TEAS Scientific reasoning study guide

What this topic actually asks of you

Scientific reasoning is one of the four sub-content areas of the ATI TEAS Science section. ATI publishes its size: it is 6% of the exam and 9 of the 44 scored science questions. The whole Science section is 50 items (44 scored, 6 unscored) and is timed separately — 60 minutes.

Two things follow, and they shape how to study it.

• This topic is about method, not memorized facts. There is almost nothing to look up here. Every item gives you a measurement, a table or a described experiment and asks you to judge it: is that variable the one being changed or the one being measured, is that reading precise or accurate, does that evidence actually support that claim.
• The words are where the points go. Independent and dependent variable, control and controlled variable, accuracy and precision, correlation and causation. These pairs sound alike and mean different things, and the items are written to separate them.

Our practice items are single-answer and text-only. The real ATI TEAS also uses other item types: ATI's own content outline lists multiple-choice, multiple-select, fill-in-the-blank numeric, hot spot and ordered-response items. Elk Ridge Academy's practice items are single-answer only, and they carry no pictures, graphs or diagrams — a data table or a described experiment is written as text — so treat the method, not any one format, as the thing you are drilling.

ATI's scientific-reasoning objectives. ATI's Content Outline names what this topic measures, and every practice item below maps to one of them: S.4.1 use basic scientific measurements and measurement tools · S.4.2 apply logic and evidence to a scientific explanation · S.4.3 predict relationships among events, objects and processes · S.4.4 apply the scientific method to interpret a scientific investigation.

The nine procedures

Every practice item in this topic is one of these nine. Each is worked out in full below — the rule, the steps, and the trap it sets. (The ATI objective each one serves is in brackets.)

1. measurement and units — the unit a quantity is measured in [S.4.1]. Length, mass, volume, time and temperature; the common metric units; and the metric prefixes that scale them.
2. choosing an instrument — matching the tool to the quantity [S.4.1]. A balance for mass, a graduated cylinder for liquid volume, a thermometer for temperature, a ruler for length, a stopwatch for time.
3. precision and accuracy — close together vs close to the truth [S.4.1]. Precision is how tightly repeated readings cluster; accuracy is how close a reading is to the true value. Repeated trials and systematic error.
4. scientific method — the steps of an investigation, in order [S.4.4]. Question, hypothesis, experiment, data, conclusion.
5. variables and controls — what was changed, what was measured, what was held fixed [S.4.4]. Independent, dependent and controlled variables, and the control group.
6. logic and evidence — does the evidence support the claim? [S.4.2]. What counts as evidence, what weakens a conclusion, and why one trial proves little.
7. correlation and causation — changing together is not causing [S.4.2]. Correlation, causation, and the confounding variable behind a coincidence.
8. predicting relationships — reading the pattern and saying what comes next [S.4.3]. A positive relationship, a negative one, no relationship; the trend words increasing / decreasing / constant; and interpolating versus extrapolating.
9. reading data — getting the numbers out of a table [S.4.3]. Which column is which variable, what the units are, the trend, and a rate worked out from two rows.

Reference tables: the facts to know cold

Recall items are drawn from these rows, one fact per row. Read the whole row — the questions ask about the row, not just the name in it.

Table A — The instruments, and what each one measures. Match the tool to the quantity, then give the unit.

Table B — The units, and the prefixes that scale them. The common metric unit names the quantity; the prefix multiplies it by a power of ten.

The prefix directions that trip people up. Going down a prefix (kg → g) makes the number bigger (× 1000). Going up a prefix (mL → L) makes the number smaller (÷ 1000).

Strictly, the SI side. The SI base unit of mass is the kilogram — not the gram — and of temperature the kelvin — not the degree Celsius. Volume has no SI base unit; the liter is a metric unit accepted for use with SI. This topic's items ask for the common metric unit, which is what the table names.

Table C — The scientific method, in order. Each step is built on the one before it.

A hypothesis, strictly. A hypothesis is a suggested explanation that can be tested; it is usually written as an if–then prediction the experiment could support or contradict.

Table D — The variable roles. A controlled experiment usually has all four.

Table E — Accuracy and precision. Two different questions about a set of readings.

Table F — The relationship shapes. How one quantity behaves as another changes.

