PSLE-SCI-REALITY-0456
Wait, What? A GPS Screen Says “HDOP = 0.8” — but 0.8 of what?
Imagine two Primary 6 students standing in the same park with two different GNSS receivers. Receiver A shows a position, eight satellites and HDOP = 0.8. Receiver B shows a position, eleven satellites and HDOP = 1.4. One student says, “Receiver A must be accurate to 0.8 metres because the number is smaller.” Another says, “Receiver B must be more accurate because it sees more satellites.” Both statements sound sensible. Neither follows from the display alone.
HDOP stands for Horizontal Dilution of Precision. It is a dimensionless number connected to the geometry of the satellites or other ranging sources used in a position solution. It helps describe how that geometry can magnify measurement uncertainty in the horizontal position. It is not, by itself, a direct statement that the position error is 0.8 metres, 1.4 metres or any other number of metres.
This is exactly the kind of real-world evidence transfer that PSLE Science can train. A display gives a number. The number looks precise. The learner’s job is not to worship the number or dismiss it. The job is to identify what quantity the number actually represents, which variables it leaves out, what independent evidence would test the claim, and how far a conclusion is allowed to travel.
Quick Answer
No. HDOP = 0.8 does not mean “the GPS error is 0.8 m”. HDOP is a geometry factor. In simplified GNSS error models, a dilution-of-precision factor combines with a scale of ranging uncertainty to estimate how geometry contributes to position uncertainty. Real user accuracy also depends on other factors such as signal blockage, reflections from buildings or other surfaces, atmospheric effects, receiver design and processing, antenna behaviour, reference information and the quality of the observations themselves.
A lower HDOP generally indicates more favourable horizontal geometry than a higher HDOP for otherwise comparable conditions. But “lower” is not the same as “this many metres of error”. The unit check is your first clue: metres measure distance; HDOP has no distance unit.
The Exact Learner Job This Reality Lab Owns
Owned learner job: evaluating a GPS/GNSS status display, specification or screenshot that reports HDOP, by separating a satellite-geometry dilution factor from a direct horizontal error in metres, then checking the additional evidence required before making an accuracy claim.
Not owned here: the complete physics of GPS signals, orbital mechanics, full GNSS surveying mathematics, generic measurement uncertainty, generic graph reading, variables, fair testing, sampling, accuracy-versus-precision, or every cause of positioning error. Those broader skills and concepts belong to existing Science owners. This page applies them to one unmistakable communication object: a displayed HDOP number.
This page also does not replace the earlier Reality Lab job about a consumer screen that reports an estimated “GPS accuracy” radius. That is a neighbouring but different object. An app’s estimated accuracy radius and an HDOP field are not interchangeable labels.
Why This Belongs in PSLE Science Reasoning
The current 2026 PSLE Science assessment is based on the 2023 Primary Science syllabus. SEAB states that candidates are assessed on Knowledge with Understanding and on Application of Knowledge and Scientific Inquiry, including interpreting and analysing information, evaluating observations, information and methods, and communicating explanations and reasoning. The MOE syllabus also encourages healthy scepticism, attention to assumptions and uncertainty, more than one plausible explanation, evidence-based model building, and understanding how Science is presented in different forms and media.
An HDOP display is therefore useful not because Primary pupils need to become satellite surveyors, but because it is a compact lesson in scientific reading: a number only becomes evidence when you know what it measures, how it was obtained, what assumptions connect it to the claim, and what it does not include.
The Real-World Communication Object
Consider this original composite GNSS status card:
| Field | Displayed value |
|---|---|
| Latitude | 1.35210° N |
| Longitude | 103.81980° E |
| Satellites used | 9 |
| HDOP | 0.8 |
| Fix age | 1 s |
| Status | 3D fix |
The card contains several different kinds of information. The coordinates are a position estimate. The satellite count says something about how many observations contributed. “3D fix” describes a solution status. “Fix age” says something about timing. HDOP says something about geometry. None of these fields, by itself, is a universal certificate that the displayed coordinates equal the true position.
