Normal distribution practice becomes more reliable when you translate the event before reaching for a table or calculator. The calculation z = (x − μ)/σ is only one stage. You must also know what the variable measures, which values the question includes and whether the supplied spread is a variance or a standard deviation. This Sialkot practice guide works through those decisions using one deliberately constructed model.
The numbers are invented for teaching. They do not describe a survey of Sialkot students or predict examination grades. For a family discussion about interpreting progress figures, connect the mathematics with the Sialkot O and A Level study guide. Here the focus is the student's explanation of the event, the transformation and the final probability.
State the model and read its parameters
Let X follow a normal distribution with mean 72 and variance 36. Using the convention that the second parameter is the variance, write X ~ N(72, 36). The standard deviation is therefore 6, not 36. State that convention when presenting a model because software and resources can display parameters differently. Check what a field asks for before entering a number.
Define Z = (X − 72)/6. Under the stated model, Z has the standard normal distribution with mean zero and variance one. Subtracting 72 centres the scale at the model mean; dividing by 6 expresses distances in standard deviations. A value above 72 gives a positive z-value, and a value below 72 gives a negative one.
These sign checks are useful before any probability evaluation. If your calculation gives a negative z-value for X = 84, something has gone wrong. It might be reversed subtraction or an input error. The model's meaning gives you a check that is separate from repeating the same calculator entry and hoping it produces a different result.
Translate an upper-tail event
Suppose the question asks for P(X > 84). The boundary standardises to (84 − 72)/6 = 2. Since division is by the positive number 6, the inequality direction stays the same. The event is therefore Z > 2. Sketch a normal curve in your own notebook and mark a region to the right of a boundary above the mean.
Let Φ(z) represent the cumulative probability P(Z ≤ z). Then P(Z > 2) = 1 − Φ(2). Numerically, Φ(2) is approximately 0.977249868, so the required probability is approximately 0.022750132. A rounded answer of 0.0228 may be appropriate depending on the required accuracy. Preserve useful digits until the final step.
A common error is to report 0.9772 because it appears directly in a cumulative table or calculator display. That number represents the region below the boundary, whereas the question asks for the upper tail. State the event in symbols and words beside the numerical answer. The small result also makes sense because 84 lies two standard deviations above the mean.
Translate both ends of an interval
Now find P(66 < X < 84). The lower boundary gives (66 − 72)/6 = −1, while the upper boundary gives 2. Thus the required event is −1 < Z < 2. Its probability is Φ(2) − Φ(−1), approximately 0.977249868 − 0.158655254 = 0.818594614.
Do not standardise only the upper boundary and leave the lower one in the original units. The expression 66 < Z < 2 mixes two scales and no longer describes the original event. Apply the same increasing transformation to the entire inequality. Writing one intermediate line with both transformed boundaries is often enough to make the error visible.
The probability should be larger than one half here because the interval includes the mean and extends beyond it on both sides. That observation does not determine the exact answer, but it helps reject an implausibly tiny result. Use such checks to question the calculation, then inspect the event and inputs. Do not replace the calculation with a vague statement that an answer looks reasonable.
Use symmetry with an explicit event
Consider P(X < 60). The boundary standardises to (60 − 72)/6 = −2. By symmetry of the standard normal distribution, P(Z < −2) equals P(Z > 2), approximately 0.022750132. This is the same numerical tail probability as the earlier upper-tail question because the boundaries lie equally far from the mean on opposite sides.
Symmetry does not mean every positive and negative boundary gives the same cumulative probability. Φ(−2) is the small lower tail, while Φ(2) is the large cumulative region below positive two. The equal pair is Φ(−2) and 1 − Φ(2). Name the two regions before invoking symmetry so that the relationship is attached to an event.
The broader normal distribution standardising guide can support a second review. Use a fresh boundary rather than memorising the decimal for two standard deviations. A student who can explain the sign, identify the tail and evaluate a new event has stronger evidence of understanding than one who recognises a familiar numerical result.
Work backwards from a probability
Suppose a new question asks for c such that P(X > c) = 0.10. Before calculating, predict that c lies above the mean: only the highest tenth of the model is above it. A negative z-value would conflict with that description. Convert the event to a cumulative probability, P(X ≤ c) = 0.90, before using an inverse normal function.
