DoAssignment study guide

S2 · How to Read Confidence Intervals

Learn to interpret confidence intervals, confidence levels, and margins of error through clear examples and targeted practice.

Alberta Grade 11 Mathematics

Statistics

Interpreting estimates, confidence levels, and margins of error

A school survey can give an estimate of how many students prefer a later start time. The estimate comes from the students who answered, but the school wants to understand what may be true for all students. A confidence interval communicates a reasonable range of values for that larger group, along with a confidence level describing the method used to make the interval. This lesson focuses on interpreting supplied intervals and margins of error. You will not need to calculate them from survey data.

What you will learn

  • Explain what a confidence interval tells us about an unknown population value.
  • Interpret a confidence level without claiming that it is the probability the value lies in one particular interval.
  • Identify the estimate, margin of error, and endpoints in a supplied confidence interval.
  • Use the context and units to judge what a reported interval means.

1. Review: estimates and populations

A population is the whole group a question is about. For example, it could be every student at a school. A sample is the smaller group asked for information. A sample estimate is a value calculated from the sample, such as the percentage of responding students who prefer a later start.
The sample estimate may not match the population value exactly. The sample might not reflect every view in the whole group. A confidence interval gives a range of plausible values for the population value, based on the sample and the method used.
A percentage is a way to describe a part of a whole. When a report gives a percentage interval, its endpoints are percentages too. Keep the units and the group in mind as you interpret it.
  • Population means the whole group of interest; sample means the group that provided information.
  • An estimate is based on the sample and may differ from the population value.
  • Interpret an interval in the context of the question and with the correct units.

2. Interval, estimate, and margin of error

A confidence interval has a lower endpoint and an upper endpoint. The interval includes the values from the lower endpoint to the upper endpoint. For example, a reported interval from 42% to 50% is a range of percentages, not a claim that every value in the range is equally likely.
The estimate is the central value used to describe the result. The margin of error is the distance from that estimate to either endpoint when the interval is centred on the estimate. It describes the interval's half-width. In a report, the margin of error is often supplied directly.
For an estimate of 46% with a margin of error of 4 percentage points, the interval runs from 42% to 50%. The phrase percentage points matters: the interval's endpoints are percentages, and the difference between 46% and 42% is 4 percentage points. Do not confuse that difference with a 4% change.
A margin of error does not guarantee that the population value is within the reported range. It describes the range produced by the stated method and confidence level. Other issues, such as how people were selected or whether they answered accurately, can also affect a survey's usefulness.
interval\text{interval} = estimate\text{estimate} ± margin of error
  • The margin of error is the distance from the estimate to an endpoint in a centred interval.
  • The two endpoints are the estimate decreased and increased by the margin of error.
  • A margin of error is not a guarantee that the interval contains the population value.

3. What a confidence level means

The confidence level describes the long-run success rate of the method used to make intervals. Imagine repeating the same kind of sampling process many times and making an interval each time. A 95% confidence level means that, over the long run, about 95% of the intervals made by that method would contain the true population value.
The confidence level describes the method across repeated samples. It does not mean that there is a 95% chance the true population value is inside one particular interval after that interval has been reported. For the particular population value, either the reported interval contains it or it does not. We do not know which from the interval alone.
A higher confidence level asks the method to capture the true value more often over repeated samples. When other conditions are held the same, this usually leads to a wider interval and a larger margin of error. A wider interval gives a broader range of plausible values, so it is less precise. A narrower interval is more precise, but precision alone does not show whether the sample or survey method was good.
When reading a report, connect all three pieces: the estimate says what the sample suggests, the margin of error describes the interval's width, and the confidence level describes how the interval-making method performs over repeated samples.
  • A confidence level describes the long-run performance of the interval-making method.
  • Do not describe a 95% confidence level as a 95% chance that a fixed population value is in this particular interval.
  • With other conditions held the same, greater confidence generally means a wider interval.

4. A practical interpretation routine

First identify the population and the measured quantity. Then identify the estimate, confidence level, and margin of error, if they are supplied. Read the lower and upper endpoints as a range for the population value, in the stated units. Finally, explain the confidence level as a statement about repeated use of the method.
Check that the interval makes sense in context. A percentage interval should have percentage endpoints, and the lower endpoint should not be greater than the upper endpoint. If a report gives an estimate and a margin of error, the endpoints should be the same distance from the estimate for a centred interval. These checks help catch misread numbers; they do not prove that a survey is free from bias.
  • Name the group, the quantity, and the units before interpreting the numbers.
  • Separate what the interval says from what the confidence level says.
  • Use the reported values; this expectation asks you to interpret them, not calculate them from raw data.