The dangerous shortcut is to look at the smallest or most technical-looking number and turn it into an accuracy guarantee. Scientific reasoning asks a slower question: What does this field mean before I use it?
First Evidence Check: Look at the Units
Suppose a claim says, “HDOP is 0.8, so the receiver is wrong by only 0.8 m.” The claim has quietly changed a dimensionless geometry number into a distance. That is a warning sign.
- Position error can be expressed in metres.
- Horizontal dilution of precision is a factor with no metre unit.
- Satellite count is a count.
- Signal-to-noise measures have their own definitions and units or scales.
- Map coordinates use a coordinate reference system.
A useful Primary Science habit is to ask, “Can these two quantities even have the same unit?” If not, an unstated conversion or model must sit between them. That hidden step deserves inspection.
What “Dilution” Means Here
A GNSS receiver estimates position from distance-related observations to several satellites. The geometry of those satellites matters. If the useful directions to the satellites are spread widely across the sky, the geometry can constrain position better. If the directions are poorly arranged—for example, crowded into a similar part of the sky—the same amount of ranging error can produce a larger position uncertainty.
HDOP is one way of summarising that horizontal geometric effect. NOAA guidance describes DOP as a unitless value related to satellite geometry and explains that error components in the observations are multiplied by the DOP factor in simplified error propagation. The U.S. GPS programme likewise emphasises that user accuracy depends on a combination of satellite geometry and other error sources.
For a Primary learner, keep the model simple: HDOP tells you how favourable or unfavourable the geometry is for turning imperfect distance information into a horizontal position. It does not tell you how large all the original imperfections were.
A Paper Model: Why Geometry Changes the Result
Imagine trying to locate a hidden point using two broad strips drawn on paper. Each strip represents a measurement that narrows the point to a region rather than a perfect line.
If the strips cross at nearly a right angle, the overlap region is compact. If the strips cross at a shallow angle, the overlap region becomes long and stretched. The width of each strip has not changed; only the geometry has. Yet the uncertainty in where the point could be has changed dramatically.
GNSS uses more observations and more sophisticated mathematics, but the evidence lesson survives: measurement quality and measurement geometry are different ingredients. HDOP focuses on the geometry ingredient.
The Simplified Relationship — and Its Limit
A simplified teaching model can be written as:
horizontal position uncertainty scale ≈ HDOP × ranging-error scale
This is not a universal consumer-device formula that lets you multiply any HDOP by a guessed number and announce the true error. It is a reasoning aid. It shows why HDOP cannot become metres by itself: another quantity carrying distance information must be involved.
Suppose two situations both have HDOP = 0.8. If one situation has clean signals and small ranging errors while the other has strong multipath reflections and larger ranging errors, the same HDOP can accompany very different position quality. Geometry stayed similar; the observation errors did not.
Observed, Displayed, Claimed, Inferred
| Layer | Example | What it can support |
|---|---|---|
| Observed by receiver | Timing/range-related signal observations from available satellites | Inputs to the positioning solution, subject to measurement limits |
| Derived | HDOP = 0.8 | A summary of horizontal solution geometry |
| Displayed | Coordinates on screen | The receiver’s estimated position at that time |
| Claimed | “The position is accurate to 0.8 m” | Requires evidence beyond HDOP alone |
| Overreach | “Because HDOP is below 1, the coordinate must be correct” | Not supported |
Case 1: Same HDOP, Different Environment
Two fictional students record GNSS data using the same receiver model.
| Condition | HDOP | Environment | Independent check against known point |
|---|---|---|---|
| A | 0.9 | Open school field | 1.6 m horizontal difference |
| B | 0.9 | Beside tall glass-and-concrete buildings | 7.4 m horizontal difference |
The HDOP values are the same. The observed position differences are not. A plausible explanation is that the building environment changed signal blockage and reflections even though the satellite geometry factor displayed at those moments happened to be similar.
The lesson is not that HDOP is useless. The lesson is that HDOP is one variable in a larger evidence system. If the claim concerns actual position accuracy, direct comparison against an appropriate reference carries evidence that HDOP alone cannot provide.
