Skip to Main Content

Respiratory Therapy & Respiratory Care

What is the scientific method?

The scientific method has six steps, described on the following tabs. They will help you solve all kinds of problems, in and out of school.


Nelson, M. R. (2010). Scientific method. In K. Krapp (Ed.), Experiment centralUnderstanding scientific principles through projects (2nd ed., Vol. 5, pp. 1005-1009). UXL.


Step 1: State a problem or ask a question.

To begin using the scientific method, think about the world around you. You may see something that makes you curious, such your sandwich drying out by lunchtime on some days but not on others.

Put your curiosity into the form of a problem or question, such as these:

  • Why does my sandwich dry out some days but not others?

Step 2: Gather background information.

Read more about the problem or question. Observe it closely.

Step 3: Form a hypothesis.

Now use what you know about the situation to think of a possible answer for your question. This answer, or guess, is your hypothesis. A hypothesis is an idea in the form of a statement that can be tested by observations and/or experiment. You will use what you already know about the situation to form a hypothesis, Here are possible hypotheses to answer the questions above:  Plastic bags that seal keep more moisture in bread than waxed paper or plastic bags without seals do.

A hypothesis should be testable: You can make observations, do research, or set up experiments to determine whether each hypothesis is correct. Here is one example of hypotheses that are vague and difficult to test:

  • Sandwiches taste better when you seal them in plastic bags. How can you measure “taste better?”

Step 4: Design and perform an experiment.

You might be able to establish whether your hypothesis is accurate by research, with the sandwich problem, the best approach is an experiment.

An experiment is a controlled observation. The experimenter carefully changes one condition at a time, such as the type of sandwich wrapping, and observes what happens. In most experiments, a control experiment is set up with the same conditions as the actual experiment. The conditions remain the same in the control experiment but are changed in the actual experiment, one condition at a time. If something happens only in the actual experiment and not in the control, it is clear that it was caused by changing a condition in the actual experiment. The control experiment for our sandwiches might be leaving a slice of bread unwrapped to see what happens to it and comparing it to those in various wrappings.

Conditions that change during an experiment and affect the results are called variables. The variables in our sample experiment include the type of bread, how fresh it is, the size of the piece of bread being wrapped, any fillings used with the bread, the length of time the bread is wrapped, the temperature of the wrapped bread during the experiment, and the type of sandwich wrapping. Only one variable is changed at a time during the experiment. The variable being changed is called the independent variable, which in our experiment is the type of sandwich wrapping.

What might happen if we change two variables at a time, such as wrapping wheat bread with waxed paper and putting rye bread in a sealed plastic bag? If the rye bread is fresher than the wheat bread at the end of the experiment, we cannot be sure which variable is the cause—the type of bread or the type of wrapping.

The condition that change during an experiment is called the dependent variable. In our example, the dependent variable is the amount of moisture in the bread. Results of experiments must be measurable, so we need a way to measure this moisture. We decide to weigh each slice of bread before and after the experiment. The difference in the weight would be the amount of moisture that evaporated.

Experiments must also be repeatable. We must write down our procedure and follow it carefully, so that someone else could carry out the same procedure and see if the same results occur.

Step 5: Draw a conclusion.

The next step in the scientific method is to graph or chart our results, analyze them, and determine whether our hypothesis was correct. For some experiments, we might have quite a bit of data to analyze. For our sample experiment, we compare the loss in weight of each bread slice after the wrapping is removed. What is our conclusion? Did our results support our hypothesis?

Even if the results did not support our hypothesis, we have learned something just by asking the question and doing the experiment. Often there is no “right” answer when we use the scientific method. Instead, we simply gather more information about the problem, which is valuable in itself.

Step 6: Report the results.

Reporting our results allows other scientists to build on our work—and to repeat our experiment to see if they get the same results. Without the sharing of results, little scientific progress would be made. Scientists publish their findings in scientific journals as a way of sharing what they have learned.

In the two experiments that follow, you will use information you gather to identify mystery powders, and you will use the scientific method to prove or disprove a hypothesis.

Types of Research Methods

Research methods are the strategies, processes or techniques utilized in the collection of data or evidence for analysis in order to uncover new information or create better understanding of a topic. There are different types of research methods which use different tools for data collection.

Qualitative Research gathers data about lived experiences, emotions or behaviours, and the meanings individuals attach to them. It assists in enabling researchers to gain a better understanding of complex concepts, social interactions or cultural phenomena. This type of research is useful in the exploration of how or why things have occurred, interpreting events and describing actions.

