The Essential Elements of Experimental Design, Explained

UPI Study › Courses › Research Methods in Psychology › Free lessons › Lesson 12.1 Free with a free account

The Essential Elements of Experimental Design

Research Methods in Psychology · Chapter 12: Experimental Design FundamentalsSection 12.1 · lesson 70 of 9410 min · free with a free account

← Practical Strategies to Boost External ValidityNext lesson: Why Random Assignment Matters →

Earn 3 college credits for Research Methods in PsychologyRecommended for credit by NCCRS · Psychology 111 · self-paced · one payment, lifetime access
Enroll in Research Methods in Psychology · $250 →Download the syllabus (PDF)

In this lesson

The four pillars that turn an investigation into a true experiment.

  • Define the four essential elements of an experiment
  • Distinguish independent, dependent and extraneous variables
  • Identify the two conditions required for a causal claim
  • Separate true experiments from quasi- and non-experiments

Free study notes

Get this lesson's notes by email

The key points and 4 practice questions with answers, in your inbox. One email, free.

One email. Unsubscribe in one click.

Transcript

Everything said in the video, section by section (1,337 words). Read along, or search it.

Experimental Design Fundamentals

The Essential Elements of Experimental Design

Psychology runs thousands of studies a year, and only some of them are experiments. That distinction is not a technicality. It decides what you are allowed to conclude. This section unpacks the architecture beneath every experimental study. Four elements, and two conditions that those elements exist to satisfy. By the end you will be able to define the four essential elements of an experiment. You will separate independent, dependent, and extraneous variables in any study you read. You will name the two conditions that have to be met before a causal claim is legitimate. And you will tell a true experiment from a quasi-experiment at a glance. That last skill is the practical payoff of the whole section.

What Counts

What Makes a Study an Experiment

Start with what does not make a study an experiment. It is not the use of statistics. Correlational studies use statistics too. It is not the presence of a hypothesis. Almost every study has one. And it is not the formality of the laboratory setting. A study can happen in a windowless lab and still be purely observational. The defining feature is a pair of structural choices the researcher makes. First, deliberate manipulation of one variable, while holding others stable. Second, random allocation of participants to conditions. Those two choices give experiments a special power. They support claims about cause and effect, rather than mere association. Everything else in this section follows from that one sentence.

The Architecture

Four Interlocking Components

An experiment is built from four interlocking components. The independent variable, the thing you change. The dependent variable, the thing you measure. Random assignment, the way participants reach their condition. And control of extraneous variables, everything you deliberately hold still. The word interlocking is doing real work here. These are not four boxes to tick independently. Removing any one of them weakens the study's ability to identify a causal effect. Often it destroys that ability outright. So take them one at a time, and notice what each one is protecting against.

The Presumed Cause

The Independent Variable

The independent variable is the factor the researcher actively manipulates. Notice the word actively. The researcher sets it, not the participant. It is the presumed cause in the relationship being investigated. Independent variables normally take at least two levels. Presence versus absence of a treatment, for instance, or different doses of a drug. Two levels are the minimum, because with only one there is nothing to compare. Now here is the part that decides a study's whole category. Without manipulation, you are merely observing variation that occurred for some other reason. That is the territory of correlational research, not experimental research. No amount of statistical sophistication later can convert one into the other.

The Measured Outcome

The Dependent Variable

The dependent variable is the outcome you measure, to detect the effect of the manipulation. Good dependent variables meet three standards at once. They are sensitive, meaning they register real changes when those changes occur. A measure too coarse to move will show you nothing, even from a real effect. They are reliable, meaning they give consistent readings. And they are valid, meaning they actually capture the construct of interest. Consider a study on whether sleep deprivation affects memory. It might use recall accuracy as its dependent variable. That single choice shapes what the experiment can, and cannot, reveal. Pick a different memory measure and you may be studying a different question entirely.

Chance, Not Choice

Random Assignment

Random assignment is how participants get placed into experimental conditions. By a chance process. A coin flip, a random number generator, or an equivalent unbiased mechanism. Now here is the part almost everybody gets wrong. The purpose of random assignment is not to make the groups identical. No procedure can do that with finite samples. The groups will differ. The purpose is to control how those differences got there. Pre-existing differences are distributed across conditions in a way that owes nothing to systematic selection. So the differences are still present, but they are unsystematic. With random assignment, the only systematic difference between groups is the manipulation itself. That sentence is the entire justification for causal inference from an experiment.