Case 2: Lower HDOP, Worse Actual Error
Now consider two different moments:
| Moment | HDOP | Signal environment | Difference from known reference |
|---|---|---|---|
| 1 | 0.7 | Strong reflections near a wall | 5.8 m |
| 2 | 1.2 | Open sky | 2.1 m |
If you ranked the two fixes using HDOP only, Moment 1 would look better. Yet the independent reference shows Moment 2 was closer to the known point. This does not prove that higher HDOP is better. It proves that a favourable geometry factor cannot compensate for every other source of error.
When comparing real scientific evidence, change one sentence in your head from “HDOP predicts the exact error” to “HDOP describes an important geometry contribution to the uncertainty”. That small wording change prevents a large reasoning mistake.
Case 3: More Satellites Does Not Automatically Mean Better Geometry
A common status screen shows both the number of satellites and the HDOP. Learners may assume these two numbers are interchangeable. They are not.
Imagine one receiver uses six satellites well spread across the visible sky. Another uses nine satellites, but many are clustered in similar directions. More observations can help, but their geometry also matters. Satellite count alone does not tell you the shape of the solution geometry, just as the number of rulers in a box does not tell you whether you placed them in useful directions for a measurement.
A strong claim therefore avoids “more satellites = definitely more accurate”. Better wording is: “More usable observations can improve redundancy and geometry, but the actual positioning result also depends on where the satellites are, signal quality, the receiver and other conditions.”
Case 4: A Social-Media Screenshot
An original social-style post says:
“My phone says HDOP 0.6, so this pin is scientifically proven to be within 60 centimetres of my actual location.”
How should a Primary 6 learner respond? Do not call the person dishonest. Do not say GPS never works. Ask what evidence connects the HDOP field to the centimetre claim.
- Is HDOP being treated as a distance when it is a dimensionless factor?
- What does the device actually mean by the displayed field?
- Was the receiver in open sky or near obstructions?
- Was the screen showing a fresh fix?
- Was there an independent reference point?
- Could the map layer itself be offset from the true feature?
- Are we looking at one fortunate reading or repeated performance?
The proper conclusion is not “the pin is wrong”. It is “the screenshot does not establish a 0.6 m error bound from HDOP alone”. That is evidence-based scepticism rather than automatic disbelief.
Representation Check: A Status Screen Mixes Different Kinds of Numbers
Scientific interfaces often place many fields next to each other. Physical proximity on a screen can make unrelated quantities feel comparable. A learner might see:
- HDOP: 0.8
- Altitude: 35 m
- Speed: 0.4 m/s
- Satellites: 10
- Accuracy: 4 m
Only some fields are distances. Some are rates. Some are counts. Some are derived quality indicators. A clean screen layout does not turn them into the same measurement family. Read the label, definition and units of each field before comparing magnitudes.
Comparison Check: Was Everything Except Geometry Comparable?
Suppose an advertisement compares two receivers:
| Receiver | HDOP shown | Claim |
|---|---|---|
| A | 0.8 | “More accurate” |
| B | 1.5 | “Less accurate” |
Before accepting the comparison, ask whether both were observed at the same place and time, with similar sky visibility, antenna placement, constellations, correction services, device settings, receiver algorithms and reference standard. If not, the comparison has mixed the geometry factor with many other variables.
Even if both values are genuine, the phrase “Receiver A is more accurate” is broader than the evidence “Receiver A displayed lower HDOP at this moment”. The first statement is about performance; the second is about one geometry indicator.
Method Check: What Was Actually Tested?
When a product page, field report or class investigation uses HDOP as evidence, reconstruct the method.
- Where? Open field, under trees, beside buildings, indoors?
- When? One instant or repeated times?
- Which receiver? Same hardware and antenna or different systems?
- Which satellite constellations? Same set or different availability?
- Which processing? Standalone, augmented, differential or other?
- What reference? Was actual error checked against a known position?
- How many repeats? One fix or many independent observations?
- What outcome? HDOP itself, coordinate repeatability, or distance from reference?
