Cite specific evidence and determine the central ideas of a science text

Learning goal: Read an original science and technical text closely, cite specific textual evidence to support an analysis of it, determine its central ideas and conclusions, analyze how the text develops them, and write an accurate, objective summary that keeps the author's conditions.

Original technical text — How a Rain Garden Reduces Runoff. This text was written for this lesson. Its measurements are hypothetical and describe no actual school, storm or county guide.

1. The problem. When rain falls on a roof, a driveway or a packed path, almost none of it soaks in. The water runs off, collects in gutters and storm drains, and reaches a creek within minutes. A fast rush of runoff carries loose soil, lawn fertilizer and oil from pavement, and it arrives all at once, which scours stream banks. On bare or compacted soil the same thing happens more slowly: the surface seals, and water that cannot enter the ground has to move across it.

2. How a rain garden works. A rain garden is a shallow basin with a flat bottom and gently sloping sides, usually 10 to 20 centimeters deep, dug where runoff already flows and planted with deep-rooted plants that tolerate both wet and dry spells. It reduces runoff in three ways. First, the basin holds water: a bed whose top opening covers 10 square meters and that is 15 centimeters deep can hold about 1.5 cubic meters, or 1,500 liters, before overflowing; that figure treats the sides as vertical, so a basin whose sides slope inward from that opening toward a smaller flat bottom holds somewhat less, and the number is an upper limit for a bed of that size. Second, the water in the basin infiltrates, meaning it enters the soil through the pores between particles instead of running across the surface. Loosened soil mixed with compost has larger connected pores than compacted soil, so it infiltrates faster. Third, plant roots keep the pores open and take up some of the stored water, which the leaves later release to the air as vapor. Water that infiltrates does not disappear; it moves downward toward groundwater or sideways through the soil far more slowly than it would move across pavement, and the slow route is what protects the creek.

3. Procedure: siting and sizing the bed. (1) Find where water already flows during rain; a rain garden receives runoff, it does not create it. (2) Keep the bed at least 3 meters from any building foundation, because water soaking in beside a wall can enter a basement. (3) Before any digging, an adult obtains the site owner's permission (at a school, the office's) and has the spot checked for buried utility lines; in the United States the free call-before-you-dig service is reached by dialing 811, and no hole is dug until the lines have been marked or the spot cleared. Then test the soil: dig a hole 20 centimeters deep, fill it with water, let it drain, refill it, and time how long the second fill takes to empty. If it drains within 24 hours, the site passes the drainage test; if it takes between 24 and 48 hours, the site can still work with a larger, shallower bed; if water still stands after 48 hours, the soil is too tight and needs amending or a different site. Passing the drainage test is one site check among several, not a verdict on the whole site. (4) Estimate the area that drains to the site, the roof section and pavement that feed it, and make the bed about one fifth of that area if the soil drains well, larger if it drains slowly. (5) Dig (within the cleared area only), mix compost into the loosened soil, adding coarse sand only if a soil test and the local guide for your soil call for it, because a little sand mixed into heavy clay can make drainage worse, then plant, mulch, and direct the downspout or path runoff into the basin with a shallow channel.

4. What one class measured. A seventh-grade class dug a 6-square-meter bed 15 centimeters deep beside a school walkway, fed by one downspout that drains about 40 square meters of roof (a little smaller than the one-fifth guide in step 4, because the walkway left no more room). During a storm that delivered 12 millimeters of rain, they measured the depth of standing water in the bed every hour. The bed filled to 8 centimeters by the end of the rain and was empty 14 hours later. On a neighboring stretch of lawn, water from a similar downspout pooled and then flowed across the grass to the storm drain within 40 minutes. The class concluded that the bed had held and infiltrated water that would otherwise have reached the drain during the storm.

5. What this does not show. One storm at one site cannot show how the bed behaves in a long, heavy rain that exceeds its capacity; overflow then leaves by the channel as runoff. Nor does the measurement show what happened to the water after it entered the soil. The claim the evidence supports is narrow: this bed held this storm's runoff and drained within a day. The class planned to record several more storms before saying anything about the whole year.

