Science Practice 4: Argumentation

Teaching issue: a learning center wants a policy for adults studying unfamiliar vocabulary. Should it replace some rereading with retrieval attempts spread across days, or replace every form of instruction with that method? Use the three source summaries below to evaluate a bounded recommendation. They are complete original hypothetical teaching sources, not published studies, quotations, journal articles or official exam sources. Their labels are citation aids within this packet only.

Source 1 · Original hypothetical retrieval experiment: researchers ask whether practicing recall improves one-week vocabulary memory relative to rereading. Eighty adult volunteers aged 18–24 respond to an advertisement at one university. A computer randomly assigns 40 to retrieval practice and 40 to rereading. Everyone first studies the same 24 unfamiliar word–meaning pairs for four minutes. For the next 12 minutes, retrieval participants see word cues, try to produce meanings, and then receive the correct answer; rereading participants see the complete pairs repeatedly. Both groups work in the same quiet room with identical materials and time. One week later, everyone writes meanings for all 24 cues without notes, with one point for each correct meaning (0–24). A scorer who does not know condition uses a predefined answer key. All 80 complete the test. Mean recall is 18/24 after retrieval and 14/24 after rereading. These are invented study summaries, not actual participants or a published finding. The described safeguards are institutional review before recruitment, plain-language consent explaining possible frustration and withdrawal without loss of compensation, breaks when requested, coded answer sheets with the identity key stored separately under restricted access, and a post-study explanation. Convenience recruitment, ages 18–24 at one university, and a 24-pair recall outcome limit generalization. No p-value or uncertainty interval is supplied in this packet; do not infer one from the mean difference.

Source 2 · Original hypothetical spacing experiment: researchers ask whether spreading retrieval practice across days improves one-week vocabulary recall compared with one concentrated retrieval session. They recruit 120 university volunteers aged 18–24 by campus advertisement and randomly assign 60 to each schedule. Everyone first studies the same 24 unfamiliar word–meaning pairs for four minutes on day one. The spaced group then completes four minutes of cue–attempt–feedback retrieval on each of days one through four; the concentrated group completes the same total 16 minutes of cue–attempt–feedback practice on day four. Both groups finish practice on day four and take the same 24-point cued-recall test one week later. A scorer unaware of schedule applies one answer key. All 120 complete the test. Mean recall is 19/24 after spaced practice and 15/24 after concentrated practice. No significance test or effect-size statistic is supplied. Adults consent after an explanation, may withdraw without penalty and have coded confidential records. The convenience recruitment, narrow age range and word-pair task limit application elsewhere. This study compares schedules within retrieval practice; it does not directly compare spacing with rereading or test the combined policy against every alternative.

Source 3 · Original hypothetical transfer experiment: researchers ask whether a recall advantage for definitions transfers to solving new problems. Eighty adult volunteers aged 18–65 recruited through a community-library notice are randomly assigned 40 each to definition retrieval or rereading. Everyone reads the same short introductory science text for four minutes. During the next 16 minutes the retrieval group uses cues to attempt definitions and receives feedback, while the rereading group sees the complete definitions again. Both receive identical content, time and room conditions. One week later, a scorer unaware of condition uses fixed keys for a 24-point definition-recall test and a separate 12-point set of novel application problems. All 80 complete both. Group means for definitions are 18/24 after retrieval and 14/24 after rereading. Group means for novel problems are 7/12 in both conditions. No significance tests, uncertainty intervals or effect-size statistics are supplied. Review approval, understandable consent, breaks, withdrawal without penalty, confidential coded records and a post-study explanation are described. These convenience volunteers and tasks do not represent every learner or type of understanding. The equal observed problem means do not prove that the true transfer effect is exactly zero.

Read a source in this order: question and participants, selection and design, procedure and outcome measure, result and limitations. This is a teaching procedure, not an official scoring formula. Distinguish a reported result from an explanation you propose. For example, Source 1 reports a recall advantage; it does not measure brain activation, prove every learner benefited or establish improved reasoning on new problems.

A claim states a position about the question; evidence supplies observations that bear on it. A defensible claim can be evaluated using relevant evidence and reasoning. It need not be absolute. It should be specific enough that a reader can tell what action, relationship or consequence is being proposed. ‘Studying is interesting’ does not answer the learning center’s policy question.