Qualitative research refers to any research based on something that is impossible to accurately and precisely measure. It uses methods such as interviews, open-ended questions, participant observations, case studies, focus groups, etc. to identify patterns, themes, and features. These factors cannot easily be reduced to numbers. Qualitative research is common in the social sciences.

Quantitative Research gathers numerical data which can be ranked, measured or categorised through statistical analysis. It assists with uncovering patterns or relationships, and for making generalisations. This type of research is useful for finding out how many, how much, how often, or to what extent. This research based on something that can be accurately and precisely measured might also be referred to as "empirical research."

Pro tip: Statistical analysis in an article usually indicates quantitative research. Check the articles you find to see if some sort of numerical measuring and statistical analysis is present along with the characteristics listed below.

  • Introduction with a statement of background or purpose (what was being studied and why). May review prior studies on the same topic.
  • Description of the design and/or method of the study (the experimental group or sample, control, variables, number of test subjects, test conditions, etc.)
  • Results, or report of the findings (in numeric form as tables, charts, or graphs, etc., often with statistical analysis)
  • Conclusions that can be drawn from the results (may be labeled discussion or significance)
  • Footnotes and/or a bibliography

Mixed Methods Research integrates both Qualitative and Quantitative Research. It provides a holistic approach combining and analysing the statistical data with deeper contextualised insights. Using Mixed Methods also enables Triangulation, or verification, of the data from two or more sources.

Finding Mixed Methods research in the Databases 

  • PubMed - there are no suitable MeSH terms for mixed methods research in Medline. Search your topic with the following suggested free text keywords using the quotation marks and truncation symbol*:

    “mixed model*” OR “mixed design*” OR “multiple method*” OR multimethod* OR triangulat*

    CINAHL - the following Subject Headings may be of use: Multimethod Studies or Triangulation. You can also include in your search the following free text keywords: mixed model*, mixed design*, multiple method*, multimethod*, or triangulat*.

A systematic review usually focuses on a single, well-defined research question and seeks to comprehensively gather all existing studies that address this research question. 

meta-analysis takes the results of several existing quantitative studies and analyzes them in a new way. Meta analysis looks for previously unnoticed patterns or trends among existing study results, or seeks to pull out new data from them. Meta analysis is usually considered another form of quantitative research.

literature review, also known as a review article, is an article whose sole purpose is to provide an overview of previous important research on a particular topic. Although valuable to researchers, literature reviews are not considered primary research. However, they can help you identify research trends and major articles published on a topic. No new study is conducted in a true literature review.

Data Collection Techniques & Tools

Techniques or tools used for gathering research data include:

Qualitative Techniques or Tools Quantitative Techniques or Tools
Interviews: these can be structured, semi-structured or unstructured in-depth sessions with the researcher and a participant. Surveys or questionnaires: which ask the same questions to large numbers of participants or use Likert scales which measure opinions as numerical data.
Focus groups: with several participants discussing a particular topic or a set of questions. Researchers can be facilitators or observers. Observation: which can either involve counting the number of times a specific phenomenon occurs, or the coding of observational data in order to translate it into numbers.
Observations: On-site, in-context or role-play options. Document screening: sourcing numerical data from financial reports or counting word occurrences.
Document analysis: Interrogation of correspondence (letters, diaries, emails etc) or reports. Experiments: testing hypotheses in laboratories, testing cause and effect relationships, through field experiments, or via quasi- or natural experiments.
Oral history or life stories: Remembrances or memories of experiences told to the researcher.  

Dartmouth's Evidence Pyramid

Dartmouth's Evidence Pyramid provides a way to visualize both the quality of evidence and the amount of evidence available. For example, systematic reviews are at the top of the pyramid, meaning they are both the highest level of evidence and the least common. As you go down the pyramid, the amount of evidence will increase as the quality of the evidence decreases. Image credit: EBM Pyramid by Trustees of Dartmouth College and Yale University.

Levels of evidence in Evidence based research

Here's another take on the evidence pyramid indicating primary vs. secondary sources and explaining each level.

6-level pyramid of primary and secondary research evidence

Study Design Tutorial

Evidence Based Practice

What is evidence-based medicine?

Evidence based medicine is the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients" (Sackett et al., 1996, p. 71). Also called evidence-based practice, EBM incorporates the practitioner's expertise and clinical judgement with relevant scientific evidence to honor patients' values and preferences in recommending treatment.

evidence based medicine venn diagram


Sackett, D. L., Rosenberg, W. M., Gray, J. A., Haynes, R. B., & Richardson, W. S. (1996). Evidence based medicine: what it is and what it isn't. BMJ (Clinical research ed.)312(7023), 71–72.,