If a supplied table or instruction gives the rounded standard normal quantile 1.282, use (c − 72)/6 = 1.282. Then c = 72 + 6 × 1.282 = 79.692. This value follows from the rounded quantile. A calculator using a more precise inverse value gives approximately 79.6893, so do not interpret their small difference as a conceptual disagreement.
The reversal also provides a check. Substitute the boundary back into the standardising expression and confirm that it is about 1.282. Then confirm that you used an upper probability of one tenth rather than a lower probability of one tenth. These two checks inspect different parts of the reasoning: the algebraic rearrangement and the probability convention.
Distinguish a model value from a rounded observation
For a continuous normal variable, P(X = 84) = 0. This does not mean that a recording system can never display 84. If the underlying quantity is rounded to the nearest whole unit, a displayed value of 84 usually corresponds to an interval such as 83.5 ≤ X < 84.5 under a specified rounding convention. The interval has positive probability.
Keep the measurement rule explicit. The choice of included endpoints does not change a probability under a continuous normal model, but the interval width does. A question about a rounded display is therefore different from a question about one exact underlying value. Read what is measured and how it is recorded before translating the words into symbols.
Likewise, do not apply a continuity correction automatically whenever the word normal appears. A continuous quantity model and a normal approximation to a discrete count require different reasoning. Establish the variable and the modelling instruction first. If the task is directly about the continuous X defined here, calculate its stated interval without inventing a separate discrete process.
Separate standardised position from real-world importance
Consider another invented model, Y with mean 76 and standard deviation 4. The value 84 gives (84 − 76)/4 = 2, the same standardised position as X = 84 in the first model. Both are two model standard deviations above their respective means. Their raw distances from those means are different, and the underlying quantities need not have the same practical meaning.
Standardisation does not show that two measurements are equally desirable, equally difficult or equally important. It describes position relative to a specified model. A family should not use the invented models to compare actual school results. The Sialkot family guide develops the broader habit of asking what reference point supports a numerical comparison.
Keep an error record that names the decision
When reviewing a wrong answer, identify its first meaningful error. Did you use 36 as the standard deviation? Did you select the lower tail for an upper-tail event? Did you forget to transform one interval boundary? These observations lead to different follow-up questions. Writing normal distribution mistake in an error log is too broad to guide the next attempt.
Complete a fresh task after the correction with the original worked example closed. Record any assistance using the independent-attempt approach. Fit the follow-up into the wider Pakistan AS and A2 study plan, and confirm your syllabus and component route with your teacher when selecting further questions.
Frequently asked questions
Why do we divide by six when the model says N(72, 36)?
In this guide the second parameter is explicitly the variance. Its square root is the standard deviation, six. Standardising divides by the standard deviation. Always check the convention used by the question or software rather than assuming that every display uses the same parameter order and meaning.
How do I know whether to subtract a table value from one?
First identify what the table provides. If it gives Φ(z) = P(Z ≤ z), an upper-tail probability P(Z > z) is one minus that value. Translate and sketch the requested event before consulting the table. A memorised subtraction rule without an event can produce the wrong tail.
Why are strict and inclusive inequalities equivalent here?
A continuous normal variable assigns probability zero to one exact boundary value. Consequently P(X < a) and P(X ≤ a) agree under this model. This property does not automatically apply to a discrete count, where an individual supported integer value can have a positive probability.
Should every normal answer be rounded at each step?
Keep adequate precision during intermediate calculations and round the final result according to the task instructions. A supplied rounded quantile can produce a slightly different boundary from a full-precision inverse calculation. State which value you used so the small numerical difference can be understood correctly.
How can I ask for help with a normal distribution question?
Show your variable definition, event, standardised boundaries and probability expression, even if the final number is wrong. That makes the difficult decision visible. For current teaching arrangements, use Abeel Eschool admissions and explain the student's syllabus and the specific type of question needing support.
Ask about O and A Level courses
Share your subjects, current study stage and the support you need. The Abeel Eschool team can help you check the current course options.