If the method only recorded HDOP, it can support a claim about observed geometry conditions. If it also compared positions against a trustworthy reference over repeated conditions, it can support a much stronger statement about actual positioning performance.
Alternative Explanations: Why Might a Position Be Wrong Even With Low HDOP?
A good scientific explanation does not stop at the first plausible cause. A low HDOP says the horizontal geometry was favourable according to the model. If the position is still far from a reference, other explanations remain.
- Signal blockage: some useful paths may be weakened or lost.
- Multipath: signals can reflect from structures before reaching the receiver.
- Atmospheric effects: signal travel can be affected on the path from satellite to receiver.
- Receiver and antenna limitations: hardware and processing differ among systems.
- Reference error: the point used as “truth” may itself be poorly known.
- Coordinate-system mismatch: two datasets may use different reference frames or projections.
- Map-layer offset: the background map feature may not perfectly match the coordinate system.
- Timing: the displayed status may not describe exactly the same fix being judged.
The aim is not to memorise a long list. The transferable habit is: when one indicator looks good but the result looks bad, search for variables outside that indicator’s scope.
What Evidence Would Strengthen the Claim “This Position Is Accurate”?
- Repeated positions are compared with a well-established reference point.
- The reference coordinate system and the receiver output are matched correctly.
- The environment and sky visibility are documented.
- HDOP, signal quality and fix status are recorded for the same observations.
- Measurements are repeated at different times to avoid depending on one satellite geometry.
- Independent observations or equipment give similar results.
- The claimed accuracy range is supported by the actual distribution of errors rather than by one lucky reading.
- Limitations and poor-quality observations are reported rather than silently removed.
What Would Weaken the Claim?
- The only evidence is a single HDOP value.
- The author converts HDOP directly to metres without explaining the model.
- The reference point is not independently known.
- Different devices are compared at different places or times.
- Urban reflections, tree cover or indoor use are ignored.
- The display is old but described as current.
- Only the best reading is shown while other observations are omitted.
- Satellite count is treated as a substitute for geometry and signal quality.
A Useful Distinction: Precision, Accuracy and Geometry Are Not the Same Job
A receiver can produce repeated positions close to one another yet still be offset from the true reference. That is a repeatability or precision problem versus an accuracy problem. HDOP is different again: it describes geometric sensitivity in the position solution.
These concepts interact, but they are not synonyms. A scientifically careful learner avoids using one word as a replacement for another merely because all three appear in discussions of measurement quality.
Worked Case 5: The Playground Survey
A class marks a known point on a school field. They record five GNSS observations with the same receiver. The data below are constructed for learning:
| Observation | HDOP | Horizontal difference from reference |
|---|---|---|
| 1 | 0.7 | 1.8 m |
| 2 | 0.8 | 2.2 m |
| 3 | 0.8 | 1.5 m |
| 4 | 1.0 | 2.0 m |
| 5 | 1.1 | 1.7 m |
Can the students conclude, “Error equals HDOP”? No. The units differ and the numerical pairs do not match. Can they conclude, “Higher HDOP always caused larger error”? No. Observation 5 has higher HDOP than Observation 2 but a smaller observed difference. Can they say, “All observations occurred under fairly favourable geometry”? That may be a reasonable description if the receiver’s documentation supports interpreting these HDOP values as favourable and the class states the context carefully.
The strongest direct evidence about actual error in this table is the independent comparison with the known reference, not the HDOP field alone.
Worked Case 6: A Comparison That Looks Scientific but Is Not Fair
A fictional product comparison says:
- Receiver X: HDOP 0.9, tested on a rooftop.
- Receiver Y: HDOP 1.6, tested beside a covered walkway.
- Conclusion: “Receiver X is almost twice as accurate.”
There are two separate reasoning errors. First, 0.9 versus 1.6 is not a direct ratio of metre accuracy. Second, the receivers were tested under different signal environments. A fairer design would expose both receivers to comparable conditions, use a known reference, repeat observations and compare actual errors as well as quality indicators.
