C226 Research Design and Analysis, catalog number EDUC 5113, is the two-CU third course in the School of Education research sequence, and it is the one where the reasoning gets tested rather than described. Two things are assessed: designing an empirical study that will actually produce valid results, and drawing interpretations from the resulting data that a sceptical reader would accept. Those are separate skills, and educators tend to be much stronger at the first than the second.
Design is a set of decisions, not a label
Students often think design means choosing a name: experimental, quasi-experimental, correlational, case study. The name is the last thing that gets decided, and it emerges from choices made earlier. Who is in the study and how did they get there. What is measured, with what, and when. What is compared with what. What else could explain the result you expect to see.
That last question is the whole of research design compressed into a sentence. Every design is an argument that a rival explanation has been ruled out. Random assignment rules out preexisting group differences. A pre-test rules out the possibility that groups started apart. A comparison group rules out the chance that everyone improved anyway over eight weeks of normal teaching. When you write a design section, you are not describing a procedure. You are listing the rival explanations and showing which design feature disposes of each.
Classroom research is where this gets uncomfortable, because a teacher cannot randomly assign pupils and would not want to. That constraint is not a flaw to hide; it is a condition to design around and state honestly. A quasi-experimental design with a stated threat and a stated mitigation is stronger work than an experimental design that could not have happened.
The analysis half of the course is where the returns cluster. Analysis has to be chosen before data exists, has to match the type of data being collected, and has to answer the research question rather than an adjacent one. A question about whether a group changed needs a comparison across time. A question about whether two things move together needs an association. A question about what an experience meant to participants needs a coding process with rules, not a paragraph of impressions. Choosing an analysis after seeing the data is how findings become unpublishable.
Turning scored aspects into a section plan
Scored detail lives inside your Course of Study, not in the public catalog, so open your rubric first and count what is actually being assessed. Every aspect is scored independently against a three-point scale and a score of 2 in each aspect passes a task. Nothing compensates for anything else, which in a design course means a beautiful sampling section cannot rescue an analysis plan that does not fit the data.
Where your version is assessed by submitted work, use the rubric's own nouns for headings and give each scored aspect its own home. Design writing invites long integrated paragraphs where sampling, instrument and analysis blur together. That style makes an evaluator hunt, and hunting is how aspects get scored as unmet.
The word budget, worked. Take a rubric with six scored aspects and directions calling for roughly 1,600 words. Reserve 120 words for an opening that restates the research question exactly as C225 left it, and 100 for a close, leaving about 1,380 for the scored body. Six into 1,380 is 230 words per aspect. Now weight it: the validity threats aspect wants nearer 350 because each threat needs naming, explaining and answering, and the participants aspect is complete at about 150 once population, sampling method, number and access are stated. Take the difference from the descriptive aspects and the total holds.
One planning discipline specific to this course. Write the analysis section before you write the data collection section, then go back and check that every piece of data your analysis needs is actually being collected. Doing it in the other order produces designs that gather information nothing will ever use and omit the one variable the analysis required.
A structure that fits a design and analysis task
Where the directions specify their own arrangement, follow it. Where they do not, this order makes the alignment chain visible, and alignment is what evaluators are checking.
| Section | What belongs in it | How it gets scored |
|---|---|---|
| Question restated | The research question verbatim, with variables or constructs identified | Not always scored alone, but every alignment judgment starts here |
| Design named and justified | The design, and why this question requires it rather than a neighbouring one | Scored for fit; the name alone earns little without the reasoning |
| Participants and sampling | Population, how participants are selected, how many, and how access is obtained | Scored for feasibility and for honesty about convenience sampling |
| Instruments | What measures or collects the data, and evidence that it measures the intended construct | Scored on the link between instrument and construct, not on instrument quality alone |
| Procedure and timeline | What happens, in order, with when and how often stated | Scored for replicability; another teacher should be able to run it |
| Analysis plan | What will be done with the data, matched to data type and to the question | The highest-return aspect in the course when it does not match the data |
| Threats to validity | Named threats, each with the design feature or limitation statement that answers it | Scored for specificity; a generic limitations paragraph scores as unmet |
| Interpretation rules | What result would support the question, and what result would not | Scored where the rubric asks for defensible interpretation |
| Ethics | Consent, pupil anonymity, and your dual role as teacher and researcher | Scored wherever ethics is named; teacher-researcher conflict belongs here explicitly |
The interpretation rules row is the one experienced educators skip and the one that most improves a submission. Stating in advance what would count as the intervention not working is the clearest possible evidence that you are conducting research rather than gathering support.
