Conduct sustained research: refine the question, search effectively, weigh each source for the task and synthesize without overreliance on one source

Learning goal: Conduct short as well as more sustained research projects to answer a question, including a self-generated question, or solve a problem; narrow or broaden the inquiry when appropriate; synthesize multiple sources on the subject, demonstrating understanding of the subject under investigation. Gather relevant information from multiple authoritative print and digital sources, using advanced searches effectively; assess the strengths and limitations of each source in terms of the task, purpose and audience; integrate information into the text selectively to maintain the flow of ideas, avoiding plagiarism and overreliance on any one source, and following a standard format for citation.

The scenario this unit uses. Castlemere Regional Library is invented, and so is every number attached to it. Its board is deciding whether to stop charging overdue fines. The simulated sources are these. Budget sheet (audited): the system has 9 branches, 62,000 active cardholders and an annual budget of 8,000,000 dollars; fines brought in 48,000 dollars last year, which is 0.6 percent of the budget. Finance office estimate: staff time spent collecting and recording fines cost about 21,000 dollars, an estimate built from average minutes per transaction multiplied by wages, not from a time study. Circulation report: 7,400 of the 62,000 cards, about 12 percent, are blocked because the holder owes 10 dollars or more, and 4,440 of those 7,400 blocked cards, 60 percent, belong to cardholders under 18. Pilot report: two branches went fine-free for 12 months; their on-time return rate was 91 percent in the year before and 89 percent during the pilot, and 310 of their 820 blocked cards, about 38 percent, were reactivated. Staff survey: 54 of 80 staff responded; 36 favor ending fines, 12 oppose and 6 are unsure. Public comments: 212 comments were submitted; 131 favor ending fines and 81 oppose. Dunmore report: another invented library went fine-free and saw circulation rise 6 percent in the same year it added Sunday hours. None of this describes a real library, and the evidence supports more than one reasonable position.

Short and sustained research. A short research project answers a focused question in a class period or two: "What share of Castlemere's budget comes from fines?" A sustained project runs for weeks, changes as the researcher learns, and ends in a substantial product: "What happens to borrowing and returns when a public library ends overdue fines, and what should Castlemere do?" Both begin with a question, and the best questions are often self-generated, arising from something the researcher noticed and could not explain.

Narrowing or broadening the inquiry. A question is too broad when the sources on it would fill a library and no single paper could synthesize them ("Are libraries good?"). It is too narrow when the answer is a single fact or when too few sources exist ("How many books did the Millbrook branch lend on one Tuesday?"). The researcher adjusts as evidence arrives. If the first search on "library fines" returns thousands of opinion pieces, narrow to "overdue fines and on-time returns in public libraries". If a narrow search on one branch returns nothing, broaden to comparable systems. Record each change and the reason; the log becomes evidence of method.

Advanced searches. An advanced search uses the tools a search engine or database provides to control what is returned. Most general search engines and many library databases support some of these: quotation marks for an exact phrase ("fine-free"), a minus sign to exclude a word, a site or domain limit (for example, restricting results to government or university domains), a date range and, in library databases, subject headings and peer-review filters. Features differ from tool to tool, so a researcher checks the tool's help page instead of assuming. The purpose is not cleverness; it is to cut the list down to results that are on the subject, of the right date and from the kinds of sites likely to hold what you need. No filter certifies authority. A domain limit selects where a page is hosted, a date range the date the tool records for it, a subject heading what it is about and a peer-review filter whether the journal is refereed (not whether every item in it was); the researcher still has to judge authorship, expertise, method and accountability.

Authoritative does not mean perfect. An authoritative source has an identifiable author or organization with relevant expertise or responsibility, a clear method and accountability for errors. Every source, even an authoritative one, has strengths and limitations, and those are judged in terms of the task, purpose and audience. The Castlemere budget sheet is authoritative on revenue and useless on returns. A national library association's survey might be authoritative on how many libraries have dropped fines but not on what happened to returns at each one. A local newspaper column is weak as evidence of effects but may be strong evidence of public opinion. For an audience of board members who will spend public money, a source's method matters more than it would in a personal reflection.

Synthesis versus summary. Summary reports what each source says, one after another. Synthesis puts sources in relation: where they agree, where they conflict, and what the combination shows that no single source shows. "The pilot report shows a two-point drop in on-time returns; the circulation report shows 7,400 blocked cards; together they frame a trade between a lower share of on-time returns and restored borrowing access for many cardholders" is synthesis. Synthesis demonstrates understanding of the subject because only a researcher who understands the evidence can relate it.

Integrating information selectively. Selective integration means using the part of a source that does work in your paragraph and no more. Three ways to integrate are quotation (exact words, in quotation marks, when the wording itself matters), paraphrase (the source's idea fully restated in your own words and sentence structure) and summary (a condensed version of a longer passage). Each is cited. A signal phrase ("According to the library's finance office, ...") keeps the flow of ideas and tells the reader where your thinking ends and the source begins.