This is a useful example of why scientific-looking numbers do not rescue a poor comparison design. The method still matters.
Worked Case 7: The App Shows Both “Accuracy 4 m” and “HDOP 0.8”
A learner asks, “Which one is the real accuracy?” The correct response begins by refusing to collapse two different fields into one.
“Accuracy 4 m” may be the application or operating system’s estimate of positional uncertainty according to its own model and available sensor information. HDOP 0.8 is a geometry factor. One may contribute to the other in some systems, but they are not definitions of the same quantity. To understand the app’s 4 m field, read that device or software documentation rather than reverse-engineering it from HDOP alone.
A strong learner asks, “What exactly does each field mean on this system?” rather than “Which number looks more precise?”
Worked Case 8: Good Geometry, Bad Reference
A student measures a point with HDOP 0.7 and compares it with a pin copied from an online map. The GNSS position differs by 6 m. The student concludes the receiver has 6 m of error.
Not necessarily. The map pin may not be a surveyed reference point. It may be manually placed, based on imagery with its own registration error, or intended only as a visual label. To measure receiver accuracy, the comparison reference must itself be fit for that purpose.
This creates an important evidence chain: a measurement can only be judged against a reference whose own quality is understood.
Tempting Reasoning That Fails
| Tempting claim | Why it fails | Better question |
|---|---|---|
| “HDOP 0.8 means 0.8 m error.” | HDOP is dimensionless, not a distance. | What error scale and model connect geometry to position uncertainty? |
| “Lower HDOP means the coordinate is definitely closer to truth.” | Other error sources can dominate. | What does an independent reference show? |
| “More satellites means better accuracy.” | Count does not fully describe geometry or signal quality. | How are the satellites arranged and how usable are the signals? |
| “HDOP below 1 means perfect GPS.” | No finite geometry factor removes all other measurement errors. | What uncertainties remain? |
| “The app shows many decimals, so the location is exact.” | Display precision is not measurement accuracy. | What uncertainty accompanies the coordinate? |
| “The map pin and GPS disagree, so GPS is wrong.” | The map layer or reference can also be uncertain. | Which source has a traceable or surveyed reference? |
Model Limit: HDOP Describes Geometry, Not the Whole World
Every scientific model keeps some features and leaves others out. HDOP is useful because it compresses complex geometry into a manageable indicator. That usefulness is also its limitation. A geometry indicator cannot, by itself, encode every property of the signal path, antenna, receiver, atmosphere, correction source, map or reference point.
This is not a flaw that makes HDOP “fake”. Scientific models are valuable precisely because they simplify. The correct question is whether the simplified quantity is being used for the job it was designed to do.
Measurement Limit: Exact-Looking Coordinates Are Still Estimates
A status screen may show latitude and longitude with many decimal places. A Primary learner can easily mistake decimal detail for certainty. But numerical display resolution and real-world accuracy are different ideas.
Writing more digits does not create better observations. A coordinate can be printed to a tiny numerical step while the true position uncertainty remains several metres. Likewise, HDOP can be displayed as 0.82 rather than 0.8 without proving that the system knows its geometry contribution to hundredth-level practical accuracy.
How Far Can the Conclusion Travel?
From a well-defined HDOP value, you can make a limited statement about the horizontal geometry of the observations used in that position solution. You can compare geometry indicators cautiously when the systems and conditions are sufficiently comparable.
You cannot travel directly from HDOP to the exact true horizontal position error without additional assumptions and evidence. You also cannot use one low-HDOP observation to prove that a receiver model is always accurate, that a map pin is correct, or that another receiver with higher HDOP is universally worse.
A scientifically responsible conclusion keeps the claim at the same scale as the evidence.
PSLE-Style Transfer Case
A fictional investigation compares GNSS positions at the same marked point. Students obtain the following constructed data:
| Trial | HDOP | Distance of measured position from reference |
|---|---|---|
| A | 0.6 | 4.5 m |
| B | 0.9 | 1.8 m |
| C | 1.3 | 2.2 m |
Question: A student says, “Trial A is the most accurate because it has the smallest HDOP.” Evaluate the statement.
