatsoever. The only thing peer review tells you is: does the research meet the norms of the field.

This is a huge problem in a field that is supposed to be linked to a profession that happens outside academia, however. If our research isn’t assessed based on its impacts beyond academia, then TPC becomes simply another academic discipline that only speaks to internal audiences.

Quantitative descriptive research narrows the scope of ethnographies (qualitative studies) to include only the most important variables, quantifies them and then looks for patterns and cause and effect relationships. It’s different from qualitative research in that its focus is more detailed and specific than a broad, overall description; it includes a statistical analysis of isolated variables. Unlike experimental quantitative research, it does not use a control group and there are no treatments applied. (p. 82)

Quantitative descriptive research could be considered a “sequel” to a qualitative study – the qualitative study extracts a broad, highly descriptive narrative of a situation, and then using that description the researcher determines the most critical unknowns within and designs a quantitative descriptive study to extract and analyze the unknowns independently. – Susan (Section 602)

Final Steps for Module #3: Teacher Response to Drafts

As you work on revising your drafts for Wednesday, keep in mind the following tips.

Don’t Make Too Many Assumptions About Your Participants

Good research questions are open-ended enough to ask a genuine question, but also come in with some assumptions that you know from past research. You have to be careful, though: even though other researchers may have studied similar populations, this doesn’t mean that your participants will fall into that population.

Instead of saying something like “why does X group do Y behavior” ask “is Y behavior present in X group, and if so: why does it exist?” The difference is subtle, but important. In the first formulation, the question assumes what you’re going to find. In the second, it asks if the behavior is present.

Student Example from a Past Class

Original:

How can intercultural health communication help health care providers with their struggles with the communication, linguistic difficulties, cultural traditions of patients from different ethnic backgrounds?

Revised:

What communication issues (linguistic, cultural, ethnic, etc.) do health care providers struggle with most?

How can intercultural health communication help health care providers with these issues?

Pare Down Your Research Questions

Many of you have too many research questions for one study. Research questions are the boundaries of a study: they delimit what you’re looking at and what you’re not. You can create very complex research projects that look at a lot of different variables, but those are very hard to create. It’s much better to take on simpler questions that are focused on a very specific variable.

Student Example from a Past Class

Original:

Conducting research across cultures presents many challenges. These unique challenges are not exclusive to the target audience, but also includes the researcher, who brings his own inherit bias. I wanted to explore how to best gather objective results absent a cultural bias when observing a demographic different from the researcher. I will do this by exploring online app dating from the LGBTQ perspective.

Revised:

Conducting research across cultures presents many challenges. These unique challenges are not exclusive to the target audience, but also includes the researcher, who brings his own inherent biases. In order to examine researcher bias, I will conduct an authoethnography of cultural stereotypes in online data apps targeted at LGBTQ users. I will begin by identifying stereotypes in marketing landing pages for online dating targeted at LGBTQ users. Then I will explore actual dating profiles to see how they display and/or subvert these stereotypes.

Here the researcher starts out with a lot of potential variables (intercultural research, bias, demographics, sexual preference, stereotypes, etc.). I have paired it down to one variable: cultural bias in online dating apps.

Cite Your Claims

It is important in your introduction/problem state to cite claims you are making. Anyone looking at a research proposal will want to see that you’ve done your homework.

How Will You Record Data?

In the age of omnipresent video-recording, I’m amazed by how few researchers make use of it. Taking notes will always be problematic, because you’ll lose about 80% of what actually transpired, including direct quotations, body language, etc. I only use note-taking when video-recording is impossible or completely impractical, such as during a busy activity involving a lot of people that participants are mixed in with. I video-record about 90% of my study data, however, because that’s the way to get the best data.

Final Steps for Module

5) 10/18/21 by Midnight ET >>

Post an answer to the following research question as a comment on the posts of each person’s webpage on this course website.

Each module, I will ask you a research question, which you must post a response to as a comment on the posts of each of your peers. Your research question for this module is the following:

  • Looking at this author’s Research Instrument: do the activities mentioned form a cohesive instrument? Does this instrument appear rigorous (systematic, credible, etc.)? Are there additional activities they might want to add to this instrument? Is tweaking their research question called for? Are there activities that don’t seem to fit their research question?

6) 10/20/21 by Midnight ET >>

Revise all your documents and hand them in:

Section 601