Positive, negative, or none. Two quantities either move the same way (positive), move in opposite directions (negative), or show no pattern at all — and that last case is answered none, or no relationship. Those are the only three answers to "what kind of relationship is this?". When the question asks instead about the trend of one series, the words are increasing (the value rises as you read across the table), decreasing (it falls) or constant (it does not change from row to row).

A four-step method for reading an experiment

Almost every item in this topic hands you a described experiment, a set of readings or a table, and asks one of four questions. Run these steps in order.

1. Find what was changed. The independent variable is the thing that differs between the groups or the trials on purpose — the temperature you set, the concentration you used, the group that got the treatment. There is usually only one.
2. Find what was measured. The dependent variable is the number the experiment produced, and it almost always comes with a unit — the height in centimeters, the time in seconds, the count of colonies. It is the effect.
3. Find what was held fixed. The controlled variables are everything else that could have changed and did not. If something that could affect the result was left free, then the comparison is no longer fair, and that thing is a confounding variable.
4. Compare and judge. Only if the change is the one difference that matters, with everything else held fixed, can you say the change caused the result. If the two things merely moved together with something else left free, they are correlated, and the movement may be a coincidence or a shared cause.

The one rule that keeps this honest: a conclusion may say only what the data show. The data support "the plants given more light grew taller"; they do not support "light is the only thing that affects plant height", because that was never tested.

The nine worked examples

1. Measurement and units — convert a mass

A sample has a mass of 2.5 kg. Given that 1 kg = 1000 g, what is the mass in grams?

1. A kilogram is the larger unit, so going down to grams makes the number bigger.
2. Multiply by the factor the row gives: 2.5 × 1000.
3. 2.5 × 1000 = 2500 g.

The trap: moving the decimal the wrong way. Going kg → g multiplies by 1000; going g → kg divides by 1000. Read which direction the question asked.

2. Choosing an instrument — match the tool to the quantity

A student needs the length of a metal rod to the nearest millimeter. Which instrument should be used?

1. The quantity is a length, and the unit is millimeters.
2. A metric ruler is the instrument whose scale reads length in millimeters and centimeters.
3. So the answer is the metric ruler.

The trap: naming a balance or a graduated cylinder. A balance measures mass and a cylinder measures a liquid's volume; neither reads a length. Match the instrument to the quantity the question names.

3. Precision and accuracy — average repeated readings

Three trials of a volume give 12.4 mL, 12.6 mL and 12.5 mL. What is the average of the three readings?

1. Add the three readings: 12.4 + 12.6 + 12.5 = 37.5.
2. Divide by the number of trials: 37.5 ÷ 3.
3. 37.5 ÷ 3 = 12.5 mL.

The trap: calling the average the "true value". Averaging three trials reduces random error; it does not make the average correct, because a systematic error (an instrument that always reads high) survives the average untouched.

4. The scientific method — name the step

A student notices that the water in the sink feels colder than the water in the bath, and records: "If water is colder, it will take longer to boil." Which step of the scientific method is that recorded sentence?

1. The sentence predicts what will happen next, in an if–then form.
2. The steps run: ask a question, form a hypothesis, run the experiment, collect the data, draw a conclusion.
3. A testable if–then prediction is the hypothesis — the step that follows the question.

The trap: calling any early statement "the question". The question asks does water temperature affect boiling time?; the hypothesis answers it with a prediction that could be shown wrong. An observation comes before both.

5. Variables and controls — name the two roles

A student grows bean plants at three light levels and measures each plant's height at the end of two weeks. What is the independent variable, and what is the dependent variable?

1. Ask which quantity the student set: the light level. That is the independent variable.
2. Ask which quantity the student measured: the height. That is the dependent variable.
3. Independent variable = the light level; dependent variable = the height.

The trap: swapping them. The independent variable is the one you change; the dependent variable is the one that responds. Everything else (the water, the soil, the pot) had to be held the same — those are the controlled variables.

6. Logic and evidence — what the data support

A detergent is tested on 5 stained cloths and 4 come out clean. A student concludes that the detergent removes every stain. What is wrong with that conclusion?

1. The data are 4 of 5 — they show the detergent removed most of the stains in this test.
2. "Every stain" is a claim about all stains everywhere, and one cloth in this test was not cleaned.
3. So the conclusion claims more than the evidence supports: the data support "most of these stains came out", not "every stain".

The trap: reading a claim as if the words "always" and "every" were free. A conclusion may say only what the data show, and one counter-example is enough to break "every".

