Why Random Assignment Matters | Research Methods in Psychology

UPI Study › Courses › Research Methods in Psychology › Free lessons › Lesson 12.2 Free with a free account

Why Random Assignment Matters

Research Methods in Psychology · Chapter 12: Experimental Design FundamentalsSection 12.2 · lesson 71 of 949 min · free with a free account

← The Essential Elements of Experimental DesignNext lesson: The Purpose and Power of Control Groups →

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 single most important procedural step in experimental research.

  • Explain the logical role of random assignment
  • Separate random assignment from random sampling
  • Apply simple, block and stratified procedures
  • Recognize when randomization fails, and why

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 (1,227 words). Read along, or search it.

Experimental Design Fundamentals

Why Random Assignment Matters

If you had to protect just one step in experimental design, this would be it. Random assignment is the single most important procedural step in experimental research. Not the most sophisticated. The most load-bearing. By the end of this section you will explain its logical role in supporting causal inference. You will distinguish it from random sampling, which sounds similar and does something else. You will apply the three common procedures, simple, block, and stratified. And you will recognize the situations where randomization fails, or is simply impractical. Then you can judge whether a published study used it correctly.

The Problem

Two Groups That Differ in Everything

Start with the problem that random assignment exists to solve. A researcher wants to test whether a new study technique improves exam performance. She forms two groups. Students who already use the technique, and students who do not. The groups differ in their average exam scores. So far, so promising. The trouble is that they probably differ in many other ways too. Motivation. Prior achievement. Study time. Family background. Any one of those differences could explain the gap on its own. And notice that the groups formed themselves, by self-selection. Without some way of equating the groups before treatment, the comparison is uninterpretable. The score difference is real. The explanation is simply not available to her.

The Solution

Remove the Choice of Who Goes Where

Random assignment solves this by removing the choice of who goes where. Instead of letting participant characteristics determine the condition, a chance procedure does it. And that procedure is blind to those characteristics. It does not know who is motivated, or who did well last semester. Across many trials, this distributes individual differences across conditions in roughly equal measure. Here is the phrase to hold on to. Known and unknown. It handles the differences you thought of, and the ones you never considered. What remains as the systematic difference between groups is the manipulation itself. That is the whole logic, in one sentence.

A Common Confusion

Assignment Is Not Sampling

Two procedures share a name and a tool, randomness. They serve different purposes, and answer different questions. Random sampling randomizes the selection of participants from a larger population. The question it answers is whether the findings generalize beyond the sample. That is external validity. Random assignment randomizes the allocation of those participants into conditions. The question it answers is whether the manipulation caused the effect. That is internal validity. Confusing the two is a common error among students reading their first methods papers. So keep the two verbs separate. Selecting, and allocating.

One Without the Other

The Recurring Trade-Off

A study can use one without the other, and most studies do. Take a laboratory experiment that recruits an opportunity sample of undergraduates. It then randomly assigns them to conditions. That study has strong internal validity, but limited external validity. You know the manipulation caused the effect, in undergraduates. Now take a national survey that samples randomly from the population. It manipulates nothing, so it has good population level inference and no causal claims. Strong studies aim for both, of course. But the trade-off between them is a recurring feature of empirical work.

Procedure One

Simple Random Assignment

There are three common ways to actually carry out random assignment. The first is simple random assignment. Every participant has an equal probability of being placed in any condition. There are no constraints at all on the resulting group sizes. A coin flip works for two conditions. A random number generator handles more. The procedure is unbiased, which is exactly what you want. But no constraints has a cost. It can produce unequal group sizes purely by chance. That happens particularly with small samples, where a run of luck is not unusual. Unequal sizes matter more for some statistical analyses than others.

Procedure Two

Block Random Assignment

The second procedure is block random assignment. It guarantees equal group sizes by working in batches. Suppose you have four conditions. The researcher generates random orderings of the conditions, in blocks of four. Each successive participant is assigned according to that ordering. So by the end of every block, each condition has received exactly one participant. Notice what this does and does not give up. The order within a block is still random, so the logic of randomization is preserved. What you gain is balance, guaranteed rather than hoped for. That is especially valuable in small studies, and where analysis needs equal cell counts.

