Integrate words with a graph, separate fact from judgment and speculation, and compare sources on one topic

Learning goal: Integrate quantitative and technical information expressed in words with the same information expressed visually in a graph and a table; distinguish among facts, reasoned judgment based on research findings, and speculation in a science text; compare and contrast the information gained from an experiment, a simulation and a video with the information gained from reading a text on the same topic; and read, comprehend and respond to a science and technical text at this band independently.

Original technical text — Three Surfaces, One Afternoon: What the Infiltration Tests Showed. Written for this lesson; the readings are hypothetical.

Method. On a dry afternoon after a week without rain, the class measured infiltration at three surfaces within 30 meters of each other: the amended soil of the rain-garden bed, the school lawn, and the packed earth path. They used the can-and-timer procedure with one change: each run lasted 60 minutes, and whenever the water level fell 25 millimeters they refilled the can to the fill line and added the drops together, so every surface started at a head of 50 millimeters and was refilled by the same rule (the head fell between refills, so at any one moment the three cans held different depths). Each surface had a wetting run first. The class planned to record the total depth of water that had entered the soil every 10 minutes, but missed the 50-minute reading at all three surfaces, so the record gives totals at 10, 20, 30, 40 and 60 minutes.

Findings. In the rain-garden bed, 18 millimeters had entered by 10 minutes, 32 by 20 minutes, 44 by 30 minutes, 54 by 40 minutes and 72 by 60 minutes. In the lawn, the totals were 6, 11, 15, 19 and 26 millimeters. On the path, they were 1, 2, 3, 4 and 6 millimeters. Over the full hour the bed accepted 72 millimeters, the lawn 26 and the path 6. The bed's rate fell during the hour: 18 millimeters entered in the first 10 minutes, a rate of 108 millimeters an hour, but the same 18 millimeters took the last 20 minutes, a rate of 54 millimeters an hour, half the starting pace. The lawn's rate also fell, from 36 millimeters an hour in the first 10 minutes (6 millimeters) to 21 millimeters an hour in the last 20 (7 millimeters). The path's readings show no change in rate as far as a ruler marked in millimeters can tell: 1 millimeter in each measured 10-minute interval, and 2 millimeters over the 20 minutes from 40 to 60, the same average of 6 millimeters an hour.

Interpretation. Because the bed accepted nearly three times as much water as the lawn and about twelve times as much as the path in the same hour under the same starting head and refill rule, we judge that the bed's loosened, amended soil is the most likely reason it emptied within 14 hours after the storm reported earlier, and that the lawn would have taken several times longer to accept the same depth. Three spots on one afternoon do not test the whole bed, and the bed differs from the lawn in more than its soil, since it is a planted basin, so this judgment is tentative and would change if a test of the whole bed pointed elsewhere. The can rate is far higher than the bed's average drain rate after that storm, about 6 millimeters an hour (8 centimeters in 14 hours); a small can, with water spreading sideways under its rim, usually accepts water faster than a whole bed does, so we use the can numbers to rank the surfaces, not to predict how fast the bed empties. The falling rate in the bed and the lawn is consistent with the explanation that pores fill as the soil wets. At 10 minutes the bed had accepted three times as much as the lawn (18 against 6) and eighteen times as much as the path (18 against 1); by 60 minutes the ratios had fallen to about 2.8 (72 against 26) and 12 (72 against 6), so we judge that a run shorter than the full hour would have overstated the bed's advantage, most of all over the path. The path's low rate may be because delivery trucks have compacted it for years; nobody measured that, and it is possible that the path simply sits on a clay layer. The class also wonders whether the bed's rate would drop in winter when the soil is cold; no one tested it.

Other sources. Before the tests, the class watched a four-minute video of a rainfall simulator spraying water on a bare hillside and a grassed hillside, ran an online model in which the user chooses a soil type and a slope and reads off a runoff percentage, and read a chapter explaining pores, compaction and infiltration. The video showed runoff forming on bare soil within seconds and mud leaving the slope, but it named no soil type, said nothing about whether the hillside had been compacted, and gave no numbers for any soil the class could visit. The model produced a number for every setting but used built-in soils, and it reported that a “loam” at a 2 percent slope produced 20 percent runoff without saying how that figure was obtained. The chapter explained why compacted soil resists infiltration but described no particular place. The can tests gave numbers for the school's own three surfaces, on one dry afternoon, at three spots.

