Defining and Understanding Quasi-Experimental Designs

UPI Study › Courses › Research Methods in Psychology › Free lessons › Lesson 7.1 Free with a free account

Defining and Understanding Quasi-Experimental Designs

Research Methods in Psychology · Chapter 7: Quasi-Experimental and Time-Based DesignsSection 7.1 · lesson 40 of 9411 min · free with a free account

← Capturing Lived ExperienceNext lesson: Working with Ex Post Facto Research →

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

Studying interventions and changes when random assignment is impossible.

  • Place quasi-experiments on the design spectrum
  • Identify when random assignment is unavailable
  • Distinguish the major quasi-experimental designs
  • Recognize and address threats to internal validity

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 (1,496 words). Read along, or search it.

Quasi-Experimental and Time-Based Designs

Defining and Understanding Quasi-Experimental Designs

Random assignment is the engine of causal inference. But some of the most consequential questions in social science cannot use it. You cannot randomly assign children to a school reform. You cannot randomly assign citizens to a national health policy. This section is about what you do instead. Quasi-experimental designs keep the comparison, and give up the randomizer. By the end, you will be able to place them on the design spectrum. You will recognize the conditions that make them necessary. You will distinguish the major families of quasi-experimental design. And you will name the threats to internal validity that every one of them must manage.

Between Two Worlds

When the Gold Standard Is Out of Reach

Start with what a true experiment actually does. The researcher manipulates an independent variable, assigns participants at random, and measures the outcome. The random assignment is doing something specific, and it is worth being precise about it. It ensures that differences between the groups before the intervention are due only to chance. That leaves the manipulation as the most plausible explanation for any difference afterwards. Now think about where research actually gets commissioned. A new educational policy is introduced across an entire state. A hospital reorganizes its emergency department. A corporation rolls out a new training program to all of its staff. In none of these can you allocate people to conditions by chance. Quasi-experimental research exists in exactly this gap. It keeps the logic of comparison, and lets go of the coin flip.

The Defining Feature

Manipulation Without Random Assignment

So hold the two designs side by side. In a true experiment, participants are allocated to conditions by chance. Any group difference before treatment is therefore just random variation. In a quasi-experiment, people end up in conditions through some non-random process. They select themselves, or they happen to live in a particular place. Or an institution decided for them, or they already differed on some characteristic. Here is the part students most often get backwards. A quasi-experiment can still involve real manipulation. Something genuinely was done, and often deliberately. What is missing is control over who receives it. That single missing ingredient is the whole difference between the two designs. Everything else, the comparison, the measurement, the inference, still follows experimental logic.

The Spectrum

Three Positions on One Continuum

It helps to picture research designs on a single line. At one end sits non-experimental work. Surveys and archival studies, where nothing is manipulated at all. Causal claims from that end are weak, because nothing rules out reverse causation. At the other end sits the true experiment. The researcher both manipulates the variable and randomly assigns participants. Quasi-experimental designs occupy the middle of that spectrum. They are stronger than purely descriptive or correlational work. They are weaker than true experiments. That middle position is not a failure. It is a deliberate trade, and the next slides say what you get for it.

Side by Side

What Each Design Family Buys You

Four features separate the three design families cleanly. Manipulation of the independent variable. A true experiment always does it. A quasi-experiment often does. Non-experimental work does not. Random assignment is present only in the true experiment. Neither of the other two has it. Causal inference follows directly from that. Strong for the true experiment, moderate for the quasi-experiment, weak for non-experimental work. The settings differ too. Laboratories and randomized controlled trials on one side. Field studies and policy evaluation in the middle, surveys and archival work on the other side. Read that middle column carefully, because it carries the definition. Often yes on manipulation is what separates a quasi-experiment from a survey. Something really was done. It just was not assigned at random.

When You Need Them

Five Circumstances That Force the Choice

Five circumstances commonly force a researcher into a quasi-experimental design. The first is ethics. A study of how a school reform affects students cannot randomly assign children. The second is practicality. Evaluating a national health-insurance change cannot randomly assign citizens. The third is the unit of delivery. A new safety protocol gets rolled out to an entire factory or hospital, not to individuals. The fourth is the variable itself. Nobody can assign a region its climate, its language, or its history. The fifth is opportunity. Earthquakes, policy changes, and organizational restructurings create natural experiments. Notice what all five share. The constraint comes from the world, not from the researcher's laziness.

The Trade

Internal Validity for Ecological Validity

There is a real compensation for giving up random assignment. Quasi-experiments happen in the settings where policies and programs actually operate. Real schools, real hospitals, real workplaces. So their findings often translate more directly into practice. A tightly controlled laboratory result may not survive contact with a messy institution. Two terms let you name the trade precisely. Internal validity is the extent to which the study supports the conclusion that the intervention caused the effect. Not something else. The intervention. Ecological validity is the extent to which findings reflect what happens in real-world settings. The loss of internal validity here is genuine, and you should not pretend otherwise. But the gain in ecological validity is genuine too. For many applied questions, it is the larger of the two.