Procedure Three

Stratified Random Assignment

The third procedure is stratified random assignment. Use it when a known characteristic is likely to affect the dependent variable. Gender, for example. Or age band. Or a pre-test score. The researcher stratifies the participant pool on that variable. Then randomizes separately within each stratum. The result is a guarantee where you most need one. Each group contains a balanced mix of the stratification variable. Meanwhile randomization is preserved on every other characteristic. This is particularly useful when sample sizes are modest. And when the stratifying variable is strongly related to the outcome.

Not Random

Convenient Is Not the Same as Unbiased

Now a warning about procedures that feel random but are not. Letting the first ten arrivals form the treatment group, for instance. Or assigning participants by birthday. Both are convenient. Neither is random. First arrivals may be more punctual, or more motivated, than the people who show up late. And birthdays correlate with school placement, so they carry information about background. Any rule that quietly tracks a participant characteristic is a selection rule. Real random assignment requires an unbiased mechanism, such as a random number generator. It also requires pre-specified rules, applied before any data are collected. Deciding the rule afterwards is not randomization. It is choosing.

Residual Imbalance

When Chance Still Leaves a Gap

Even a properly conducted randomization can leave you with unbalanced groups. Purely by chance, the groups can differ on an important variable. This is more likely when sample sizes are small. It is also more likely when many individual differences exist in the participant pool. Researchers handle that residual imbalance in two ways. First, they run pre-treatment checks on key variables. Then they report any imbalance honestly, rather than quietly hoping nobody looks. Second, they can include those variables as covariates in the analysis. That statistically adjusts for whatever imbalance the randomization happened to produce.

Serious Failures

When Randomization Breaks Down

More serious failures happen when randomization breaks down in practice. Sometimes experimenters reassign participants, to balance ongoing recruitment. That is well intentioned, and it destroys the procedure. Sometimes participants drop out of one condition more than another. That is called differential attrition, and it is easy to miss. The groups were equivalent at the start, and are not equivalent at the end. And sometimes intact groups, like whole classrooms, are used instead of individual assignment. Each of these failures reintroduces the very confounding that randomization was meant to eliminate. So the procedure has to hold for the whole study, not just the first day.

Why It Cannot Be Skipped

Protection Against the Unknown

The strongest argument for random assignment is what happens without it. Groups that differ for any reason other than chance carry that reason with them. Voluntary enrollment. Geographic convenience. Scheduling. Each leaves open dozens of alternative explanations for any observed difference. Statistical controls help, but they have a hard limit. They only work for variables the researcher thought to measure. Random assignment protects against variables the researcher never identified. It can do that because chance does not select on any criterion at all. That is why it is sometimes called the gold standard of experimental design. But it is a tool for specific questions, not for all of them. Traumatic events, biological sex, and economic deprivation cannot be assigned.

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 Why Random Assignment Matters

Why is random assignment important in experiments?
Random assignment is important because it lets a chance procedure, not participant characteristics, decide who receives each condition. Pre-existing differences in motivation, ability, or background are then spread across groups without systematic bias, including differences nobody thought to measure. The only systematic difference left between groups is the manipulation, which justifies a causal conclusion. It does not make groups identical, and chance imbalances can still occur, especially in small samples.
What is the difference between random assignment and random sampling?
Random sampling selects participants from a population by chance and supports external validity, the ability to generalize to that population. Random assignment allocates participants already in a study to conditions by chance and supports internal validity, the ability to infer causation. A lab experiment with an opportunity sample of undergraduates may use random assignment without random sampling, while a national survey may sample randomly without assigning anyone to conditions.
What are simple, block, and stratified random assignment?
Simple random assignment gives every participant an equal chance of each condition, as with a coin flip, but can produce unequal group sizes in small samples. Block random assignment randomizes the order of conditions within successive blocks, such as blocks of four for four conditions, guaranteeing equal group sizes. Stratified random assignment randomizes separately within levels of a key variable, such as gender or pretest score, so each group is balanced on it.
Is assigning participants by arrival order random assignment?
No, assigning participants by arrival order, birthday, or any similar rule is not random assignment. The first people to arrive may be more punctual or motivated than latecomers, and birthdays can relate to school placement and background. Any rule that tracks a participant characteristic is a selection rule. True random assignment uses an unbiased mechanism, such as a random number generator, with rules fixed before data collection begins.

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 →