Draft — content is not yet teacher-reviewed
NCMath
Grade 7
Strands
Ratio and Proportional Relationships3 ✓The Number System3 ✓Expressions and Equations4 ✓Geometry5 ✓Statistics and Probability8 ✓
NC.7.SP.5 · NC.7.SP.6 · NC.7.SP.7Print or save as PDF
Probability Models
Will it rain tomorrow? Will the spinner land on red? These are chance events, and probability tells how likely they are. In this guide you will put a number on a chance. You will collect data, build models, and compare what you expect with what really happens.
Source: Authored for NCMath (not from the PDF prototype) — needs teacher review
What is probability?
Words to know
When you spin, roll, or draw, the result is left to chance. Here are the words you will see in this guide.
WordWhat it means
Chance eventSomething that might or might not happen, like rolling a 6.
OutcomeOne possible result, like landing on red.
EventAn outcome, or a group of outcomes, that you care about. "An even number" is an event.
ProbabilityA number from 0 to 1 that tells how likely an event is.
LikelihoodHow likely something is. Words like unlikely and certain describe it.
TrialOne time you do the chance activity, like one spin.
Relative frequencyThe number of times the event happened, divided by the number of trials.
Experimental probabilityThe relative frequency you find from real data.
Theoretical probabilityThe probability you work out from a model, without doing the experiment.
Probability modelA list of all the outcomes with the probability of each one.
Uniform modelA model where every outcome is equally likely.
01/41/23/41ImpossibleEqual chanceCertainUnlikelyLikelyLess likelyMore likely
Probability is a number from 0 to 1
Picture a number line from 0 to 1. Every chance event sits somewhere on it.
An event that can never happen has probability 0. An event that must happen has probability 1.
Everything else is in between. The closer to 1, the more likely the event is. The closer to 0, the less likely.
WordProbabilityExample
Impossible0A number cube shows a 7.
UnlikelyBetween 0 and 1/2A number cube shows a 6. The probability is 1/6.
Equally likely1/2A coin lands on heads.
LikelyBetween 1/2 and 1A number cube shows a 2 or more. The probability is 5/6.
Certain1A number cube shows a number less than 7.
Trap: A common mistake is writing a probability bigger than 1 or smaller than 0. A probability can be 0 or 1 or anything between. If you get 7/5, you put the parts in the wrong order. The top number can never be bigger than the bottom.
01green 1/5blue 1/2red 3/10
Writing a probability
Probability of an event Count the outcomes that fit the event. Divide by the number of equally likely outcomes in all. — probability = outcomes that fit ÷ all outcomes
Think of a bag with 10 marbles. Three are red, five are blue, and two are green.
You pick one without looking. Every marble is equally likely to be picked. So the probability of a color is the number of marbles of that color, out of all 10.
Example
A bag holds 10 marbles: 3 red, 5 blue, and 2 green. You pick one without looking. What is the probability of red? Which likelihood word fits?
Three of the 10 marbles are red. So the probability is 3/10, which is 0.3. That is between 0 and 1/2, so red is unlikely. Check: blue is 5/10 and green is 2/10. Then 3/10 + 5/10 + 2/10 = 10/10 = 1, because you must pick some color.
Uniform probability models
123456788 equal sections
Equally likely outcomes
Uniform model Each outcome has the same probability. Add up the outcomes that fit the event. — probability of an event = outcomes that fit ÷ all outcomes
A fair spinner has 8 equal sections numbered 1 to 8. No section is favored over another.
When every outcome is equally likely, the model is called uniform. You give each outcome the same probability.
There are 8 outcomes, so each one gets 1/8. To find the probability of an event, count the outcomes that fit it. Then divide by 8.
Example
A fair spinner has 8 equal sections numbered 1 to 8. What is the probability of landing on an even number? What is the probability of a number greater than 5?
Each number has probability 1/8. The even numbers are 2, 4, 6, and 8. That is 4 outcomes, so the probability is 4/8 = 1/2. The numbers greater than 5 are 6, 7, and 8. That is 3 outcomes, so the probability is 3/8. Check: the shaded half of the spinner is 4 of 8 sections, and both answers are between 0 and 1.
Trap: A common mistake is thinking two possible results always mean a 50-50 chance. A bus is either on time or late, but those two results are not equally likely. Use 1/2 only when the outcomes are truly equally likely.