Integrating words and a visual. The findings paragraph and the graph below express the same quantitative information in two forms, and the standard asks you to integrate them: read each, check that they agree, and use each for what it does best. To integrate, locate every number in the words on the graph. The sentence “72 by 60 minutes” is the bed's last point, at the 60-minute mark; “26” and “6” at 60 minutes are the lawn's and the path's last points. Then read what the graph shows that the words state less clearly. The bed's line bends: it is steep at first and flatter later, which is the falling rate the words describe as “half the starting pace.” The path's line is low and straight, because every measured interval gives the same average, 1 millimeter per 10 minutes; the readings are consistent with an unchanging rate but cannot show what happened between them. The gap between the bed and the lawn widens across the hour, from 12 millimeters at 10 minutes to 46 at 60. The words report the end result as a ratio, “nearly three times”, and that ratio actually shrinks slightly, from 3 to about 2.8, which the widening gap on the graph does not show. Now read what the words give that the graph cannot: the method, the dry week, the refilling rule, the wetting run, and the definition of the axes. A graph without the method is a picture; the words without the graph hide the shapes. When words and visual disagree, say so and trust neither until you find the source of the difference. Notice also what neither form has: there is no 50-minute reading, so each line runs straight from 40 to 60 minutes, and that straight segment shows the average over those 20 minutes, not what happened within them.

Facts, reasoned judgment based on research findings, and speculation. A fact in a science text is a statement that was measured or directly observed, or that follows from such measurements by arithmetic alone with no further assumption, and that another person could check: “72 millimeters had entered by 60 minutes,” “the bed accepted nearly three times as much water as the lawn” (72 ÷ 26 ≈ 2.8), “the class watched a four-minute video.” Reasoned judgment based on research findings is a conclusion that goes beyond the measurements but is tied to them by stated reasoning: “because the bed accepted nearly three times as much water … we judge that the bed's loosened, amended soil is the most likely reason it emptied within 14 hours …” The signals are a reason (because), the findings it rests on (nearly three times), a hedge (we judge, most likely), and a stated limit (three spots on one afternoon). Speculation is a guess about a cause or a future result that no finding in the text supports: “the path's low rate may be because delivery trucks have compacted it for years; nobody measured that.” The signals are may, possible, wonders, and an admission that nothing was measured. Signals help but do not decide. A speculation can be stated flatly, and a judgment can be stated with confidence; the test is whether you can point to the findings and the reasoning that connect them. The sentence about winter is speculation even though it is honest, because no one tested it. The sentence about a run shorter than the full hour is judgment, because it rests on the measured ratios of bed to lawn and bed to path at 10 and at 60 minutes.

Comparing and contrasting information from experiments, simulations, video and text. Each source gave information the others could not. The experiment gave numbers for the school's own surfaces under a stated method, which neither the video nor the model could supply; its limit is that it covers three spots on one dry afternoon. The simulation let the class change one variable at a time and see a result instantly, which the experiment could not do, but its soils were built in, and its 20 percent runoff figure came without a method, so the class could not check it. The video showed the process itself, runoff forming and carrying mud, in a way that a table of depths never shows, but it gave no measurements. The chapter gave the explanation, the vocabulary of pores and compaction, and the reason the falling rate was expected, but it described no particular place. Comparing means asking, for each source, what question it answers, what its numbers rest on, and what it leaves out. Contrasting means naming where they differ: the model said loam gives 20 percent runoff; the experiment gave no runoff percentage at all, only infiltration depths, so the two figures are not the same quantity and cannot simply be compared. A careful reader does not rank the sources; the reader says what each contributed and what each could not.

A simulation whose rules you can see. The class's online model was a black box: it gave a number without showing how. Here is a simulation written for this lesson whose rules are all on the page, so you can run it by hand and see where its numbers come from. It is a model, not a measurement; its built-in capacities are illustrative values, not results from any soil. Rule 1. Each built-in surface has an infiltration capacity: sandy loam 50 mm/h, loam 20 mm/h, clay 5 mm/h, compacted ground 2 mm/h. Rule 2. Choose a rain intensity in mm/h. If the rain is slower than the capacity, runoff is 0 percent; otherwise runoff percent = (rain − capacity) ÷ rain × 100. Rule 3. On a slope steeper than 5 percent, add 10 to any runoff percent that is already above zero, but never above 100. Run it: loam at 25 mm/h of rain on a 2 percent slope gives (25 − 20) ÷ 25 × 100 = 20 percent, the same figure the online model reported for loam. Now you can see the kind of thing the online model hid: in this model the 20 percent is not a fact about any soil; it follows from an assumed capacity of 20 mm/h and an assumed rain of 25 mm/h, and it changes the moment either assumption changes. The online model's 20 percent must likewise rest on assumptions of its own, but it does not say which, so this paper model shows how such a figure can arise, not how that one did. Run clay at the same rain and it gives (25 − 5) ÷ 25 × 100 = 80 percent. Nothing in either model comes from the school's path; the only number about the path is the experiment's 6 millimeters in an hour, and even that is a depth measured with a head that started at 50 millimeters and fell between refills, not a runoff percentage in rain. Comparing a simulation with an experiment therefore means asking where each number came from: the experiment's from a can on the ground under a stated method, the simulation's from rules someone chose.