Reasoned answer: The statement is not supported by these data. Trial A has the smallest HDOP, so its satellite geometry indicator is the most favourable of the three. However, its measured position is 4.5 m from the reference, which is farther than Trials B and C. HDOP does not include every source of positioning error. The independent reference comparison provides direct evidence that Trial A was not the closest position in this set.
Notice the answer does not say HDOP “does not matter”. It says what HDOP supports and what the observed error supports. That separation is the core habit.
Practice 1: Units First
A receiver reports HDOP = 1.1. A learner writes, “The error is 1.1 m.” What is the first scientific objection?
Answer: HDOP is a dimensionless geometry factor, so the displayed value cannot be interpreted directly as a distance in metres without additional information and a model connecting it to measurement error.
Practice 2: Same Number, Different Conditions
Two readings both have HDOP = 0.8. One is taken in an open field and the other beside a tall reflective building. Must the two positions have the same accuracy?
Answer: No. Similar geometry indicators do not guarantee similar signal errors, blockage, multipath or receiver behaviour. Check the actual measurements and an appropriate reference.
Practice 3: Satellite Count
Receiver A uses 12 satellites; Receiver B uses 8. Can you conclude A has better horizontal geometry?
Answer: Not from count alone. Geometry depends on the directions of the usable satellites as well as their number. More satellites can help, but the spatial arrangement still matters.
Practice 4: One Lucky Fix
A receiver with HDOP 1.5 lands exactly on a known point in one trial. Does that prove HDOP is useless?
Answer: No. A single result can be unusually good by chance or because other error sources happened to cancel. Evaluate repeated observations and the role of geometry across many trials.
Practice 5: A Product Claim
An advertisement says, “Our receiver maintained HDOP under 1.0 for 95% of the test, proving sub-metre accuracy.” What missing evidence would you request?
Answer: Ask how actual positions were compared with a trustworthy reference, what the observed error distribution was, where and when testing occurred, what signal environment existed, which corrections and settings were used, and whether the claimed sub-metre result held across repeated independent conditions. HDOP alone does not prove the metre-level claim.
Practice 6: Map Pin Versus Surveyed Point
Why is a surveyed reference point stronger evidence for evaluating GNSS accuracy than a hand-placed map pin?
Answer: The surveyed point has a documented measurement basis and uncertainty suited to position comparison, while a hand-placed map pin may only be an approximate visual label.
Practice 7: Better Geometry, Worse Fix
Trial X has HDOP 0.7 and is 6 m from the reference. Trial Y has HDOP 1.2 and is 2 m from the reference. What is the best conclusion?
Answer: Trial X had the more favourable geometry indicator, but Trial Y produced the position closer to the reference in these trials. Other error sources affected the final results.
Practice 8: Display Precision
A screen changes from HDOP 0.83 to 0.84. Can you conclude the actual position became exactly 1 cm worse?
Answer: No. HDOP is not a distance and the tiny display change does not map directly to centimetres of actual position error.
Practice 9: Missing Time Match
A screenshot shows coordinates from 10:00 and an HDOP field updated at 10:05. Why is that a problem?
Answer: The quality indicator may not describe the same position solution being judged. Evidence must be matched to the correct observation.
Practice 10: Healthy Scepticism Without Cynicism
A classmate says, “HDOP is only a model, so we should ignore it.” How would you improve the statement?
Answer: HDOP is useful evidence about solution geometry. The correct limitation is that geometry is only one contributor to position uncertainty. Use the indicator for its proper scope and combine it with other evidence when evaluating actual accuracy.
Delayed Independent Return: A Different Measurement System
Weeks later, a learner encounters a scientific system in which several sensors observe the same event from different directions. A “geometry quality” field is shown beside the final estimate. The learner has never seen this exact instrument before.
The Reality Lab habit transfers: determine whether the quality field describes the geometry of observations, the raw sensor error, or the final result. Ask whether the indicator has units. Identify the other error sources. Look for an independent reference. Do not turn one quality metric into a universal guarantee.
