Skip to main content

Health Sciences Advanced Search

This guide is designed from postgraduate level students working on primary research projects or dissertations.

The Research Question

When searching the literature to discover evidence, the searcher MUST have a focused research question so that the search will be productive.

To create a productive research question, follow these steps:

1. Identify your topic. What is "IT" that you want to know?     

The answer to this questions will be a noun.   

"I'm interested in obesity."     "I want to research transductive reasoning."

2. Identify exactly what you want to KNOW about that topic. What about "IT?"     

The answer to this question could be difficult to state succinctly. Talk it through with somebody. Use the "talk-back method."

Find a partner. State what you want to look for, and let your partner state back to you what you just said. Your partner will usually repeat what you said in MUCH FEWER words.

You will also hear your question in another person's voice, and then can fine-tune the question more readily.   

"I haven't been sleeping well, and I've noticed I'm gaining weight. I want to know if sleeping more will help me lose weight. Dr. Oz says it will, but I want to know for sure. I want to see the science myself."

Just stating what you want to know will help you see/hear additional details that could be important. Just how much sleep will be enough to prove or disprove effectiveness of weight loss?

The better research question would be as follows:     

"I want to know if 6-8 hours of sleep per night is as effective as walking for 30 minutes 3 times a week to promote weightloss."

3. Now, continue to identify the parameters of your research question.

Think about what you want and what you don't want to show up in your search results.     
Are you interested in ALL people over ALL time? Or are you interested in a specific demographic? Are you interested in power walking or strolling or 20,000 steps a day?     

"I want to know if 6-8 hours a sleep per night is as effective as walking for 30 minutes 3 times a week for Hispanic women who are postmenopausal."   

"I want to know if problem-based learning is more effective than face-to-face lecture for family nurse practitioners enrolled in clinical courses."

GOOD RESEARCH QUESTIONS versus BAD RESEARCH QUESTIONS
  • If your research question can be answered with a YES or a NO, then it's not a good research questions.
  • If your research question can be answered with a statistic, then it's not a good research question. 
  • If your research question is so broad that people could write an entire books about it, then it's not a good research question.

"Is social media harmful to adolescents?"

  • If your research question telegraphs the answer you want to find, then it's not a good research question. 

"Is social media as harmful to teenagers as verbal abuse by parents?"

  • If your research question is too new, too current, then scientific research has not had time to be adequately performed to discover the answer. It's not a good research question.  These extremely current research questions could concern the effectiveness of drugs or devices that have been "on the market" for less than a year or so.   
  • Is your research question "arguable?" Could two well-meaning adults disagree on the answer? If so, the search of the literature will resolve the disagreement.  This research question is a good one.
  • If you asked a knowledgeable co-worker your research question and that person does not know the answer, you MAY have a good research question.

Research projects that require advanced search skills BEGIN with a Research Question that targets the evidence the researcher is seeking to find.

The PICO is one of many mnemonic tools used to help researchers formulate a Research Question that will guide a literature search.

P -- Population Problem Patient+Diagnosis (+ demographic characteristics) 
I  -- Intervention   (the new thing; the thing the researcher suspects is correct)
C -- Comparison  (the traditional thing; the default thing; the normal choice)
O -- Outcome       (the desired result; the measure of success -- BMI < 23 or temperature < 99º)


Example: 

Population:      females, +65yrs, African American, diagnosed with Diabetes Mellitus
Intervention:     metformin
Comparison:    diet & exercise
Outcome:         target outcome of A1C level < 5.8


Use the "question" words below to begin creating your Research Question:

Is?
Does?
Do?
Will?
Which?
What?
Can?
How?
In?
Are?
etc.....


Now, create your question using the PICO tool:


             In 
P:  African American females over the age of 65,
             is
I:    Metformin (I)
             more effective than
C:   Diet & Exercise 
             at reducing
O:   A1C levels to 5.7 or lower 

In African American females over the age of 65, is Metformin more effective than Diet & Exercise at reducing A1C levels to 5.7 or lower? 


The PICO has 4 different research question types:

  1. Therapy Questions
  2. Etiology Questions
  3. Diagnosis Questions
  4. Prognosis Questions

The handout below offers assistance with creating Research Questions depending on your PICO question type.


For additional mnemonic tools that assist in creating Research Questions, see the document below.