Video: watch one and make one. Your teacher will choose a short public video of a rainfall simulator, the kind a university extension service or a soil and water conservation district demonstrates at a fair: a sprayer rains on trays of different surfaces, and the runoff and the water that soaks through collect in separate jars. Watch it with a viewing log in three columns: what you saw happen, in order; every number the video stated, with its unit, or “none”; and what the video did not say (which soil, how compacted, how much rain per hour). Then make your own: film sixty seconds of one can-and-timer run at the path, with the ruler in view. Show the clip to a classmate who was not there and ask them to write down the infiltration rate. They cannot, unless the clip also shows the readings and the times; a video shows the process and hides the measurement unless the measurement is put on the screen. Compare what your classmate learned from the clip with what the data table tells them in ten seconds. That comparison, made from things you watched and measured yourself, is the comparison the standard asks for.

Reading at this band independently and proficiently. By the end of grade 8 you should be able to read a text like this one on your own and respond to it. The habits that make that possible are the ones this lesson has practiced: preview the headings to learn the kinds of section; read every number with its unit and the condition attached to it; when a sentence has a because or a may, stop and classify it; find each number from the words on the visual; and, before responding, write one sentence stating what the text claims and one stating what it says it cannot claim. Comprehension in a technical text is not a feeling of understanding; it is the ability to reproduce the numbers, the reasoning and the limits in your own words. Because this text has now been explained to you, reading it again is not a test of independent reading; the practice section supplies a second text you have not seen, and that is where independence is shown.

A second invented text for independent reading — The Bed in a Dry August (written for this lesson; all readings are hypothetical). Read this one on your own, when the practice section sends you to it. Through the second half of August no rain fell for sixteen days. The class had covered the northern half of the rain-garden bed with 8 centimeters of shredded bark mulch in June and left the southern half bare so that the two could be compared. On the sixteenth dry day they pushed a soil-moisture probe 10 centimeters into the soil at six points on each half and averaged the readings. The mulched half read 21 percent water by volume; the bare half read 9 percent. Two of the five plants on the bare half had wilted leaves at midday and recovered by evening; none on the mulched half wilted. The class judges that the mulch slowed evaporation from the soil surface, because the two halves received the same rain before the dry spell, hold the same amended soil and carry the same plants, and the only difference the class introduced was the cover. They could not say how much of the difference came from the mulch shading the soil and how much from it blocking the wind, since the mulch does both at once. One student suggested that the bare half's plants might by now have deeper roots; nobody measured roots. The class plans to repeat the probe readings after the next rain to see whether the difference closes.

The misconception. Students often treat hedged sentences as weak and confident sentences as strong. In a science text the opposite is usually true: the hedged sentence tied to findings is the author's most careful claim, and a confident sentence with no finding behind it is the least trustworthy. A second misconception is that a graph proves more than a table or a paragraph because it looks scientific. The graph in this lesson shows exactly the fifteen readings the words state, plus a starting point of zero at zero minutes for each surface, which is the condition every run began from and not a reading; its authority comes from the method in the words.

A fully worked model. Question: what is the bed's average infiltration rate over the hour, and what is it in the last 20 minutes; do the words and the graph agree; and is the sentence “the lawn would have taken several times longer to accept the same depth” a fact, a judgment or speculation? Step 1, from the words: 72 millimeters in 60 minutes is 72 mm/h. From 40 to 60 minutes, 72 minus 54 is 18 millimeters in 20 minutes, which is 18 ÷ 20 × 60 = 54 mm/h. The first 10 minutes gave 18 millimeters, which is 108 mm/h, so 54 is half of 108, matching “half the starting pace.” Step 2, on the graph: the bed's point at 60 is at 72 and at 40 is at 54, and the line is visibly flatter between them than between 0 and 10; the words and the visual agree. Step 3, classify the sentence: it is a reasoned judgment. It goes beyond the measurements (no one ran the lawn until it accepted 72 millimeters), but it rests on the finding that the lawn accepted 26 millimeters in the hour against the bed's 72, a ratio of about 2.8, and it is stated with the hedge would have. It is not speculation, because the finding and the reasoning are in the text.