Evidence craft when the data has not been collected yet
Design work has a peculiar evidence problem: the study is hypothetical, so every claim has to be warranted by methodological authority or by the literature rather than by results.
- Warrant design choices with methods sources. A statement that a pre-test and post-test arrangement controls for a specific threat needs a citation, not confidence.
- Warrant expectations with the literature from C225. If you expect an effect, the reason lives in the studies you already reviewed.
- Keep hypothetical data clearly hypothetical. Where a task asks you to interpret sample data, say plainly that the numbers are illustrative. Presenting invented results as real classroom findings is a serious integrity problem, not a shortcut.
- De-identify everything. Pupils, colleagues and schools should be unidentifiable in the document, and saying that you have done so is part of the ethics aspect.
- Report any statistics with the test, the numbers behind them and the meaning in plain words. A figure with no interpretation and an interpretation with no figure both score as half an answer.
- Match citation style to your directions and check the reference list against every in-text citation.
A habit that separates strong work: naming what the design cannot show. A single-classroom study cannot support a claim about a district, and an eight-week study cannot support a claim about lasting change. Saying so, then proceeding, is what defensible interpretation means in practice.
What separates Competent from a submission sent back
Aspects are scored independently, so a returned design task is usually one broken link in the chain rather than a bad plan.
- Every scored aspect has its own heading in the rubric's own wording.
- The question, the design, the instrument and the analysis all align, and each section names the one before it.
- Every threat to validity is a named threat with an answer attached, not a general acknowledgment that limitations exist.
- The analysis matches the data type. Counting things that were never counted, or averaging categories, is an instant return.
- Interpretation is bounded by the design, so no conclusion reaches past what the study could show.
- The teacher-researcher role is addressed explicitly rather than left implied.
Performance assessment work at WGU can be revised and resubmitted with no grade penalty, so a return costs calendar rather than standing. In a six-month flat-rate term running a four-course sequence, calendar is exactly what you cannot spare. Design problems also propagate: a misaligned analysis plan here becomes a proposal in C227 that cannot be defended without redesigning the study.
If your plan attaches a proctored objective assessment to this course, the line is absolute. Proctored exams are yours to sit. We prepare only: design decision drills, practice matching analyses to data types, worked interpretation exercises and an honest readiness call. We do not sit assessments and we never ask for portal credentials.
Six mistakes that cost time in C226
- Naming a design before making the decisions. The label follows from sampling, comparison and timing choices. Choosing the label first produces a study that does not match its own name.
- Claiming random assignment in a classroom. Teachers rarely can, and an unrealistic claim damages a submission more than an honestly stated constraint would.
- Choosing analysis to suit the data type you wish you had. Ordinal ratings are not interval measurements, and treating them as such is a return waiting to happen.
- Writing limitations as a ritual paragraph. Each threat needs a name and an answer. A sentence saying results may not generalise is not a threat analysis.
- Letting the research question drift from C225. If the question changed, say so and justify it, because otherwise the design answers something the literature review never set up.
- Skipping the interpretation rules. Deciding after the fact what counts as success is the failure mode this course exists to prevent.
How support works on this course
Send the rubric from your Course of Study, the task directions and the research question you carried out of C225. Work comes back aspect-mapped with the alignment chain visible: question to design to instrument to analysis, each threat named and answered, interpretation bounded by what the study can support, and the ethics of researching your own pupils handled directly. The walkthrough shows where each decision was made and why, so the same reasoning survives into the proposal course.
If your question has shifted since C225, bring both versions. Reconciling them now is far cheaper than discovering the mismatch in C227, where the proposal has to hold every earlier decision together in one document.
Questions students ask about C226
Is C226 the same course as EDUC 5113?
Do I have to run the study I design in C226?
Can you take a proctored assessment for this course?
Designing a study that has to hold together?
Send your Course of Study rubric, the task directions and your research question. You get an aspect-mapped design with the alignment chain visible and every validity threat answered.
Where C226 sits in WGU's programs
The July 2026 catalog places this code in 1 current WGU program. Open a program page for the complete standard path and term positions. The live Degree Plan remains authoritative after transfer credit, substitutions, and mentor planning.
The assessments, one by one
The public catalog does not publish this course's PA/OA identity or task count. WGU Tutors publishes at most one PA manual per course and only from a WGU-controlled public rubric. Until that source exists, PA help begins from the student's real Course of Study and OA support remains preparation only.