4.9 Scientific Reasoning and the Scientific Method
Key Takeaways
- The scientific method moves from observation to question to testable hypothesis to experiment to analysis to conclusion
- The independent variable is manipulated, the dependent variable is measured, and controlled variables are held constant
- A control group provides a baseline; randomization, blinding, and a placebo reduce bias; larger samples are more reliable
- Correlation does not prove causation—watch for confounding variables before concluding cause and effect
- Validity means a study measures what it intends to (accurate); reliability means repeated measurements agree (consistent)
Thinking Like a Scientist
The ATI TEAS 7 Science section reserves a block of questions for scientific reasoning — not memorized facts, but the ability to design a fair test, identify variables, read a graph, and judge whether a claim is justified. Nurses use these skills constantly to read research, interpret monitor data, and avoid jumping to conclusions.
The Scientific Method
| Step | What You Do | Example |
|---|---|---|
| 1. Observation | Notice a phenomenon | Patients with low vitamin D report fatigue |
| 2. Question | Ask why or how | Does vitamin D reduce fatigue? |
| 3. Hypothesis | Propose a testable, falsifiable answer | If patients take vitamin D, fatigue will decrease |
| 4. Experiment | Test it under controlled conditions | Randomized controlled trial |
| 5. Analyze data | Examine results, often statistically | Compare fatigue scores between groups |
| 6. Conclusion | Support or reject the hypothesis | Evidence does or does not support it |
| 7. Communicate | Share findings for peer review | Publish in a journal |
A strong hypothesis is testable and falsifiable — phrased as an "if/then" prediction that an experiment could prove wrong. "Plants are pretty" is not a hypothesis; "if plants get more light, then they grow taller" is.
Variables in an Experiment
| Variable | Definition | Example |
|---|---|---|
| Independent | What the researcher changes | Vitamin D dose |
| Dependent | What the researcher measures (the outcome) | Fatigue score |
| Controlled | Held constant for a fair test | Age, diet, sleep |
| Confounding | An uncontrolled variable that distorts results | Exercise level |
A memory hook: the dependent variable depends on the independent variable. In a graph, the independent variable goes on the x-axis and the dependent variable on the y-axis.
Experimental Design
| Component | Purpose |
|---|---|
| Control group | Baseline that receives no treatment |
| Experimental group | Receives the treatment |
| Randomization | Reduces selection bias |
| Blinding | Subjects do not know their group |
| Double-blind | Neither subjects nor researchers know |
| Placebo | Inactive treatment for the control group |
| Sample size | Larger samples give more reliable results |
The control group is the heart of a fair experiment: without an untreated baseline, you cannot tell whether the treatment or something else caused the change.
Correlation vs. Causation
Correlation means two variables move together; causation means one directly produces the other. The TEAS repeatedly tests the rule that correlation does not prove causation.
Worked Example: A researcher notices that as monthly ice cream sales rise, drowning deaths also rise, and the two are strongly correlated. Does eating ice cream cause drowning? No. A confounding variable — hot summer weather — independently increases both ice cream sales and swimming (and therefore drowning). Two variables can rise together without one causing the other. To claim causation you need a controlled experiment that rules out confounders.
Types of Studies
| Study | Description | Strength |
|---|---|---|
| Experimental | Researcher manipulates a variable | Can show causation |
| Observational | Researcher only observes | Cannot prove causation |
| Case study | In-depth look at one subject | Rich detail, not generalizable |
| Survey | Self-reported data | Fast, large samples |
| Longitudinal | Follows subjects over time | Shows change and development |
Only a controlled experiment can establish cause and effect, because only there does the researcher actively manipulate the independent variable while holding others constant.
Reading Data and Graphs
Match the graph to the data:
- Bar graph — compares categories (e.g., fatigue scores by treatment group)
- Line graph — shows change over time or a trend
- Pie chart — shows parts of a whole (percentages)
- Scatter plot — shows the relationship/correlation between two variables
Always read the title, axis labels, and units first, then identify the trend before drawing any conclusion.
Measurement and Units
Science uses the SI (metric) system. Prefixes scale a base unit by powers of ten: kilo- (1,000x), centi- (1/100), milli- (1/1,000), micro- (1/1,000,000). So 1 kilogram = 1,000 grams and 1 milliliter = 0.001 liter.
Validity, Reliability, Accuracy, and Precision
| Term | Meaning |
|---|---|
| Validity | The study measures what it claims to measure |
| Reliability | Repeated measurements give consistent results |
| Accuracy | A measurement is close to the true value |
| Precision | Repeated measurements are close to each other |
A classic distinction: a scale that always reads 5 pounds heavy is precise (consistent) but not accurate (off from the true value). A study can be reliable (repeatable) yet still invalid if it measures the wrong thing — for example, using shoe size to estimate intelligence would be reliable but not valid.
In an experiment testing a new medication, what is the purpose of the control group?
A study finds that students who eat breakfast tend to score higher on tests. What is the most appropriate conclusion?
In an experiment, which variable is manipulated by the researcher?
A bathroom scale consistently reads 3 pounds too high every time. This scale is BEST described as:
Match each research term to its meaning.
Match each item on the left with the correct item on the right