Worked example

A school start-time survey

A school report says that 46% of surveyed students prefer a later start time. It gives a 95% confidence level and a margin of error of 4 percentage points. Interpret the interval and the confidence level.
Later-start preference estimate
Later-start preference estimateThe interval runs from 42% to 50%; the supplied estimate is 46% and the margin of error is 4 percentage points.42 %46 %50 %95% confidence levelMargin of error: ±4 %

The interval runs from 42% to 50%; the supplied estimate is 46% and the margin of error is 4 percentage points.

  1. Find the endpoints
    The margin of error is 4 percentage points on either side of the estimate. The reported estimate is 46%, so the interval extends 4 percentage points below and above it.
    46%−4=42%,46%+4=50%46\% - 4 = 42\%,\quad 46\% + 4 = 50\%
  2. Interpret the interval
    The interval is from 42% to 50%. In context, the report's method gives a range of plausible percentages of all students at this school who prefer a later start time.
    42%≤p≤50%42\% \leq p \leq 50\%
  3. Interpret the confidence level
    If the same kind of sampling and interval-making method were repeated many times, about 95% of the intervals made this way would contain the true percentage for all students at the school. This is not a statement that this one interval has a 95% chance of containing that percentage.
Answer: The reported interval is 42% to 50%. The method has a 95% long-run success rate: about 95% of intervals made by this method over repeated samples would contain the true percentage of all students who prefer a later start.
Check: The estimate, 46%, is 4 percentage points from both endpoints. The interval uses percentage units, as required for a percentage estimate.

Worked example

A community survey

A community survey estimates that 62% of residents support adding a cycling path. The report gives an interval from 57% to 67% and a 90% confidence level. Interpret the interval and state the margin of error.
Support for a cycling path
Support for a cycling pathThe reported range is 57% to 67%, centred at an estimate of 62%, with a 5-percentage-point margin of error.57 %62 %67 %90% confidence levelMargin of error: ±5 %

The reported range is 57% to 67%, centred at an estimate of 62%, with a 5-percentage-point margin of error.

  1. Read the range
    The reported endpoints are 57% and 67%. They give a range of plausible percentages of all residents who support the cycling path, according to the survey's method.
    57%≤p≤67%57\% \leq p \leq 67\%
  2. Relate the range to its estimate
    The estimate, 62%, is 5 percentage points above the lower endpoint and 5 percentage points below the upper endpoint. That distance is the margin of error. 62%−57%=67%−62%=562\% - 57\% = 67\% - 62\% = 5 percentage points
  3. State the confidence meaning
    A 90% confidence level means that about 90% of intervals made by this method over many repetitions would contain the true percentage of residents who support the path. It does not assign a 90% chance to the true percentage being in this one reported interval.
Answer: The interval is 57% to 67%, and the margin of error is 5 percentage points. The 90% confidence level describes the method's long-run success rate over repeated samples.
Check: The estimate is halfway between the endpoints, so the interval is centred on 62%. The estimate and endpoints all refer to percentages of residents.

Worked example

Comparing two reports carefully

Two reports estimate the share of residents who use a public library each month. Report A gives 48% with a margin of error of 3 percentage points at a 90% confidence level. Report B gives 48% with a margin of error of 6 percentage points at a 95% confidence level. Interpret what can be concluded from these supplied values.
Report B: monthly library use
Report B: monthly library useReport B's estimate is 48% with a 6-percentage-point margin of error at a 95% confidence level.42 %48 %54 %95% confidence levelMargin of error: ±6 %

Report B's estimate is 48% with a 6-percentage-point margin of error at a 95% confidence level.

  1. Interpret Report A
    Report A's estimate is 48%. Its margin of error gives an interval from 45% to 51%. The report describes the share of all residents who use a public library each month.
    48%±3%=[45%,51%]48\% \pm 3\% = [45\%, 51\%]
  2. Interpret Report B
    Report B's estimate is also 48%, but its margin of error is larger. Its interval runs from 42% to 54%.
    48%±6%=[42%,54%]48\% \pm 6\% = [42\%, 54\%]
  3. Compare without overclaiming
    Report B has a wider interval and a higher confidence level. The values are consistent with the idea that a method aiming for higher confidence can give a wider interval, when other conditions are the same. The reports alone do not establish that their other conditions are the same, so do not claim that confidence level is the only reason for the difference.
Answer: Report A gives an interval from 45% to 51% at a 90% confidence level. Report B gives an interval from 42% to 54% at a 95% confidence level. Both estimates are 48%, but Report B's interval is wider. The supplied information does not show whether all other conditions were the same.
Check: Each interval is centred on 48%. Report B's margin of error is larger, so its interval is wider. Both intervals describe percentages of residents.