Everything Else

Control of Extraneous Variables

Extraneous variables are influences other than the independent variable that could affect the outcome. Features of the setting. The time of day. The experimenter's behavior. Even characteristics of the materials themselves. The researcher has three tools for handling them. Hold them constant across conditions, so they cannot differ. Counterbalance them systematically, so their influence cancels out. Or measure them, so they can be statistically accounted for later. The danger point is specific and worth memorizing. When an extraneous variable moves in step with the manipulation, it becomes a confound. At that moment the experiment loses the ability to attribute the effect to the independent variable. The data are still real. The interpretation is no longer available to you.

A Common Mistake

Two Groups Is Not Enough

Here is a mistake that shows up constantly in student write-ups. Labeling any study with two groups an experiment. Two groups is not the test. The test is why the groups differ. If the researcher randomly assigned them, the study is experimental. Say, to a twenty minute meditation session, or a twenty minute reading session. But if participants chose their group, the study is correlational. Same if they were born into it, or already belonged to it. The two studies can look almost identical on the page. The label matters because only the first design supports a causal claim. So when you read a method section, look for the assignment sentence first.

Causal Inference

The Two Conditions

Establishing that one variable causes another requires meeting two structural conditions. Not one. Two. And each is supplied by a different design element. The first is covariation. The cause and the outcome must vary together. When the cause changes, the outcome changes too. That condition is supplied by manipulating the independent variable across at least two levels. The second condition is internal validity. Alternative explanations for the covariation must be ruled out. The change in the outcome must be attributable to the manipulation, and to nothing else. That condition is supplied by random assignment, plus control of extraneous variables. Meet the first without the second and you have a correlation with extra steps.

Three Categories

True, Quasi, and Non-Experiments

Researchers sort studies into three broad categories. They are distinguished by which of the essential elements each one includes. A true experiment has manipulation and random assignment. It supports a causal claim, with strong internal validity. A quasi-experiment has manipulation, but no random assignment. It uses intact groups, so its causal claim is limited. Alternative explanations remain live, because the groups may have differed beforehand. A non-experiment has neither. It is observation only. It can show association, and nothing stronger than association. So the strength of a causal claim depends directly on which category the study belongs to.

The Trade-Off

When a True Experiment Is Impossible

Now for the honest complication. Many of the most important questions in psychology cannot be studied experimentally at all. The effects of trauma. The effects of poverty. The effects of chronic illness. You cannot randomly assign a person to poverty. Manipulating those variables would be unethical, or simply impossible. So researchers turn to quasi-experimental and observational designs. They accept weaker causal claims, in exchange for studying phenomena that matter. That is a bargain, not a failure. Recognizing the trade-off openly is part of mature research practice. The problem is never an imperfect design. It is a design whose limits go unstated.

Reading Critically

Four Diagnostic Questions

Finally, a checklist you can run on any published experiment. Four questions cover the essential elements and expose most design weaknesses at a glance. What was manipulated? That reveals whether the study is experimental at all. It also tells you what the authors are treating as the cause. What was measured? That reveals whether the dependent variable is sensitive and reliable. And whether it is conceptually appropriate for the question being asked. How were participants assigned to conditions? That tells you whether random assignment was used, or whether the groups were intact. And finally, what was held constant or controlled? That reveals whether any extraneous variables were free to confound the results. Answer them in order, and the ceiling on the study's conclusions becomes obvious.

About Research Methods in Psychology at UPI Study

Research Methods in Psychology (Psychology 111) is a self-paced online course of 94 video lessons in 16 chapters, from the scientific method and research ethics to designs, data collection, sampling, measurement, validity, experiments, descriptive statistics and hypothesis testing. It carries a NCCRS credit recommendation for 3 semester credits.

Every lesson transcript is free on pages like this one, and the videos are free with a free UPI Study account. Enrolling adds a practice quiz after each lesson, the graded work and final exam that earn the credit, lifetime access, and an official transcript on completion. UPI Study has helped over 130,000 students with college credits. ACE and NCCRS have over 2,100 cooperating colleges that will consider accepting credits; the final decision is with the universities.

Enroll now → Course details Download the syllabus All Psychology and social science courses Every course with PRO →

Compare: Research Methods vs Statistics

Why take it here

Why students take Research Methods in Psychology with UPI Study

How psychologists know what they knowResearch Methods in Psychology teaches how psychological studies are designed, run and judged: research ethics, surveys and observation, correlational and experimental designs, sampling, reliability and validity, and the statistics used to test hypotheses. It is a core course in psychology and many social-science degrees.
Credit at a fraction of the priceThe same 3 credits can cost $750 or more at a US college. Here it is one payment, self-paced, with lifetime access.
Finish your degree soonerTransfer credits reduce the courses you still owe, so you can graduate earlier or lighten a heavy semester.
Strengthen an applicationA completed college-level course on an official transcript shows admissions and scholarship committees you can do the work.
Study on your own scheduleNo start dates, no deadlines. Watch the lessons, practice, and take the final when you are ready.
130,000+students helped with college creditsACE and NCCRS have over 2,100 cooperating colleges that will consider accepting credits; the final decision is with the universities.