What “cite specific textual evidence” requires. Evidence in a science and technical text is a sentence, phrase, number or step that the text actually contains. Specific means you can point to it: quote it, or paraphrase it closely, and say where it sits (section 2, step 3, the last paragraph). “The text says rain gardens are good” cites nothing; it names the topic. “Section 2 states that a bed whose top opening covers 10 square meters and that is 15 centimeters deep holds about 1,500 liters before overflowing” cites a specific claim with its numbers. Analysis is the claim you make about the text, and the evidence must support that claim, not merely sit near it. If your analysis is that the author treats infiltration as the most important of the three ways a rain garden works, your evidence has to show that emphasis: section 2 explains infiltration and ends on why the slow route through the soil protects the creek, the soil test in step 3 of the procedure measures drainage and nothing else, and the class's measurement in section 4 records how long the bed took to empty. Three citations from three sections support the analysis; one quotation of the word infiltration does not.

A citation must also keep the author's conditions. The text does not say the bed holds 1,500 liters; it says a bed of a stated size holds about that much before overflowing. Dropping “about”, dropping the size or the shape, or dropping “before overflowing” turns the author's careful claim into a different, stronger one, and a reviewer counts that as misquoting. In science and technical texts the conditions are part of the evidence. You can check the number yourself: 10 square meters times 0.15 meters is 1.5 cubic meters, and one cubic meter is 1,000 liters, so the figure is consistent with the stated size and shape, and the author's own note that sloping sides hold less is why “about” and “upper limit” are part of the claim.

Determining the central ideas or conclusions. A central idea is what the sections keep returning to; a conclusion is what the author is finally willing to claim after the explanation, the procedure and the measurement. To find them, ask what each section contributes and what they have in common. Section 1 establishes that runoff is a problem because water that cannot enter the ground moves fast across it. Section 2 explains three ways a bed slows that water down. Section 3 gives a procedure whose steps mostly protect those three ways; one, the setback from foundations, protects the building instead. Section 4 reports a measurement of the bed holding water and emptying. Section 5 limits the claim. The idea that runs through all five is that a rain garden reduces runoff by holding water and letting it enter the soil slowly, and the conclusion is stated in section 5 in narrow form: this bed held one storm's runoff and drained within a day. Notice that the conclusion is not “rain gardens stop flooding.” The author never says that, and the fifth section exists to stop readers from saying it for the author.

Analyzing content-area-specific text development. Science and technical texts develop an idea in a recognizable order: a problem, a mechanism, a way to test or build, a result, and the limits of the result. Analyzing development means explaining how each part builds on the one before. The mechanism in section 2 is what makes step 3 of the procedure make sense: the soil test times how fast water drains because infiltration is the mechanism the bed depends on. The measurement in section 4 matters because section 2 predicted it: a bed that infiltrates should empty, and it did, in 14 hours. Development in a technical text is therefore not just “first this, then that”; it is a chain in which the later parts test or apply the earlier parts, and an analysis of development names those links.

Writing an accurate and objective summary. A summary of a science text includes the central ideas or conclusions, keeps the author's conditions, and leaves out the reader's opinion. Accurate means every statement in the summary can be traced back to the text and means what the text meant. Objective means the summary adds no praise, blame or claims the author did not make. Here is a model summary in three sentences, written after the analysis above. Runoff from roofs and pavement reaches streams quickly and carries pollutants, so the text explains how a rain garden, a shallow planted bed placed where water already flows, reduces runoff by holding water, letting it infiltrate through loosened soil and using plant roots. A five-step procedure sites and sizes the bed, including a drainage test that a site passes if a refilled hole empties within 24 hours. In a hypothetical trial during a 12-millimeter storm, a 6-square-meter bed filled to 8 centimeters and emptied within 14 hours, which the author says supports only the narrow claim that this bed handled this storm. Every number in the summary is in the text, and nothing in the summary says whether the class should build another bed.