The Families

Six Common Quasi-Experimental Designs

Quasi-experimental research is not one design. It is a family of them. The non-equivalent control group design compares a treated group with an untreated comparison group. The two were never randomly assigned, hence non-equivalent. The single-group pretest-posttest design measures outcomes in one group, before and after. An interrupted time series tracks an outcome at many points on both sides of a change. You look for a discontinuity right at the moment of the intervention. Ex post facto designs compare groups that already differ on a characteristic nobody could assign. Regression discontinuity compares units just above and just below a cut-off score. The cut-off is what determines who receives the treatment. And the natural experiment treats a real-world event as if it were a planned manipulation. A policy change, or a natural disaster. Each of these suits a different evidentiary situation, and later sections take several of them in depth.

Rival Explanations

Four Threats to Internal Validity

Give up random assignment, and rival explanations rush in. Each one competes with your intervention as the cause of the change you observed. Selection is the first and the biggest. The treated and untreated groups differed in important ways before anything happened. History is the second. Other events during the study period affect outcomes alongside, or instead of, the intervention. Maturation is the third. Participants change naturally over time, growing older and gaining experience. If you measure children a year apart, some improvement was going to happen anyway. Testing is the fourth, and it is the sneakiest. Repeated measurement can itself influence later scores, regardless of any intervention. People get better at the test, not better at the thing the test measures.

Rival Explanations

Three More, and the Standard That Matters

Three more threats complete the list. Instrumentation. Changes in the measuring instrument or the procedure produce apparent effects. If your raters get stricter over the year, outcomes appear to worsen for no real reason. Regression to the mean. Extreme initial scores tend to drift toward average on retest. This one is a trap, because programs are usually given to the most extreme cases. Those cases would have improved somewhat with no help at all. Attrition. Participants who drop out may differ systematically from those who remain. Your final sample is then not the sample you started with. Now the standard that actually matters. Quasi-experimental work is judged less by whether threats exist, because they always do. It is judged by how thoroughly the researcher identifies, measures, and addresses them. A study that takes its threats seriously can produce credible causal evidence. One that ignores them produces nothing stronger than a correlation.

Building Strength

Six Strategies That Sharpen the Inference

So how do you actually address them? Six strategies do most of the work. Pretest measurement attacks selection, by establishing what the baseline differences were. You cannot correct for a gap you never measured. Comparison groups attack history and maturation. If both groups live through the same outside events, those events cannot explain a difference between them. Multiple measurement points attack maturation and regression to the mean. Many observations reveal the underlying trend, so a dip and a rebound stop looking like an effect. Statistical control of covariates handles selection effects on the variables you measured. Note the limit built into that sentence. Only the measured ones. Replication across settings guards against the quirks of any single site or sample. And triangulation with other methods guards against the limitations of any single design. Used together, these turn a weak comparison into a defensible causal argument.

Putting It Together

The Vocabulary You Now Own

Three definitions are worth holding onto. Quasi-experimental research compares groups or conditions to estimate intervention effects. It does so without random assignment. That clause is the definition. Manipulation is the deliberate introduction or alteration of an independent variable. It can come from the researcher, or from a real-world process. A natural experiment is an opportunity created by a real-world event. The researcher treats it as if it had been a planned manipulation. Carry one question into every quasi-experiment you read from here on. Which threats apply to this study, and what did the researcher do about each one? Answering that is the entire skill.

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 Defining and Understanding Quasi-Experimental Designs

What is a quasi-experimental design?
A quasi-experimental design compares groups or conditions to estimate the effect of an intervention without random assignment. Something is often genuinely manipulated, such as a new school policy or hospital protocol, but people end up in conditions through self-selection, location, institutional decisions, or pre-existing differences. Quasi-experiments sit between correlational studies and true experiments in the strength of the causal conclusions they support.
What is the difference between a true experiment and a quasi-experiment?
The key difference is random assignment. In a true experiment, the researcher manipulates the independent variable and assigns participants to conditions at random, so pre-existing group differences reflect only chance. A quasi-experiment may also involve manipulation but lacks control over who receives it. As a result, it usually has weaker internal validity but often stronger ecological validity, because it takes place in real schools, hospitals, and workplaces.
When do researchers use quasi-experimental designs?
Researchers use them when random assignment is impossible, unethical, or impractical. Common situations include school reforms that cannot ethically be assigned to individual children, national policies that cannot be randomized across citizens, interventions delivered to an entire factory or hospital, variables like a region's climate or history, and events such as earthquakes or policy changes that create natural experiments.
What is a regression discontinuity design?
A regression discontinuity design compares units just above and just below a cut-off score that determines who receives a treatment. Because people near the threshold are otherwise similar, a jump in outcomes right at the cut-off suggests a treatment effect. It is one of several quasi-experimental designs, along with non-equivalent control groups, interrupted time series, ex post facto comparisons, and natural experiments.

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 →