Where the credit goes

Universities that work with NCCRS

Research Methods in Psychology carries a NCCRS credit recommendation. These universities work with NCCRS; some are UPI Study's signed partners. ACE and NCCRS have over 2,100 cooperating colleges that will consider accepting credits; the final decision is with the universities.

Thomas Edison State UniversityThomas Edison State UniversitySigned partnerUniversity of the PeopleUniversity of the PeopleSigned partnerCharter Oak State CollegeCharter Oak State CollegeSigned partnerStrayer UniversityStrayer UniversitySigned partnerColumbia Southern UniversityColumbia Southern UniversitySigned partnerIndiana TechIndiana TechSigned partnerPenn StatePenn StateWorks with NCCRSSouthern New Hampshire UniversitySouthern New Hampshire UniversityWorks with NCCRSSUNYSUNYWorks with NCCRS +many more cooperating universitiesFind my college →

Find my college → Not listed? Ask Athena in the corner of this page, or book a 15-minute call.

FAQ

Questions students ask us

About The Essential Elements of Experimental Design

What are the four essential elements of an experiment?
An experiment has four interlocking elements: an independent variable that the researcher actively manipulates, a dependent variable measured to detect the manipulation's effect, random assignment of participants to conditions, and control of extraneous variables. Removing any one weakens, and often destroys, the study's ability to identify a causal effect. Manipulation combined with random assignment is what lets experiments support cause-and-effect claims rather than mere associations.
Is every study that compares two groups an experiment?
No, comparing two groups does not make a study an experiment; what matters is how people ended up in each group. If the researcher randomly assigned participants, for example to a 20-minute meditation session or a 20-minute reading session, the study is experimental. If participants chose their group, were born into it, or already belonged to it, the comparison cannot support the same causal claim. When reading a method section, look first for the sentence describing assignment.
What makes a good dependent variable?
A good dependent variable is sensitive, reliable, and valid. Sensitive means it registers real changes when they occur rather than being too coarse to move. Reliable means it gives consistent readings. Valid means it actually captures the construct of interest. In a study of sleep deprivation and memory, choosing recall accuracy as the dependent variable defines which aspect of memory the experiment can speak to.
Why does an independent variable need at least two levels?
An independent variable needs at least two levels because an effect can only be detected by comparison; with a single level there is nothing to compare. Typical levels are presence versus absence of a treatment, or different doses of a drug. Manipulating the variable across levels is what shows that cause and outcome vary together. If the researcher only observes variation that arose for other reasons, the study is correlational, and no later statistical analysis can make it experimental.

About the Research Methods in Psychology course

How many college credits is Research Methods in Psychology?
Research Methods in Psychology (Psychology 111) carries a NCCRS credit recommendation for 3 semester credits. ACE and NCCRS have over 2,100 cooperating colleges that will consider accepting credits; the final decision is with the universities.
How much does Research Methods in Psychology cost and what is included?
$250, paid once, with lifetime access: all 94 video lessons, a practice quiz after each, the graded work and final exam, and an official transcript when you finish. Students who want several courses choose UPI Study PRO instead.
Which universities accept UPI Study credits?
UPI Study has signed partner agreements with universities such as Thomas Edison State University, University of the People and Charter Oak State College, and works with universities such as Penn State, SNHU and SUNY. ACE and NCCRS have over 2,100 cooperating colleges that will consider accepting credits; the final decision is with the universities. Use "Find my college" on this page to check yours.
How long does Research Methods in Psychology take?
It is self-paced with no deadlines. The 94 lessons total about 16 hours of video; most students add reading and practice time and finish in a few weeks to a few months, on their own schedule.
How are the exams done?
Quizzes and assignments are online in the course. The final exam is taken online with proctoring; a person on the UPI Study team checks assignments, usually within two days.
How do my credits reach my college?
When the course shows 100% complete, request the official transcript from your dashboard. It is sent to the college you name, and you can send it to more than one. Check your college first with "Find my college" on this page, or ask Athena.

Ready for the full course? Earn 3 credits in Research Methods in Psychology.

94 video lessons, a quiz after each one, and the graded work that earns the credit. Recommended for credit by NCCRS.

Enroll in Research Methods in Psychology →Download the syllabusEvery course with PRO →
Research Methods in Psychology
3 credits · NCCRSEnroll →