A specific, testable prediction about an expected relationship or difference.

The factor a researcher manipulates or uses to define study conditions.

The outcome measured to assess whether it differs across conditions.

A relationship between variables; by itself, it does not establish causation.

Repeating a study to test whether a finding is reliable and to identify its limits.

A specific prediction that can be tested with evidence.

A statistical relationship between variables that does not, by itself, show causation.

A repeat of a study used to test the reliability and boundaries of a finding.

What is comparison group?

A group used as a reference for judging whether an intervention made a difference.

How Psychologists Turn Questions into Evidence — Psychology: How Minds Are Studied, Shape Experience and Dream

1. Start with a Question, Not an Answer

Make curiosity testable

A broad question becomes researchable when it identifies who is studied, what may influence them, and what outcome will be measured.

Evidence over anecdotes

Psychologists use empirical methods: systematic observation and data, rather than relying on a memorable story or personal intuition alone.

Research question: Among healthy adults, does avoiding smartphone use for one hour before bed change next-morning alertness?

2. Turn the Question into a Hypothesis

A hypothesis predicts

The team predicts that adults instructed to avoid smartphones before bed will report higher next-morning alertness than those following their usual routine.

Variables

The independent variable is the bedtime-phone instruction. The dependent variable is the alertness score measured the next morning.

Operationalization

During the sixty minutes immediately before each participant's planned bedtime, the smartphone is not used and is placed outside the bedroom. Alertness is rated from 1 to 7 at the same time each morning.

3. Build an Experiment with a Fair Comparison

Two conditions

Eighty volunteers are randomly assigned for seven nights either to avoid smartphone use and keep the phone outside the bedroom during the sixty minutes before their planned bedtime, or to follow their usual routine.

Why random assignment?

It helps make groups comparable by distributing many starting differences across them, reducing alternative explanations for a group difference.

Why compare?

Without a comparison group, better alertness could be caused by expectations, an easier week, or closer attention to bedtime rather than the instruction itself.

4. Spot the Alternative Explanation

Thought exercise

Imagine the phone-free group reports better alertness after a week. Pause before saying phones caused the change.

What else might explain it?

Participants may have gone to bed earlier because they followed a new routine.
They may expect the phone-free rule to help and rate themselves more favorably.
Some may replace phone use with another stimulating activity.

Now ask: What could the team measure, standardize, or compare to make each explanation less plausible? For example, the sleep diary can record bedtime, wake time, caffeine use, and whether the instruction was followed.

A strong study is not one that eliminates every uncertainty. It is one that anticipates plausible alternatives and makes its conclusions appropriately narrow.

5. Correlation Is a Clue, Not a Cause

Observational study

A survey that measures natural phone habits and alertness can reveal a pattern, but it does not assign anyone to change their behavior.

Three causal possibilities

If two variables correlate, A may affect B, B may affect A, or a third factor, such as stress, may affect both.

Match claim to design

Use "associated with" for correlation. Stronger causal claims require designs that better rule out competing explanations.

6. Treat One Study as a Starting Point

Evidence, not a verdict

A single positive result may be informative, but researchers still examine uncertainty and ask whether the result appears again.

Replication

Replication repeats a study with new participants or a deliberate variation, such as adding an attention task alongside self-reported alertness.

Converging evidence

Repeated, compatible findings raise confidence. Differences between studies may reveal when, for whom, or why an effect occurs.

7. Review the Evidence Toolkit

Flip the cards to review the terms that turn a question into evidence.

Hypothesis: A specific, testable prediction about an expected relationship or difference.
Independent variable: The factor a researcher manipulates or uses to define study conditions.
Dependent variable: The outcome measured to assess whether it differs across conditions.
Correlation: A relationship between variables; by itself, it does not establish causation.
Replication: Repeating a study to test whether a finding is reliable and to identify its limits.

8. Check Your Understanding

One final decision

Which design best supports a causal conclusion about a bedtime-phone instruction and next-morning alertness?

Which result would most strongly support the claim that the instruction caused a difference in alertness?

Adults were randomly assigned either to avoid smartphone use and place the phone outside the bedroom during the sixty minutes before their planned bedtime or to follow their usual routine, then completed the same alertness measure.
A survey found that people who used phones more often at night were less alert the next day.
Several people said they personally felt better after deciding to use their phones less.

Show Answer

Answer: A) Adults were randomly assigned either to avoid smartphone use and place the phone outside the bedroom during the sixty minutes before their planned bedtime or to follow their usual routine, then completed the same alertness measure.

Random assignment and a comparison group make alternative explanations less likely than in a survey or anecdote. The phone-free instruction directly prohibits smartphone use and requires the phone to remain outside the bedroom during the sixty minutes immediately before each participant's planned bedtime. The result would still need careful measurement, transparent reporting, and replication before supporting a broad conclusion.

Key Terms

hypothesis: A specific prediction that can be tested with evidence.
correlation: A statistical relationship between variables that does not, by itself, show causation.
replication: A repeat of a study used to test the reliability and boundaries of a finding.
comparison group: A group used as a reference for judging whether an intervention made a difference.
empirical method: A method that relies on systematic observation, measurement, and analysis of data.
informed consent: A process that gives prospective participants understandable information and a voluntary choice about participating in research.
random assignment: Assigning participants to conditions by chance to improve group comparability.
dependent variable: The measured outcome that may change in response to an independent variable.
operationalization: Defining an abstract concept through a concrete measurement or procedure.
independent variable: The factor manipulated by a researcher or used to distinguish conditions.