Practice Signal vs Noise in Math

Use these practice problems to test your method after reviewing the concept explanation and worked examples.

Worksheet PDF Free Answer-key PDF Family

Quick Recap

Signal versus noise describes the fundamental challenge of separating meaningful patterns (signal) from random, unpredictable variation (noise) in data — the central task of all statistical analysis.

Is this pattern real or just coincidence? The fundamental question of data analysis.

Showing a random 20 of 50 problems.

Example 1

medium
A survey of 400400 voters gives 52%52\% support, with a margin of error of ±5%\pm 5\%. A second survey of 400400 voters gives 48%48\% support. Should we conclude support has dropped?

Example 2

medium
You measure the same resistor 3 times: 99,101,10099, 101, 100 ohms. The labeled value is 100 ohms. Is the ±1\pm 1 variation signal or noise, and what is the best estimate of the true value?

Example 3

medium
A control-chart rule flags 'any single point more than 3σ3\sigma from the mean'. Mean =50=50, σ=4\sigma=4. A point reads 6363. Should it be flagged?

Example 4

medium
Two studies test a drug. Study A: n=20n=20, effect not significant. Study B: n=20000n=20000, the same tiny effect is 'statistically significant.' Is the effect necessarily practically important?

Example 5

easy
A teacher tracks class average scores over 6 months: {65,67,64,68,82,85}\{65, 67, 64, 68, 82, 85\}. Identify the noise (random month-to-month variation) and the signal (meaningful trend) in this data.

Example 6

challenge
A weak periodic signal of amplitude 11 sits inside Gaussian noise with σ=4\sigma=4. After averaging nn independent observations of a single time point, you want SNR ≥3\ge 3. Find the minimum nn.

Example 7

hard
To detect a signal of size SS with noise SD σ\sigma, you need SNR ≥2\ge 2 via averaging. If S=1S=1 and σ=10\sigma=10, how many independent samples are needed?

Example 8

hard
Two analysts share the same dataset. Analyst A fits a 9-parameter curve through 10 points; Analyst B fits a straight line. Which is more likely confusing noise for signal?

Example 9

easy
True or false: random noise can sometimes look like a pattern purely by chance.

Example 10

easy
A scoreboard tracks a player's accuracy each game: {70%,71%,69%,72%,70%}\{70\%, 71\%, 69\%, 72\%, 70\%\}. Is the variation here best described as signal or noise?

Example 11

easy
Fill in the blank: data == signal ++ ____.

Example 12

easy
You flip a fair coin 100100 times and get 5454 heads. About how many heads would you expect if the coin were fair?

Example 13

medium
A dataset of daily temperatures shows a clear seasonal cycle plus random daily wobble of about ±2\pm 2 degrees. A single day reads 33 degrees above the seasonal curve. Signal or noise?

Example 14

easy
In the equation reading == signal ++ noise, what does subtracting the estimated signal leave?

Example 15

medium
A radio receives a 10001000 Hz tone buried in static. Averaging 100100 recordings nearly removes the static but keeps the tone. Why does averaging help?

Example 16

easy
A thermometer reads 20.1,19.9,20.0,20.2,19.820.1, 19.9, 20.0, 20.2, 19.8 degrees for the same object. Is the underlying temperature (2020 degrees) the signal or the noise?

Example 17

medium
A website's daily visits look noisy from day to day but rise steadily over six months. Which part is the signal?

Example 18

medium
If random noise has mean 00, what does the long-run average of many noisy readings of a fixed signal SS approach?

Example 19

easy
A coin flipped 1010 times gives 77 heads. Is concluding 'this coin is biased' justified by signal here?

Example 20

challenge
Readings are signal S=10S=10 plus noise with standard deviation σ=4\sigma=4. If you average nn independent readings, the noise SD of the mean is σ/n\sigma/\sqrt{n}. How many readings are needed so the mean's noise SD is at most 11?
← Back to Signal vs Noise Worked Examples