A defensible teaching claim for this packet is: ‘For adult vocabulary learners similar to the university volunteers, the center should replace part of passive rereading with retrieval practice spread across several days, while keeping activities that teach and assess understanding.’ This answers the policy issue, identifies the target use and leaves room for task and population limits. Sources 1 and 2 support its practice and schedule recommendations; Source 3 motivates retaining instruction for outcomes beyond definitions.

Compare ‘Retrieval always makes everyone better at everything.’ That claim exceeds the populations, measures and results. Conversely, ‘One group scored 18 and another 14’ merely restates a result; it does not propose a position about the policy. A claim may be an explanatory conclusion or a recommendation, depending on the question. Do not substitute an unrelated moral judgment or a vague endorsement of learning.

A hypothesis predicts a testable relationship before or for a study; a claim takes a position in an argument using the available evidence. They can concern the same relationship, but they play different roles. A claim about a policy’s success also depends on what success means. Here improved delayed vocabulary recall is one outcome; universal transfer, equity of access and implementation costs are separate questions the packet does not fully test.

Scientifically derived evidence comes from systematic investigation using stated methods, measures and observations. In real work, inspect source credibility, methods, relevant findings and limits rather than accepting a claim because it appears scientific or has an attractive graph. The original packet simulates that structure for practice; it is not itself real published evidence. No invented author or journal gives it authority.

Specific evidence for the bounded claim: ‘In the randomly assigned vocabulary experiment, retrieval participants averaged 18 correct meanings out of 24 one week later, compared with 14 after rereading (Source 1).’ This identifies the task, direction and magnitude and attaches the packet citation to the result. ‘Source 1 says retrieval is good’ loses the observation and its relevance.

A weak explanation repeats: ‘This proves my claim because the retrieval group did better.’ A stronger explanation adds the psychological connection: ‘Retrieval practice requires learners to attempt access to stored meanings before checking answers, whereas rereading keeps answers available. Practicing that access helps explain why the retrieval group later recalled more meanings when only word cues were present. The result supports replacing some rereading with recall attempts for similar vocabulary tasks.’ This explains how the observed difference bears on the proposed change.

The explanation uses the concept without pretending the study directly measured an internal mechanism. Random assignment and matched time make the condition comparison more informative, but ‘random assignment,’ ‘mean’ or ‘statistical significance’ alone would be methodological or statistical description, not the psychological application shown here. The bridge is the relationship between practice of retrieval and later cue-based access.

A second bridge for Source 2 uses the spacing effect: practice distributed over time can improve later retention compared with concentrating equal practice time in one session. The spaced and concentrated groups have the same total retrieval time, so the observed 19-versus-15 difference supports the schedule part of the recommendation. Spacing is distinct from retrieval practice: Source 1 compares activities; Source 2 compares timing within the same activity. Neither study measures a special combined synergy between the two.

Supporting a claim means explaining how the evidence strengthens it. Refuting challenges a claim with evidence inconsistent with its assertion. Modifying preserves what the evidence warrants while revising an overbroad, incomplete or conditional part. These are reasoning actions, not three kinds of citation. A source can contribute to different actions depending on the precise claim.

Support: Source 1 strengthens ‘retrieval can improve delayed vocabulary recall compared with rereading for similar learners’ because the randomly assigned retrieval group has higher one-week recall under equal time and task conditions. Refute: the same result challenges the categorical claim ‘retrieval never helps delayed recall.’ Neither use establishes that it helps every individual, and a single source does not settle every implementation question.

Modify: Source 3’s higher definition-recall mean after retrieval, alongside equal observed novel-problem means, narrows ‘better definitions mean better understanding on every task.’ The observed recall advantage does not automatically extend to transfer. Keep retrieval for the supported recall purpose while preserving explanation, problem-solving and assessment of understanding. Equal sample means do not prove an exactly zero population transfer effect, but they do not provide the predicted transfer advantage either.

Nuance can concern population, task, timing, feasibility or the strength of evidence. Source 2 supports spacing under a particular total practice time and delayed test. It does not establish an ideal number of days for everyone. Source 1’s university volunteers do not represent every adult; Source 3 broadens the age range but uses another convenience sample and task. State these particular limits rather than adding a vague ‘more research is needed’ after an otherwise universal claim.

The policy might be effective for one goal and insufficient for another. If the center values remembering vocabulary meanings after one week, Sources 1 and 2 support relevant changes. If it values novel problem solving, Source 3 indicates the center should test that outcome directly instead of assuming recall scores establish it. A warranted recommendation is conditional on goals and evidence; its implementation still needs appropriate access and evaluation.