A model as a table
A probability model lists every outcome and its probability. Here is the model for a fair number cube.
The probabilities in a model always add up to 1. The cube must show some number, so all the chances together make one whole.
Outcome123456
Probability1/61/61/61/61/61/6
Example
A fair number cube is rolled. What is the probability of a number less than 3?
Only 1 and 2 are less than 3. That is 2 of the 6 equally likely outcomes. So the probability is 2/6 = 1/3. Check: 1/6 + 1/6 = 2/6, which matches. Also, the six probabilities in the table add to 6/6 = 1.
Trap: A common mistake is a model whose probabilities do not add up to 1. Suppose a cube is given 1/6 for each of the numbers 1 to 5, and 1/3 for the 6. That adds up to 7/6, which is too much. So it cannot be a model.
Experimental probability
Collecting data
Experimental probability Divide by all the trials, not by the other results. — times the event happened ÷ number of trials
Sometimes you cannot work out a probability from equal parts. Then you do an experiment. You repeat the chance activity many times and record what happens.
Each time you do the activity is one trial. The experimental probability of an event is the number of times it happened, divided by the number of trials. This fraction is also called the relative frequency.
Example
Trina spins a spinner 50 times. It lands on blue 20 times. What is the experimental probability of blue?
She did 50 trials, and blue happened 20 times. So the experimental probability is 20/50 = 2/5 = 0.4. Check: 20 is less than 50, so the answer is between 0 and 1.
00.5150100150200Number of tossesFraction of tosses that were heads
What happens in the long run
A fair coin was tossed 200 times. The graph shows the fraction of tosses that were heads so far.
After 5 tosses, all 5 were heads, so the fraction was 1. After 10 tosses, 8 were heads, so the fraction was 0.8. At first the line jumps around a lot.
As the number of tosses grows, the line settles down. After 200 tosses, 102 were heads. That is a fraction of 0.51.
This is the long run. With many trials, the experimental probability gets close to the true probability. For a fair coin, that is 1/2.
Example
Mei tosses a fair coin 10 times and gets 8 heads. Zack tosses it 200 times and gets 102 heads. Whose fraction of heads is probably closer to the true probability?
Mei's fraction is 8/10 = 0.8. Zack's fraction is 102/200 = 0.51. The true probability of heads is 1/2 = 0.5. Zack's fraction is much closer. Check: Zack did many more trials, and more trials settle closer to the true value, just like the graph.
Trap: A common mistake is expecting a short run to match the true probability. Ten tosses can easily give 8 heads. Only a long run settles down.
Predicting with experimental probability
You can use an experimental probability to predict what will happen next. Multiply the probability by the number of new trials.
The prediction is approximate. The real count will usually be close, but not exactly the same.
Example
Trina's spinner landed on blue 20 times in 50 spins. About how many blues should she expect in 200 spins?
The experimental probability of blue is 20/50 = 0.4. In 200 spins, 0.4 × 200 = 80. She should expect about 80 blues. Check: 200 spins is 4 times as many as 50 spins, and 4 × 20 = 80.
Trap: A common mistake is thinking the prediction is exact. "About 80" means the real count might be 77 or 84. It does not have to be 80.
When outcomes are not equally likely
23red15blue12green50 draws from the mystery bag
Building a model from data
A mystery bag holds 10 marbles that are red, blue, or green. You cannot look inside. So you draw one marble, record its color, and put it back. You do this 50 times.
The bars show the counts: red 23, blue 15, and green 12. The three counts add up to 50.
Divide each count by the 50 trials. That gives an experimental probability for each color. Together they make a probability model.
ColorTimes drawnProbability
Red2323/50 = 0.46
Blue1515/50 = 0.30
Green1212/50 = 0.24
Total501
Example
Use the table to build a model for the mystery bag. Which color is most likely? About how many reds should you get in 100 more draws?
The model is red 0.46, blue 0.30, and green 0.24. Red has the biggest probability, so red is the most likely. In 100 draws, 0.46 × 100 = 46. You should get about 46 reds. Check: 0.46 + 0.30 + 0.24 = 1.00, so the model is complete.
Trap: A common mistake is giving each color 1/3 just because there are three colors. This model is not uniform. The colors have different probabilities, because the bag has more marbles of some colors.