Avoiding plagiarism. Plagiarism is presenting another's words or ideas as your own. It includes copying without quotation marks, a "patchwriting" paraphrase that swaps a few words but keeps the source's structure, and omitting a citation for a paraphrase. Common knowledge that any reader could find in many general sources does not need a citation; specific figures, findings and interpretations do.

Avoiding overreliance on any one source. A paper that draws most of its evidence from one source is really a summary of that source, and it inherits every one of that source's limitations. A quick check: count citations by source. If one source supplies more than about half of them, look for others that confirm, qualify or contradict it. The count is a prompt, not a verdict: close analysis of one primary text rightly cites it again and again, and a paper can cite five sources while resting its main claim on one. The real test is which claims depend on a single source and whether the sources are independent of one another.

A standard format for citation. Use the format your discipline or teacher specifies, such as MLA or APA, and follow it exactly. For an organization that is both author and publisher, MLA's ninth edition begins the entry with the title. For the invented pilot report, an MLA entry would read: Fine-Free Pilot Report. Castlemere Regional Library, 2026. The matching in-text citation is (Fine-Free Pilot Report 4). In APA's seventh edition the organization is the author: Castlemere Regional Library. (2026). Fine-free pilot report. APA omits the publisher when it is the same as the author, and it uses sentence case for the italicized title.

A model research log (excerpt). Question 1: "Are library fines fair?" Too broad as worded: "fair" names no criterion, so no body of evidence can bear on it yet. Two workable refinements. An evaluative one states its criterion: "How unevenly do overdue-fine blocks fall across age groups at Castlemere and at comparable libraries, and does ending fines narrow the gap?" An empirical one, Question 2 (the one pursued here): "What happens to on-time returns and active borrowers when a public library ends overdue fines?" Searches: "overdue fines" "on-time returns" public library, limited to the last ten years; subject heading Library fines. Sources gathered: the Castlemere budget sheet, circulation report and pilot report; the Dunmore report; staff survey. Evaluation: the Dunmore report shows a 6 percent rise in circulation but was published the same year the library added Sunday hours, so it cannot separate the effect of ending fines from the effect of longer hours. That limitation is recorded in the log and in the paper.

Worked narrowing and broadening: Start: "Should libraries exist in the digital age?" Too broad: no single paper can synthesize the sources, and the question is not about Castlemere. Narrowed: "What happens to on-time returns when a public library ends overdue fines?" Workable. Then too narrow: "What was the on-time rate at Millbrook in March?" The answer is a single figure. Broadened back to "at the two pilot branches over twelve months" so the answer requires comparison.

Worked source evaluation for task, purpose and audience: Task: a report to the Castlemere board recommending a policy. Source: Dunmore report. Strength: a comparison library (invented, like the rest of the scenario) that actually ended fines, with circulation data. Limitation for this task: circulation rose 6 percent in the same year Sunday hours were added, so the report cannot attribute the rise to ending fines. Use: cite it as consistent with, but not proof of, the claim that ending fines raises borrowing. Source: staff survey. Strength: staff see daily effects. Limitation: 54 of 80 responded, 36 of the 54 favor the change (two thirds of respondents, not of all staff: 36 of 80 is 45 percent).

Worked integration without plagiarism: Source sentence (invented pilot report): "Over the twelve-month period, the proportion of items returned by the due date declined from 91 percent to 89 percent at participating branches." Patchwriting, not acceptable: "Over the year, the share of items returned by the due date fell from 91 percent to 89 percent at the branches that took part." It keeps the source's structure. Acceptable paraphrase with signal phrase and citation: "The library's Fine-Free Pilot Report found that on-time returns at the two pilot branches slipped two points during the fine-free year, to 89 percent (4)." Because the signal phrase names the source, the parenthetical citation needs only the page.

Worked overreliance check: A draft cites the pilot report 7 times out of 10 citations, 70 percent. Revision replaces two pilot citations with the circulation report (blocked cards) and the budget sheet (revenue), and adds the Dunmore report with its limitation stated. The pilot report now accounts for 5 of 11 citations, about 45 percent, but the count is only the prompt. A claim-by-claim check shows that the blocked-card and revenue claims now cite their own sources, while the return-rate and reactivation claims still rest on the pilot report alone, the only source of those figures, and the paper says so.

Research record: Keep a log with five columns: date, question as currently worded, search terms and tools (including any advanced features used), sources found, and the decision made (narrow, broaden, keep). For each source you keep, write one strength and one limitation in terms of your task, purpose and audience. Finish with a citation count by source.

Independent performance: Complete a sustained research project on a self-generated question over at least two weeks, ending in a 1,200-1,600-word paper. Success criteria: a research question refined at least once, with the reason recorded; at least five authoritative print and digital sources found with documented advanced searches; a strength and a limitation of each source assessed for the task, purpose and audience; paragraphs that synthesize, relating at least two sources to each other; quotations, paraphrases and summaries integrated selectively with signal phrases; no plagiarism, including patchwriting; no central claim resting on a single source where corroboration is available, checked claim by claim (a citation count by source prompts the check; it is not the test), and any unavoidable reliance on one source acknowledged with its limitation; a works-cited or reference list in one standard format followed exactly. Also complete one short research task in a single class period and write a paragraph explaining how its method differed.