The Evidence-Based Question uses three supplied sources and a claim supported by two evidence-and-explanation pairs. The current CED rubric totals seven points: A claim (1), B(i) evidence (1), B(ii) explanation/application (up to 2), C(i) different-source evidence (1), and C(ii) explanation with a different psychological application (up to 2). The worked response below is an original teaching demonstration, not an official exam question, official scoring decision or guarantee of points.

For B(i) and C(i), select correctly cited, specific, accurate and relevant evidence. C(i) uses different evidence from either of the other two sources, rather than a second result from the same source used for B. The October 2025 notes clarify that accuracy concerns the general pattern: quoting exact numbers is not required, and small numerical reporting errors can still meet the evidence criterion when the direction is correct. This is not permission to reverse a relationship, change what was measured or invent a result. Accurate reporting remains good practice.

For B(ii), explain the relationship between evidence and claim and apply a psychological perspective, theory, concept or research finding to make the connection. C(ii) uses a different psychological application. The current notes prefer CED psychological content but allow other demonstrated psychological knowledge; popular common-sense claims do not substitute. The rubric’s excluded concepts list accompanies its specific school-start-time sample; it is not a universal blacklist for every possible question. Simply repeating a task label or a statistical term still does not supply the explanatory bridge in this vocabulary example.

The notes separate evidence and reasoning at the citation: text before a citation is treated as evidence, and text after it as reasoning unless another citation accompanies it. Make that separation clear in the response. A citation such as ‘(Source 1)’ refers here only to the teaching packet above, not a real publication. Follow the actual exam prompt’s source and citation instructions.

The rubric also permits explanation/application credit despite some evidence-point failures, including specified citation or specificity issues; C(ii) can earn those points with different evidence even if the same source was used twice. That scoring nuance does not remove C(i)’s different-source evidence task. The complete demonstration below meets the intended structure with two different sources and two distinct concepts rather than presenting a fallback scoring exception as the procedure to follow.

Use the teaching issue and Sources 1–3 shown in full above. The question asks for a policy about adult vocabulary practice; the recommendation below is bounded to similar learners and does not claim a tested synergy or universal transfer. The labeled parts demonstrate a complete EBQ-style structure. They are not a restatement of the official rubric and are not scored here.

A · Claim: For adult vocabulary learners similar to the university volunteers, the learning center should replace some passive rereading with retrieval practice distributed across several days while retaining activities that teach and assess understanding. Annotation: this makes a specific policy recommendation and identifies an outcome-relevant limit rather than merely describing the topic.

B(i) · Evidence: The randomly assigned retrieval group recalled an average of 18 of 24 word meanings one week later, compared with 14 of 24 after matched rereading practice (Source 1). Annotation: this states the accurate direction, measure, delay and conditions, with a clear source citation. It supports the activity part of the claim.

B(ii) · Explanation/application: Retrieval practice makes the learner attempt access to stored meanings before feedback instead of keeping the answer available during practice. Practicing that access helps explain the retrieval group's advantage on the later cue-only recall test, so the evidence supports replacing some rereading with recall attempts for similar vocabulary learning. Annotation: the concept explains the bridge from activity to measured outcome and recommendation; it does not merely repeat that scores differ.

C(i) · Different evidence: With equal total retrieval time, adults practicing across four days averaged 19 of 24 meanings after one week, whereas those concentrating practice on the final day averaged 15 of 24 (Source 2). Annotation: this is a specific result from a different supplied source. It supports the timing part of the claim; it does not itself compare retrieval with rereading.

C(ii) · Different explanation/application: The spacing effect concerns improved later retention when practice is distributed over time rather than concentrated. The higher delayed recall after spaced sessions, with total practice time held equal, supports spreading the center's vocabulary practice across days. Annotation: spacing explains schedule, a different psychological concept from retrieval practice in B(ii). The two applications support different elements of the claim.

Qualification beyond the required pair structure: Source 3’s definition advantage does not accompany a higher observed novel-problem mean, so the recommendation preserves teaching and assessment of understanding. The convenience recruitment and particular tasks also restrict how broadly the center should generalize. Annotation: this demonstrates broader Skill 4.B’s modification and nuance; it is supplemental to the five labeled EBQ parts, not an invented additional official point.