7. Correlation and causation — the shared cause

A study finds that children who own more books also read better on a test. What can be concluded?

1. The two quantities move together — that is a correlation, and it is real.
2. Nothing in the study says the books caused the reading; the comparison that would show it was never run.
3. A third factor could explain both — the homes that buy books may also read with their children — so the honest conclusion is that the two are correlated, not that one causes the other.

The trap: reading a correlation as a cause. "Both went up together" is not evidence of "this made that happen". Look for the shared cause, and for whether the suspected cause was actually tested.

8. Predicting relationships — the next value in a pattern

A solution's temperature is 20 °C at 0 minutes and rises by 4 °C every minute. What temperature is predicted at 3 minutes?

1. Both quantities rise together — a positive relationship — and the temperature rises by the same amount each minute: a steady increase.
2. The rate is the change per minute: 4 °C for each minute, and the starting value at 0 minutes is 20 °C.
3. At 3 minutes the temperature is 20 + 4 × 3 = 32 °C.

The trap: reporting the change instead of the reading. 12 °C is the rise after 3 minutes, not the temperature; add it to the starting value.

9. Reading data — a rate from two rows

A table shows the volume of liquid collected: 15 mL at 2 minutes and 45 mL at 8 minutes. What is the average rate?

1. The quantity that changed is the volume; the quantity it changed with is the time.
2. The change is 45 − 15 = 30 mL over 8 − 2 = 6 minutes.
3. The rate is the change divided by the time: 30 ÷ 6 = 5 mL per minute.

The trap: dividing by the wrong time. The change happened over the 6 minutes between the readings, not over 8 minutes from zero — subtract the two time numbers as well as the two volumes.

The traps, collected

Most wrong answers in this topic's practice items are one of these.

• Independent and dependent swapped: called the measured quantity the independent variable (or the reverse).
• Controlled variable called independent: counted something that was held fixed as the thing the experimenter changed.
• Control group misread: called the treated group the control, or said a control group is not needed for a comparison.
• Accuracy and precision swapped: called tightly clustered but wrong readings "accurate".
• A single trial treated as proof: accepted one reading as the answer to a general question.
• Instrument mismatched to quantity: named a balance for a liquid volume, or a ruler for mass.
• A conversion moved the wrong way: divided where the unit got smaller, or multiplied where it got larger.
• A conclusion stronger than the data: read "most" as "every", or "in this test" as "always".
• Correlation read as causation: concluded that one thing caused another because they moved together.
• A confound ignored: accepted a comparison in which a second thing also differed between the groups.
• A hypothesis read as a fact: treated a testable prediction as a result already proved.
• Interpolation and extrapolation swapped: called a value beyond the data an interpolation.
• A rate divided by the wrong interval: used the total time from zero instead of the change between the two readings.

Answer-choice strategies: three moves for when the setup will not come

1. Find the verb. "Changed" → independent variable. "Measured" → dependent variable. "Held the same" → controlled variable. The stem almost always says which role it is asking about before it asks for the name.

2. Check the direction of a conversion before you choose. Going down a prefix makes the number bigger (kg → g, × 1000); going up makes it smaller (mL → L, ÷ 1000). A choice on the wrong side of the starting number is gone before you finish the arithmetic.

3. Ask what the data could actually show. If a choice says "proves", "always", "every" or "causes", and the setup describes one test or two things moving together, the choice claims more than the evidence. That is almost always the wrong one.

How to practice this

• Work the free Scientific reasoning practice set with a pencil, and name the variable roles out loud each time. The setup is where these items set their traps.
• Drill the procedures you get wrong, not the ones you already know. Sorting your errors by procedure beats counting them.
• Read the explanation on every practice item, including the ones you got right. A right answer reached with the wrong reason comes back on test day.
• Re-attempt what you missed a day later, not immediately: a day later measures whether the method stuck.

Where to go next

• Next topic — Conventions of standard English
• The rest of the section — Human anatomy and physiology, Biology, and Chemistry.

This topic sits inside the Science section and carries ATI's published 6% of the exam and 9 of the section's 44 scored items — the blueprint's own split.

TEAS® is a registered trademark of Assessment Technologies Institute, LLC (ATI). Elk Ridge Academy is an independent study resource and is not affiliated with, endorsed by, or sponsored by ATI.