Common mistakes and how to avoid them

Saying that a 95% confidence level means there is a 95% chance the true value is inside this particular interval.
Correction: Describe the 95% as the long-run proportion of intervals made by the method that would contain the true population value.
Treating the margin of error as the full width of a centred interval.
Correction: The margin of error is the distance from the estimate to one endpoint. The full interval width is twice that distance.
Calling a difference between percentages a percent change.
Correction: When comparing percentage endpoints such as 46% and 42%, describe the difference as 4 percentage points.
Assuming that a confidence interval guarantees the population value is included.
Correction: An interval is produced by a method with a stated long-run success rate; any one interval may or may not contain the true value.
Assuming a narrow interval proves that a survey is accurate in every way.
Correction: A narrow interval indicates greater precision in the reported range. It does not by itself show that the sample or survey method was free from problems.

Lesson summary

  • A confidence interval is a range of plausible values for a population quantity based on a sample and a method.
  • For a centred interval, the estimate is in the middle and the margin of error is the distance to either endpoint.
  • The confidence level describes how often the method would produce intervals containing the true value over repeated samples.
  • Interpret the interval in context, state its units, and avoid treating the confidence level as a probability for one particular interval.

Check your understanding

Question 1

A report gives an estimate of 35% with a margin of error of 2 percentage points. Which interval matches the report?
  1. 33% to 37%
  2. 32% to 38%
  3. 35% to 37%
  4. 2% to 35%
Show answer and explanation
33% to 37%
The margin of error extends 2 percentage points below and above 35%, giving endpoints of 33% and 37%.

Question 2

What does a 90% confidence level describe?
  1. The long-run proportion of intervals from the method that would contain the true population value
  2. A 90% chance that the true value lies in this one reported interval
  3. The percentage of people in the sample who answered the survey
  4. A guarantee that the interval contains the true population value
Show answer and explanation
The long-run proportion of intervals from the method that would contain the true population value
The confidence level describes the method's performance across repeated samples and intervals, not a probability for one particular interval.

Question 3

An estimate is 71% and its interval is 66% to 76%. What is the margin of error?
  1. 5 percentage points
  2. 10 percentage points
  3. 66 percentage points
  4. 76 percentage points
Show answer and explanation
5 percentage points
Each endpoint is 5 percentage points from 71%, so the margin of error is 5 percentage points.

Key terms

Population
The whole group a question or survey is about.
Sample
The smaller group that provides information about the population.
Estimate
A value based on sample information that is used to describe a population quantity.
Confidence interval
A range of values made by a method to describe plausible values for a population quantity.
Margin of error
For a centred interval, the distance from the estimate to either endpoint.
Confidence level
The long-run proportion of intervals made by a method that would contain the true population value when the sampling process is repeated.
Percentage point
The difference between two percentages, such as the difference between 46% and 42%.

Continue through Math 20-2

View the complete Math 20-2 Alberta Grade 11 Mathematics curriculum and lessons

Tutor support for S2 · Interpret confidence intervals, confidence levels, and margins of error

Working on S2 · Interpret confidence intervals, confidence levels, and margins of error in Math 20-2? Bring the question and your own attempt. In a live session, you can show each step of your attempt and compare it with a worked example. Ask about the first step that does not make sense, then try a similar question yourself.

Explore math tutoring · Request an eligible free 30-minute introduction. Your request is matched by subject and language; a tutor must accept before a time is confirmed.

برای مبحث S2 · بازهٔ اطمینان، سطح اطمینان و حاشیهٔ خطا را تفسیر کنید در درس Math 20-2، سؤال و تلاش خود را همراه بیاورید. در جلسه می‌توانید گام‌های تلاش خود را نشان دهید و آن را با یک نمونهٔ حل‌شده مقایسه کنید. راهنمای معلم ریاضی را ببینید یا جلسهٔ معارفهٔ رایگان ۳۰ دقیقه‌ای درخواست کنید. زمان جلسه پس از پذیرش معلم قطعی می‌شود.

About this lesson and its review

Published by DoAssignment. This reviewed lesson follows Alberta Grade 11 Mathematics (Math 20-2), expectation S2. It is a study resource, not an official curriculum publication.

Before publication, content is checked for structure, mathematical or chemical notation, calculations, course boundaries, and readability. Errors can still occur, so corrections are welcomed.

Official curriculum reference

Report a correction or ask a question