How to chose your methods | My Virtual Office

How to chose your methods.

Choosing your methodology is a very important part of the research process. The saying goes "garbage in...garbage out." This means that a bad methodology will produce bad results. Countless times I have been contacted by researchers to help them with their data analysis only to discover that they went about collecting the data all wrong for the question they had. It's never easy to break that news, and I am often asked if there is some "statistical magic" to fix it. The answer is no. You have to start off on the right foot. It can't be undone if you don't.

So, how can you make sure that you choose good methods? To start, you want to have a good understanding of the past research in the area you are doing research, and generate a strong research question. A good research question is a clear road map to good methodology. Let's consider an example.

If I am interested in the ways in which trauma impacts learning, there are several ideas for a research study that could come to mind quickly. For example, I could do a survey of a group of students and ask them about trauma they experienced. I will also ask them to tell me their GPA. I might be able to do one better, and get the school to give me their exact GPA. I could do a couple things with this data. I could do a correlation of how many traumas vs. GPA. I could run a t-test to determine if students grouped by the trauma they did or did not experience have higher or lower GPAs. I could "group" students by those who did and did not have ANY trauma and compare that overall trauma experience to GPA. I could group by individual traumas (Yes death in family; No death in family) and compare GPAs between those specific groups. What if I don't want to go quantitative? I could do a focus group with students and ask them about major life events, extracting from this any traumatic experiences, and probe for what happened after the major events to determine if their school performance was impacted.

Seems so easy, right? Okay, answer me this. What questions are you going to use? If you are thinking you can make some up, just stop right there. S.T.O.P. That is one of the best ways to sabotage your study. Creating questions is a research study in and of itself. You want to find questions that already exist and have been tested to be valid and reliable. At the very least, you want questions and prompts derived directly from the literature--meaning it is guided by theory and/or past empirical evidence. How do you get that? By reading the literature. There is just no way around it folks. You have to read those beastly articles (not just the abstract...). So, this brings us back to where I started. To chose good methods you need to have a good understanding of the past research. This will ensure that you do not repeat what someone else has done, help you identify questions that are unanswered and pressing, and illustrate how others have done similar research (including the questions they use!)

Before I give you some steps to follow to help you do this, I want to talk a little bit about the research question. Take a look at this example:

What is the relationship between childhood trauma and academic performance?

This is a very broad question because there is no clear indication of exactly what is meant by "childhood trauma" and "academic performance". It is VERY important that we define these specifically in a way that helps determine how to measure them. We call this operationalizing. Here is what I mean:

"In this study, I will examine the relationship between childhood trauma and academic performance. Childhood trauma is defined as "the experience of an event by a child that is emotionally painful or distressful, which often results in lasting mental and physical effects" (NIH, 2019). In this study, such emotionally painful or distressful events will be categorized as including one of the following: sexual abuse, death of a parent, caretaker, or sibling, and/or physical abuse. The severity and frequency of these events will be measured using Smith's (2019) Survey of Trauma. School performance will be defined as grade point average and provided by the principal of each student's school."

See what I did there? You could run this study now even though I came up with it. You know exactly what I mean by trauma and school performance and you know precisely how to get that information. I started with a broad question and then narrowed in on the definition and way of measuring each variable. This is what you are aiming for. It takes time. So, let's get started.

Here is your to-do list when thinking about your methods.

Write down some of the things you know about this area from personal or professional experiences.
Find research articles (from your University library) about the topic and list a few key findings that stand out to you.
What questions come to mind after reviewing these articles? What is left to be known? Sometimes the authors of research articles actually state these in their discussion section. It's a good reason to read all the way to the end!
Practice thinking about how you would conduct a study to answer some of those questions. For example:

List two strategies for quantitative data collection. Be specific. What environment (even if it’s fictional at this point) would you need? What tools would you need?
What instruments (surveys, tests, etc.) have researchers used in the articles you read that you might be able to use here?
List two strategies for qualitative data collection. Be specific. What environment (even if it’s fictional at this point) would you need? How would you operationalize ONE key variable so that you could pinpoint it when you see it in the environment/interview/conversation/text/etc.?
Can you mix them up? Is there a part of the quantitative methods and qualitative methods that could be used together to complement each other?
Consider the real life practicalities if you were going to use these methods. What resources can you access to make these methods happen? What would need to change? What could be improved?

