Estimating from Samples (Foundation)
Estimating from samples is the process of using results collected from a smaller sample of a population to predict or approximate a value for the whole population, such as a total, a proportion or a mean. This works because a well-chosen, representative sample should reflect the same patterns as the population it was taken from.
Before you start
Make sure you're comfortable with these topics first:
Method
- Identify the size of the sample and the size of the whole population.
- Find the proportion (or fraction) of the sample that has the characteristic you are interested in.
- Multiply that proportion by the total population size to estimate the number in the whole population with that characteristic.
- If estimating an average, use the sample mean as a reasonable estimate of the population mean.
- State any assumptions, such as the sample being random and representative, since a biased sample gives an unreliable estimate.
- Round the estimate sensibly, since it is an approximation and not an exact value.
Common mistakes
MistakeMultiplying the sample proportion by the sample size instead of the population size, so 9 out of 36 in a population of 800 is left as 9.
InsteadThe estimate is for the whole population, so multiply the proportion by the population size: 9/36 = 0.25 and 0.25 x 800 = 200. If your estimate equals the count in the sample, you have not scaled up.
MistakeStating an estimate as an exact figure, such as 'exactly 200 students walk'.
InsteadUse 'estimate' or 'about'. A different random sample would usually give a slightly different count, so a sample gives an estimate and not the true number.
MistakeGiving the sample mean when asked for the total for the whole population.
InsteadA sample mean estimates the population mean. For a total, multiply the estimated mean by the population size: a mean of 2.4 kg per household over 1800 households gives about 4320 kg.
Worked examples
Example 1
A researcher takes a random sample of 40 apples from a delivery of 600 apples and finds that 6 of the sampled apples are bruised. Estimate the number of bruised apples in the whole delivery.
- Find the proportion of the sample that was bruised: 6/40 = 0.15.
- Multiply this proportion by the total number of apples in the delivery: 0.15 x 600 = 90.
- So the final answer is an estimated 90 bruised apples in the delivery of 600.
Example 2
A village has 2400 households. A random sample of 42 households was asked whether they compost garden waste, and 14 said yes. A second random sample of 70 different households gave 21 yes. (a) Use each sample separately to estimate how many households in the village compost. (b) Combine the two samples into one estimate. (c) Explain why the combined estimate is preferred.
- First sample: 14/42 = 1/3, and 1/3 x 2400 = 800 households.
- Second sample: 21/70 = 0.3, and 0.3 x 2400 = 720 households.
- The two estimates differ because each sample contains different households. Neither is exact, and this is why an estimate should be quoted as 'about'.
- Combined sample: 14 + 21 = 35 households in a sample of 42 + 70 = 112, so 35/112 = 0.3125 and 0.3125 x 2400 = 750 households.
- Averaging the two estimates would give 760 and would give the smaller sample the same weight as the larger one, so combining the counts is fairer.
- The combined estimate is preferred because it rests on 112 households rather than 42 or 70, and a larger random sample tends to give a more reliable estimate.
- Final answer: estimates of 800 and 720 separately, and about 750 households combined.
Practice questions
Q1[2 marks]A sample of 25 students from a school of 500 found that 5 walk to school. Estimate how many of all 500 students walk to school.Show answer
Answer: 100 students (5/25 = 0.2, then 0.2 x 500).
Q2[2 marks]A sample of 80 light bulbs from a batch of 2000 found 4 were faulty. Estimate the number of faulty bulbs in the whole batch.Show answer
Answer: 100 bulbs (4/80 = 0.05, then 0.05 x 2000).
Q3[1 mark]A survey of 30 shoppers out of 900 in a store found the mean amount spent was 18.50 pounds. What is the best estimate of the mean amount spent by all 900 shoppers?Show answer
Answer: 18.50 pounds (the sample mean is used as the estimate for the population mean).
Q4[2 marks]A random sample of 50 packets of crisps from a production run of 8000 found 3 were underweight. Estimate how many packets in the whole run are underweight.Show answer
Answer: 480 packets (3/50 = 0.06, then 0.06 x 8000).
Q5[2 marks]A garden centre sampled 45 out of 3000 trees and found 40 had survived the winter. Estimate how many of the 3000 trees survived the winter.Show answer
Answer: About 2667 trees (40/45 x 3000 = 2666.7, rounded to the nearest whole tree).
Q6[3 marks]A quality inspector samples 120 items from a batch of 15000 and finds 3 defective, but the sample was taken only from items made in the last hour of the day. Explain why her estimate might not be reliable, and give the estimate assuming the sample were representative.Show answer
Answer: The sample may be biased as it excludes earlier production, so may not be representative; assuming it were, estimate = 3/120 x 15000 = 375 defective items.
Exam-style questions
Written in the style of a GCSE Statistics exam paper, with a full mark scheme.
Exam tips
- Write the sample proportion as a fraction first, count over sample size, and check it is not greater than 1 before scaling up to the population.
- If two random samples come from the same population, combine their counts and sizes into one larger sample rather than averaging the two proportions when the samples differ in size.
- After estimating, check that the answer is no more than the population and that its share of the population matches the sample's share.
Q1[2 marks]A random sample of 60 out of 1500 visitors to a theme park were asked if they had been before; 18 said yes. Estimate the number of the 1500 visitors who had been to the park before.Show mark scheme
Tick each line you got.
Nothing ticked yet - 2 available
Q2[3 marks]A farmer takes a random sample of 25 chickens from a flock of 400 and weighs them, finding a mean weight of 2.2 kg and that 4 of the 25 chickens weigh over 3 kg. Estimate (a) the mean weight of the whole flock, and (b) the number of chickens in the whole flock that weigh over 3 kg.Show mark scheme
Tick each line you got.
Nothing ticked yet - 3 available
Q3[4 marks]A council wants to estimate how many of the 12000 households in a town recycle regularly. A random sample of 150 households is surveyed and 96 say they recycle regularly. (a) Use the sample to estimate the number of households in the town that recycle regularly. (b) The council later discovers the sample was taken only from one street. Explain why this might make the estimate unreliable.Show mark scheme
Tick each line you got.
Nothing ticked yet - 4 available
See real GCSE Statistics past-paper questions, with official mark schemes
This topic in real past papers
We have tagged some boards' past papers question by question (not every board or year yet). This topic comes up in:
- AQA GCSE Statistics past papers: 3 questions across 3 papers, latest tagged June 2022
Free printable worksheet
Want more practice on paper? Download the Estimating from Samples (Foundation) worksheet pack - 29 pages of exam-style questions with a full mark scheme. Free to download, no card or account. Print it for personal and classroom use.
This topic is chapter 2 of GCSE Statistics Foundation Essentials 1, the whole course as one free printable PDF. Its worksheet is also chapter 2 of GCSE Statistics Foundation Topic Workbook 1 and chapter 2 of GCSE Statistics Foundation 10-Minute Tests 1, free printable practice books.
Other versions of this worksheet: