Getting Started with Factorial Designs, Explained

UPI Study › Courses › Research Methods in Psychology › Free lessons › Lesson 12.7 Free with a free account

Getting Started with Factorial Designs

Research Methods in Psychology · Chapter 12: Experimental Design FundamentalsSection 12.7 · lesson 76 of 949 min · free with a free account

← Designing Studies with Multiple GroupsNext lesson: Different Variations of Factorial Designs →

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

Manipulating two or more variables together to reveal how they combine.

  • Define a factorial design and its structural features
  • Read factorial notation such as 2 by 2 and 3 by 4
  • Distinguish main effects from interaction effects
  • Recognize the limits of expanding factorial complexity

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 (1,342 words). Read along, or search it.

Experimental Design Fundamentals

Getting Started with Factorial Designs

Up to now, every design in this chapter has manipulated one variable. This section manipulates two, and that changes what you can find. By the end you will define a factorial design and identify its structural features. You will read and interpret factorial notation, like two by two, or three by four. You will distinguish main effects from interaction effects, and explain why each matters. You will set up a simple factorial experiment with appropriate cell counts. And you will recognize the conceptual and practical limits of factorial complexity. That last one saves more studies than it sounds like it should.

Why Two at Once

Single Causes Are Rare

The world rarely offers psychological phenomena as products of single causes. Whether a study technique works may depend on the type of material being learned. Whether a medication helps may depend on the patient's age. Whether a persuasive message succeeds may depend on the audience's prior beliefs. Notice the word depends in all three sentences. That word is the thing a one variable design cannot capture. Designs that manipulate one variable at a time can describe each effect in isolation. What they cannot do is reveal how the variables combine. Run two separate studies and you will get two separate answers, and no relationship between them.

The Real Advantage

Efficiency Is the Lesser Reason

Factorial designs address that limitation directly. They manipulate two or more independent variables in the same experiment. There is an efficiency argument for doing this, and it is true as far as it goes. Factorial designs extract more information from a fixed sample than separate experiments would. But efficiency is the lesser reason. The deeper advantage is that factorial designs can detect interactions. An interaction is a pattern where the effect of one variable depends on the level of another. And here is why that matters so much for psychology in particular. Many of the most theoretically important findings in the field are interaction effects. The factorial design is the standard tool for finding them.

Notation

Reading a Factorial Label

Factorial designs are described by a string of numbers separated by multiplication signs. Each number represents one independent variable. The value of the number is how many levels that variable takes. A two by two design has two independent variables, each at two levels. That produces four conditions. A two by three design has one two-level and one three-level variable, producing six conditions. A three by four by two design has three variables, producing twenty four conditions. The rule underneath is simple, and it is the one to memorize. The total number of conditions is the product of the levels. Multiplication, not addition. That is why these designs grow so fast.

Effect Type One

Main Effects

Factorial designs decompose the data into two distinct kinds of effect. The first is the main effect. A main effect is the average effect of one independent variable. Averaged across the levels of all the other independent variables. Take a two by two design crossing teaching method with class size. Lecture versus discussion, and small versus large. The main effect of teaching method compares all lecture students against all discussion students. Class size is ignored entirely for that comparison. There is one main effect for each independent variable in the design. So a two by three has two main effects, and a three by four by two has three.

Effect Type Two

Interaction Effects

The second kind of effect is the interaction. An interaction is a pattern in which one variable's effect depends on the level of another. Return to the teaching example, one layer down. Discussion might produce better outcomes than lecture in small classes. And worse outcomes than lecture in large classes. Now look at what the main effects would report. Averaged together, the two patterns cancel, so the main effects could be small or zero. Yet the interaction would reveal that the right method depends completely on context. And here is the crucial structural point. Interactions cannot be detected by separate experiments on each variable in isolation. The variables have to be manipulated together, in one study.

A Reporting Trap

When a Main Effect Hides the Truth

That cancellation has a reporting consequence worth spelling out. If one variable's effect depends strongly on another, the main effect can obscure the pattern. Not just fail to show it. Actively mislead about it. Reporting only that there was no main effect of teaching method sounds like a null result. But it may hide the fact that method matters a great deal. It just matters in opposite directions for different class sizes. So when an interaction is present, main effects must be interpreted in light of it. And they are often decomposed into simple effects. That means examining one variable's effect within each level of the other, separately.

Setting It Up

Cells, Not Groups

Setting up a factorial experiment introduces one new unit, the cell. Participants are randomly assigned to one of the cells defined by the factorial structure. In a two by two design that means four cells. In a two by three, six. Now the requirement that catches people out. The total sample must support adequate statistical power within each cell. Not just across the overall design. A hundred participants sounds comfortable until you divide it four ways. As factorial complexity grows, the number of cells grows multiplicatively. And the per-cell sample needed to detect interactions can quickly exceed practical limits.

The Arithmetic

How Complexity Scales Sample Size

Put some numbers on that, holding the cell size at twenty five participants. A two by two design has four cells, so you need one hundred participants. A two by three has six cells, and needs one hundred and fifty. A three by three has nine cells, and needs two hundred and twenty five. A two by two by two has eight cells, needing two hundred. Then watch what one more three-level factor does. A three by three by three has twenty seven cells, and needs six hundred and seventy five. That is three times the demand of the three by three. And a four factor design, two by two by two by two, has sixteen cells and needs four hundred. None of those numbers are unusual. They are just multiplication doing its work.

The Ceiling

The Limits of Factorial Expansion

Adding factors multiplies your capacity to detect interactions, at considerable cost. The first cost is interpretability. Three-way and higher-order interactions become increasingly difficult to interpret. Think about what a significant three-way interaction actually claims. It says the two-way interaction itself depends on a third variable. That is rarely a clean theoretical statement, and it is hard to explain to anyone. The second cost is sample size, which rises rapidly. A four factor design at twenty five per cell already requires four hundred participants. So most published factorial experiments use two factors. Some use three. Designs with four or more are rare, and usually justified only by specific theoretical commitments.

A Planning Error

Do Not Add Factors Hopefully

A common error in early study planning is to add factors hopefully. Throwing in every variable that might matter, in case one of them does. The result is usually an underpowered design. One that fails to detect even the main effects of theoretical interest. You asked for everything, and you come away with nothing. The discipline that works runs the other way. Start with the simplest design that addresses your central question. Then add factors only when there is a specific theoretical reason for each one. And only when you have adequate resources to power the extra cells. That approach produces more interpretable studies, and stronger evidence.

Reading Results

Tables, Graphs, and Parallel Lines

Finally, how factorial results reach the page. They are typically presented in two forms. First, a table of means, with marginal averages for each factor. Second, a graph showing the cell means. One factor sits on the horizontal axis, and the other is shown as separate lines. That graph gives you a visual test you can run in a second. Parallel lines indicate the absence of an interaction. Lines that are not parallel suggest an interaction is present. The statistical analysis uses analysis of variance. It produces a separate F test for each main effect, and for each interaction. And when an interaction is significant, main effects are read in its light. Often by examining the simple effect of one variable at each level of the other.

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 Getting Started with Factorial Designs

What is a factorial design in psychology?
A factorial design is an experiment that manipulates two or more independent variables at the same time, crossing every level of each with every level of the others. Besides getting more information from a fixed sample, it can detect interactions, where the effect of one variable depends on the level of another. Separate one-variable experiments cannot reveal how variables combine, which is why factorial designs are the standard tool for finding interactions.
How do you find the number of conditions in a factorial design?
Multiply the numbers in the design's notation, because each number stands for one independent variable and its value is that variable's number of levels. A 2 x 2 design has four conditions, a 2 x 3 design has six, and a 3 x 4 x 2 design has 24. How many numbers appear tells you how many independent variables there are, and their product tells you how many cells.
How many participants does a factorial experiment need?
Total sample size is the number of cells times the participants needed per cell, so it grows quickly as factors are added. With 25 participants per cell, a 2 x 2 design needs 100, a 3 x 3 needs 225, a 2 x 2 x 2 needs 200, and a 3 x 3 x 3 needs 675. Power has to be adequate within each cell, not just across the whole design.
Why do researchers avoid factorial designs with many factors?
Designs with many factors become hard to interpret and expensive to power. A three-way interaction claims that a two-way interaction itself depends on a third variable, which is rarely a clean theoretical statement, and a 2 x 2 x 2 x 2 design at 25 per cell already needs 400 participants. Adding factors just in case usually yields an underpowered study, so start simple and add factors only for a specific theoretical reason.

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 →