Feel ready to define your research question and develop your methods? Practice with this form. After completing the form, it will submit the information directly to me, and I can provide you with feedback.

MY VIRTUAL OFFICE

This website is just for students. I have information categorized in five major areas below. Click on the topic that you need support with right now to access support in that area.

NEED HELP WITH YOUR RESEARCH?

I create customized plans to assist each student with what they need.

Email me and let me know what you need.

Tamarah Smith, PhD
drsmith@theresearchschool.org

How to chose your methods.

So, how can you make sure that you choose good methods? To start, you want to have a good understanding of the past research in the area you are doing research, and generate a strong research question. A good research question is a clear road map to good methodology. Let's consider an example.

Before I give you some steps to follow to help you do this, I want to talk a little bit about the research question. Take a look at this example:

What is the relationship between childhood trauma and academic performance?

This is a very broad question because there is no clear indication of exactly what is meant by "childhood trauma" and "academic performance". It is VERY important that we define these specifically in a way that helps determine how to measure them. We call this operationalizing. Here is what I mean:

Here is your to-do list when thinking about your methods.

Write down some of the things you know about this area from personal or professional experiences.

Find research articles (from your University library) about the topic and list a few key findings that stand out to you.

What questions come to mind after reviewing these articles? What is left to be known? Sometimes the authors of research articles actually state these in their discussion section. It's a good reason to read all the way to the end!

Practice thinking about how you would conduct a study to answer some of those questions. For example:

List two strategies for quantitative data collection. Be specific. What environment (even if it’s fictional at this point) would you need? What tools would you need?

What instruments (surveys, tests, etc.) have researchers used in the articles you read that you might be able to use here?

List two strategies for qualitative data collection. Be specific. What environment (even if it’s fictional at this point) would you need? How would you operationalize ONE key variable so that you could pinpoint it when you see it in the environment/interview/conversation/text/etc.?

Can you mix them up? Is there a part of the quantitative methods and qualitative methods that could be used together to complement each other?

Consider the real life practicalities if you were going to use these methods. What resources can you access to make these methods happen? What would need to change? What could be improved?

Feel ready to define your research question and develop your methods? Practice with this form. After completing the form, it will submit the information directly to me, and I can provide you with feedback.

How to chose your methods.

​

​

So, how can you make sure that you choose good methods? To start, you want to have a good understanding of the past research in the area you are doing research, and generate a strong research question. A good research question is a clear road map to good methodology. Let's consider an example.

​

​

​

Before I give you some steps to follow to help you do this, I want to talk a little bit about the research question. Take a look at this example:

​

What is the relationship between childhood trauma and academic performance?

​

This is a very broad question because there is no clear indication of exactly what is meant by "childhood trauma" and "academic performance". It is VERY important that we define these specifically in a way that helps determine how to measure them. We call this operationalizing. Here is what I mean:

​

​

​

Here is your to-do list when thinking about your methods.

​

Write down some of the things you know about this area from personal or professional experiences.

Find research articles (from your University library) about the topic and list a few key findings that stand out to you.

What questions come to mind after reviewing these articles? What is left to be known? Sometimes the authors of research articles actually state these in their discussion section. It's a good reason to read all the way to the end!

Practice thinking about how you would conduct a study to answer some of those questions. For example:

​

List two strategies for quantitative data collection. Be specific. What environment (even if it’s fictional at this point) would you need? What tools would you need?

What instruments (surveys, tests, etc.) have researchers used in the articles you read that you might be able to use here?

List two strategies for qualitative data collection. Be specific. What environment (even if it’s fictional at this point) would you need? How would you operationalize ONE key variable so that you could pinpoint it when you see it in the environment/interview/conversation/text/etc.?

Can you mix them up? Is there a part of the quantitative methods and qualitative methods that could be used together to complement each other?

Consider the real life practicalities if you were going to use these methods. What resources can you access to make these methods happen? What would need to change? What could be improved?

​

Feel ready to define your research question and develop your methods? Practice with this form. After completing the form, it will submit the information directly to me, and I can provide you with feedback.

​

​

​

​

​