That is the deeper purpose of this page. The topic is HDOP; the transferable skill is respecting the boundary of a scientific indicator.
Routes to Existing Canonical PSLE Science Owners
For broader evidence and representation work, use the Primary 5 Data, Graphs & Evidence Application Lab. For experimental comparison and method design, use the Primary 5 Experimental Design & Evaluation Application Lab. For uncertainty, anomalies and claim strength, use the Primary 6 Confidence, Uncertainty, Anomalies & Strength of Evidence guide.
A neighbouring Reality Lab page, PSLE Science Reality Lab Vol No.206, deals with a different communication object: a displayed GPS accuracy radius and whether it is a guaranteed error bound. This page keeps ownership only of the HDOP-versus-metres distinction.
Parent and Tutor Teaching Guide
You can teach the core idea without a satellite receiver. Draw two wide strips crossing on paper. First make them cross almost at right angles. Shade the overlap. Then draw the same-width strips crossing at a shallow angle and shade the new overlap. Ask the learner which overlap gives a more tightly located point.
Next tell the learner that the strip width represents measurement uncertainty and the crossing angle represents geometry. Keep the strip width unchanged while changing only the angle. The child should notice that geometry changes the final uncertainty even though the individual measurement quality did not change.
Then reverse the lesson. Keep the geometry fixed but make the strips wider. Now the final overlap grows because the measurement uncertainty grew. This creates a two-factor mental model: geometry can amplify error, but geometry is not the original error itself.
Finally present three cards:
- HDOP = 0.8
- Estimated accuracy = 4 m
- Difference from surveyed point = 2.3 m
Ask what each number means and which is direct evidence of observed error in that trial. The answer should be that HDOP describes geometry, the app accuracy field is an estimate defined by that system, and the comparison with a suitable reference directly measures the observed position difference for that trial.
Do not turn the lesson into memorising “low HDOP good, high HDOP bad”. That slogan is too thin. Ask the child to complete this sentence instead: “HDOP tells me about geometry; to judge actual position accuracy I still need…” A good learner should finish with ideas such as signal conditions, receiver quality, repeated observations and an independent reference.
A Compact Evidence Checklist for Students
- Name the field. Is it HDOP, an accuracy estimate, satellite count or something else?
- Check the unit. Does the claim preserve the quantity’s units?
- Ask what was derived. Was the number measured directly or calculated from other observations?
- Find the omitted variables. What important sources of error are outside this indicator?
- Match the time. Does the quality field belong to the same observation?
- Check the comparison. Were devices or conditions compared fairly?
- Look for a reference. What evidence tells us the actual position difference?
- Limit the conclusion. State only what the evidence supports.
Authoritative Sources
- Singapore Ministry of Education — Primary Science Teaching and Learning Syllabus (2023). The syllabus emphasises scientific inquiry, healthy scepticism, assumptions, uncertainty, evidence-based model building and evaluating Science communicated in different forms and media.
- Singapore Examinations and Assessment Board — 2026 PSLE Science syllabus. The assessment objectives include interpreting and analysing information, evaluating observations, information and methods, and communicating explanations and reasoning.
- GPS.gov — GPS Accuracy. The U.S. GPS programme explains that user accuracy depends on more than signal-in-space performance, including satellite geometry, signal blockage, atmospheric conditions and receiver design/features.
- NOAA National Geodetic Survey — User Guidelines for Single Base Real Time GNSS Positioning. The guidance describes DOP as a unitless value reflecting satellite geometry and explains its relationship to measurement error propagation.
- NOAA National Geodetic Survey — OPUS. The service documentation discusses GNSS solution quality, independent repeat observations, geometry, environmental effects and the difficulty of estimating accuracy for a particular solution.
Quiet Return
HDOP is useful precisely because it does one job well: it describes part of the geometry of a position solution. Trouble begins when a useful indicator is asked to do every job at once.
When you see HDOP = 0.8, do not translate it automatically into 0.8 metres. Translate it into a better scientific question: “The geometry looks favourable. What does the rest of the evidence say about the actual position?”