Worked words-to-graph check: Take the lawn. Words: 6, 11, 15, 19, 26 millimeters at 10, 20, 30, 40, 60 minutes. On the graph, the lawn's line passes through those five points and ends at 26 at the 60-minute mark, where the readings stop. Rate over the hour: 26 mm/h. First 10 minutes: 6 ÷ 10 × 60 = 36 mm/h. Last 20 minutes: 26 − 19 = 7 millimeters in 20 minutes, 7 ÷ 20 × 60 = 21 mm/h. The words say the lawn's rate “also fell, from 36 millimeters an hour in the first 10 minutes (6 millimeters) to 21 millimeters an hour in the last 20 (7 millimeters),” and the arithmetic confirms both rates. The graph shows the same fall as a gentle bend. Nothing in either form contradicts the other.

Worked sorting of sentences: (1) “On the path, they were 1, 2, 3, 4 and 6 millimeters.” Fact: measured, checkable. (2) “We judge that a run shorter than the full hour would have overstated the bed's advantage, most of all over the path.” Reasoned judgment: rests on the measured ratios of bed to lawn and bed to path at 10 and 60 minutes, hedged with we judge. (3) “It is possible that the path simply sits on a clay layer.” Speculation: no one dug or measured, signaled by possible. (4) “… the bed accepted nearly three times as much water as the lawn …” Fact, since 72 ÷ 26 ≈ 2.8 follows directly from the measurements. (5) “The class also wonders whether the bed's rate would drop in winter when the soil is cold.” Speculation: untested, signaled by wonders. (6) “The falling rate in the bed and the lawn is consistent with the explanation that pores fill as the soil wets.” Judgment: connects the findings to the chapter's mechanism with the careful phrase consistent with, which claims agreement, not proof.

Worked source comparison: Question: does compaction reduce infiltration at our school? The video shows runoff forming on a bare hillside but does not say whether that soil was compacted and gives no number. The model gives a runoff percentage for a built-in soil, not for the path. The chapter explains why compaction should reduce infiltration but names no place. The experiment shows the path accepted 6 millimeters in an hour against the lawn's 26, which is a finding about the school, but whether the path is compacted is, in the text's own words, unmeasured. So the four sources together support a judgment that the path infiltrates far less than the lawn, and leave the cause as speculation until someone measures compaction. The comparison names what each source gave and what each left open.

Integration record: Copy the graph's fifteen readings (not the three starting points at zero) into a three-column table (surface, minutes, millimeters) from the words alone, then check every value against the graph and mark any mismatch. Compute the hour-long rate for each surface in mm/h and the rate for the last 20 minutes, showing the arithmetic. Under the table, write one thing the graph shows more clearly than the words and one thing the words give that the graph cannot.

Simulation and video record: Run the paper simulation for all four built-in surfaces at 25 mm/h of rain on a 2 percent slope and again on a 10 percent slope, showing the arithmetic, and write one sentence on what changed the answers. Keep the three-column viewing log from the rainfall-simulator video your teacher chose, and the one-paragraph account of what a classmate could and could not learn from your sixty-second clip.

Independent performance, part one (the worked text): Write a one-page response with three parts. Part one integrates words and graph: state the three hour-long totals and rates, describe the shape of each line, and say whether the two forms agree. Part two lists six sentences from the text sorted into fact, reasoned judgment based on findings, and speculation, with the signal and the finding (or missing finding) named for each. Part three compares, on one question of your choice, the information you gained from five sources you used yourself: the can-and-timer experiment from the previous lesson, the paper simulation you ran, the rainfall-simulator video your teacher chose, the clip you filmed, and this text, stating what each contributed and what each could not. Success criteria: every number carries its unit; all three rates are computed correctly (72, 26 and 6 mm/h over the hour); at least two sentences in each of the three categories are identified with a stated reason; the comparison rests on your own runs, viewing log and clip, not only on the class's account in the text, and names at least one thing each source gave that the others did not and one limit of each.

Independent performance, part two (the unseen text): Read “The Bed in a Dry August” in the Learn section on your own, without help, and write a half-page response: (a) the text's claim in one sentence and its stated limit in one; (b) three sentences classified as fact, reasoned judgment or speculation, each with its signal and its finding or missing finding; (c) one thing a graph of the twelve probe readings would show better than the words, and one thing the words give that such a graph could not. A reader checks the response against the text, not against your notes: the claim names the mulch slowing evaporation, not merely that mulch is good; the limit names shade against wind; the sentence about roots is classified as speculation because roots were not measured; every number matches the text; and the response ends with one sentence stating what the text claims and one stating what it says it cannot claim.