Science inquiry — Intermediate

Score: 0 / 20

Level 2: Intermediate

Explain, apply, analyse, sequence — connect ideas, read patterns, use rules in new contexts (bridge to Year 11).

1. A team tests a new fertiliser but also shifts light levels between control and test at the same time. The main problem is:

   A)Confounding — fertiliser and light move together, so their effects cannot be separatedB)Still a fair test whenever fertiliser was the only variable the team meant to changeC)Changing two factors at once mainly tightens precision by averaging over more conditionsD)Independent variables cannot interact, so each factor’s effect can be read off separately

2. Why might blinding (where practical) strengthen a drug trial?

   A)It allows researchers to drop participants who respond poorly without affecting statisticsB)It replaces informed consent because participants no longer know they are in a trialC)It reduces bias from participants or researchers knowing who got which treatmentD)It ensures the placebo and active drug are chemically identical in every batch

3. Plant height in mm is plotted against days since planting. Points rise along an almost straight line. The pattern is best described as:

   A)No growth — height would stay flat, not increase on a rising lineB)Roughly steady daily growth rate while the straight trend holdsC)Height increasing faster each day because any rising line means accelerationD)Growth rate undefined at the day where both height and time read zero together

4. A student concludes that one variable caused another from a correlation only. The best critique is:

   A)A strong correlation with a neat graph is usually enough to infer causation in school labsB)If two variables rise together, the one measured first must cause the change in the otherC)Controlled follow-up is unnecessary whenever the plotted points lie close to a smooth curveD)A lurking variable or reverse causation may explain the link — design and evidence matter

5. Systematic error differs from random error because it:

   A)Tends to bias results in a consistent direction, such as from a miscalibrated scaleB)Would average to zero over many repeats even when the instrument reads consistently highC)Appears only when the sample is small, then vanishes automatically as n increasesD)Appears as scatter around the line of best fit with no consistent offset direction

6. Choosing a line of best fit through scattered points should:

   A)Drop every point that sits awkwardly far from the line you first sketch by eyeB)Balance residuals fairly and follow the trend implied by the model you justifyC)Force the line through every plotted point so none of the data is left unusedD)Chase the highest $R^2$ with a high-degree curve even when points are sparse

7. Cultural protocols may be required when research involves:

   A)Geologic samples described only as inert rock with no human story attachedB)Anonymous online game metrics collected without reference to place or peopleC)Indigenous knowledge, sacred sites, or community-held data — follow consent and partnership normsD)Bench chemistry using only factory-made glassware and stock reagents from a catalogue

8. A poster for a science fair should prioritise:

   A)Humour and decorative fonts so the display feels entertaining above all elseB)Clear question, method summary, key graph, conclusion, and limits in readable layoutC)Dense walls of raw tables so every number collected is visible at onceD)Colour schemes matched to the hypothesis, with methods explained only if asked

9. A control in a fair test is useful because it:

   A)Provides a baseline to compare against when the independent variable changesB)Must match the experimental group on every factor, including the manipulated IVC)Removes the need for repeated trials because one control run is definitiveD)Guarantees the hypothesis is accepted if the control condition is included

10. Random allocation of subjects to treatment groups (where ethical) mainly helps:

    A)Remove the need for informed consent because assignment is automaticB)Guarantee treatment groups differ on every nuisance variable in the same directionC)Replace careful measurement with random guessing about who improvedD)Spread nuisance factors across groups so averages stay broadly comparable

11. Matched-pairs designs test each subject twice or pair similar subjects to:

    A)Inflate participant counts without a clear design rationaleB)Remove biological or behavioural variability entirely from the outcomeC)Avoid plotting results because paired tables are assumed sufficientD)Hold individual differences constant so the treatment contrast stands out

12. Interpolation between known data points on a smooth trend is usually more cautious than extrapolation beyond your data because:

    A)Extrapolation is banned outright in school assessment rulesB)Trends may change outside the measured range, so far extrapolation is riskyC)Interpolation fails whenever points lie on a smooth underlying curveD)Axes on graphs cannot mathematically continue beyond the last plotted point

13. A survey posted only in one fan forum about a product may suffer:

    A)Selection bias — enthusiasts may not represent all users or non-usersB)Perfect random sampling because online posts reach everyone equallyC)No bias as long as a large number of forum members respondD)Automatic proof of what caused their product opinions to form

14. A thermometer reads 0.4 °C high every time. Measurements are precise but lack:

    A)Numeric readings because the scale does not display decimalsB)Reliability, since repeats disagree strongly at the same temperatureC)Accuracy, because a consistent positive offset shifts every readingD)Units, if students forget to write the degree symbol beside values

15. A pilot trial runs a small version of a method mainly to:

    A)Serve as the final dataset so a full run is never neededB)Debug timing, equipment, and practical issues before full data collectionC)Guarantee statistical significance once the main trials beginD)Bypass ethics review because fewer participants are involved

16. Truncating the vertical axis on a bar chart without clear indication can:

    A)Prevent viewers from misreading category labels on the horizontal axisB)Automatically make the chart more honest without further annotationC)Remove entire categories from the dataset before plotting barsD)Inflate apparent differences — honest charts show scales transparently

17. Repeated measures on the same subjects can increase power for some questions but may need care because:

    A)Earlier trials can carry over through order effects or learningB)Participants always withdraw before the second measurement is takenC)Averaging across repeats is mathematically undefined for paired dataD)Control conditions cannot be included if the same person is measured twice

18. Double-blind drug trials mean both participants and key staff analysing outcomes often do not know:

    A)The scientific hypothesis the trial was designed to evaluateB)That medicines can produce adverse effects in some patientsC)Which participants received active drug versus placeboD)Whether informed consent was collected before enrolment began

19. Triangulation in school projects can mean using more than one method or data source to:

    A)Mask inconsistent findings by mixing incompatible measurement typesB)Pad the report length without changing the evidential claimsC)Postpone any interpretive conclusion until more methods disagreeD)Build confidence when separate, independent approaches point the same way

20. A report claims a chemical ‘caused’ a colour change, but temperature was not logged and the room warmed during the run. This mainly threatens:

    A)Validity — a hidden variable may explain the outcome alongside the chemicalB)Reliability — repeats would surely disagree because the room lacks windowsC)Precision — the colour change cannot be observed without digital videoD)Communication — readers cannot understand claims unless SI units appear in the title

← Back to hub