The misconception: a summary is a list of the interesting parts. Students often summarize by collecting striking details: the 1,500 liters, the basement, the 40-minute lawn flow. A list of details is not a summary, because it does not say what the details are for. The 40-minute figure belongs in a summary only as the comparison that shows what the bed changed, and it belongs with its condition, that it came from a similar downspout on lawn. A second misconception is that objective means “no numbers.” The opposite is true in a technical text: the numbers are its most objective part, provided they keep their units and conditions.

A fully worked model. Suppose the analysis question is: what does the author think the class's measurement proves? Step 1, find the sentences where the author says what the measurement shows. Section 4 ends, “The class concluded that the bed had held and infiltrated water that would otherwise have reached the drain during the storm.” Section 5 says, “The claim the evidence supports is narrow: this bed held this storm's runoff and drained within a day.” Step 2, check that each quotation supports the analysis rather than the topic: both sentences state what the measurement shows, so they are the evidence. Step 3, state the analysis with the citations: the author treats the measurement as support for a limited claim about one bed and one storm, because section 4 attributes the conclusion to the class rather than stating it as a general law, and section 5 explicitly calls the supported claim narrow and lists what one storm cannot show. Step 4, check the conditions: the analysis keeps “one bed”, “one storm” and “within a day.” That is a cited analysis. A sentence that said “the text proves rain gardens work” would fail steps 2 and 4.

Worked citation check: Three students support the analysis “the author wants readers to be cautious about the result.” Student A writes, “The text is about rain gardens and runoff.” That is the topic, not evidence. Student B writes, “Section 5 says one storm at one site cannot show how the bed behaves in a long, heavy rain that exceeds its capacity.” That is specific, located and relevant: it quotes a limit the author states. Student C writes, “The class measured the bed every hour.” That is specific and located (section 4) but it does not support caution; it describes the method. Only B's sentence is evidence for this analysis. B could strengthen it by adding a second citation from the same section: the author calls the supported claim “narrow.”

Worked central idea and conclusion: Ask what each section returns to. Section 1: water that cannot enter the ground runs across it. Section 2: three ways a bed lets water enter the ground. Section 3: steps that make sure water can enter the ground at the site. Section 4: water entered the ground in 14 hours. Section 5: what that one result does and does not show. The central idea is that a rain garden reduces runoff by holding water and letting it enter the soil slowly; the conclusion is the narrow claim in section 5. A student who answers “the central idea is that the class dug a garden” has picked an event from one section instead of the idea shared by all five.

Worked summary repair: A draft summary reads, “Rain gardens are the best way to stop pollution, and the class proved it works in 14 hours.” It has three faults. “Best” is the writer's opinion; the text compares a bed with a lawn, not with every other method. “Stop pollution” is not claimed; the text says runoff carries pollutants and that the bed held runoff. “Proved it works” drops every condition the author kept. Repaired: “The text explains that a rain garden reduces runoff by holding water and letting it infiltrate, and it reports a hypothetical trial in which a 6-square-meter bed held a 12-millimeter storm's runoff and drained within 14 hours, a result the author limits to that bed and that storm.” Every phrase now traces to a sentence in the text.

Evidence trail: Make a four-column table: the claim you are analyzing; the quotation or close paraphrase; the section and, for the procedure, the step number; and the condition you kept (a unit, an “about”, a “within”, a “one storm”). Fill it for three claims: that the author treats infiltration as the key mechanism, that the procedure is built to protect infiltration, and that the author limits the conclusion. Each claim needs at least two rows from different sections. Trade tables with a partner and check that every quotation can be found where the row says it is.

Independent performance: Write a one-paragraph analysis answering, “How does the author develop the idea that a rain garden reduces runoff, and how far does the author say the evidence goes?” Then write a three-sentence accurate and objective summary of the whole text. Success criteria: the analysis cites at least three specific pieces of textual evidence from at least three sections, each with its location; every cited number keeps its unit and condition; the analysis names at least two links between sections (for example, how the procedure applies the mechanism, or how the measurement tests it); the summary states the central idea and the conclusion in your own words, contains no opinion words such as best, great or useless, and adds no claim the text does not make. Check each sentence of the summary against the text before you hand it in.