Comparing a model to what happened
511221138475176Dashed line: 10 expected
Theory and experiment
A fair number cube is rolled 60 times. The uniform model says each number has probability 1/6.
So you expect about 60 × 1/6 = 10 of each number. The dashed lines show 10.
The real counts were 5, 12, 11, 8, 7, and 17. They are not exactly 10. Most are fairly close, but the 6 came up 17 times.
Number123456
Expected101010101010
Observed512118717
Example
A fair cube is rolled 60 times. A 6 comes up 17 times. Is that close to what the model predicts? Give two possible reasons for the difference.
The model says the probability of a 6 is 1/6. So I expect 60 × 1/6 = 10 sixes. The cube gave 17, which is 7 more. That is a big gap. One reason is chance: random results rarely match exactly, especially in only 60 rolls. Another reason is that the cube might not be fair. To find out, roll many more times. Check: 17 − 10 = 7, and 7 is a lot to be off by in 60 rolls.
Why results may not match
When the data and the model disagree, do not panic. Small differences are normal. Big differences need a reason. Here are the usual ones.
Possible reasonWhat to do
Chance. Random results never match exactly.Collect more trials and look again.
Too few trials. Short runs can wander far.Do many more trials.
The model is wrong. The object may not be fair.Build a new model from the data.
A recording mistake. A tally was miscounted.Check the data.
Example
A spinner has 4 equal sections. In 40 spins, section 1 comes up 11 times. Does the data agree well with the model?
The model gives each section probability 1/4. So I expect 40 × 1/4 = 10 spins on section 1. The spinner gave 11, only 1 more. That is very close, so the agreement is good. Check: 11 − 10 = 1, a small gap out of 40 spins.
Trap: A common mistake is thinking a result is "due". If a coin lands on heads 5 times in a row, the next toss still has probability 1/2 of heads. A coin has no memory.
Quick guide to common mistakes
If you need to...Do thisWatch out for
Tell how likely an event isFind its probability and place it from 0 to 1A probability is never below 0 or above 1.
Find a probability with equally likely outcomesCount the outcomes that fit. Divide by all the outcomes.Two possible results do not always mean 1/2.
Find an experimental probabilityDivide the times the event happened by the number of trialsPut all the trials on the bottom, not the other counts.
Predict a future countMultiply the probability by the number of new trialsA prediction is approximate. It is not exact.
Build a model from dataDivide each count by the total. Check the sum is 1.Do not assume the outcomes are equally likely.
Compare a model to dataFind the expected count. Compare it to what you saw.Small gaps are normal. Big gaps need a reason.
Use this table as a last check before you hand in your work.
Memory hook: Probability is a number from 0 to 1. A model is a guess that data can test. More trials, closer to the truth.
Practice
A bag of 8 tiles
1. A bag holds 8 tiles: 1 red, 3 blue, and 4 green. You pick one tile without looking. (a) What is the probability of picking red? Which likelihood word fits? (b) What is the probability of picking a tile that is red, blue, or green?
2. Which of these numbers could be the probability of an event? Explain. 7/10, 1.2, 0, −0.3, 1, 3/2
12345678910
3. A fair spinner has 10 equal sections numbered 1 to 10. (a) What is the probability of landing on each number? (b) What is the probability of landing on a multiple of 3?
mademissed
4. Maya shoots 30 free throws and makes 18 of them. The dots show her shots. A filled dot is a made shot. (a) What is the experimental probability that she makes a shot? (b) About how many shots should she make in 50 more tries?
5. A gumball machine has red, green, and yellow gumballs. Sam cannot see how many of each. He turns the handle 40 times. He puts each gumball back after he looks at it. The table shows what he got. (a) Build a model: find the probability of each color. (b) About how many yellow gumballs should he get in 200 turns?
ColorTimes
Red26
Green10
Yellow4
6. Jo tosses a fair coin 30 times. She gets 19 heads. (a) How many heads does the model predict? (b) Is Jo's result close to the prediction? Give a reason it may not match exactly.
7. Two classes toss a fair coin. Class A tosses it 20 times and gets 16 heads. Class B tosses it 200 times and gets 108 heads. (a) Find the fraction of heads for each class. (b) Which class's fraction is more likely to be close to the true probability? Why?
RRRBBGGY8 equal sections
8. A spinner has 8 equal sections: 3 red, 2 blue, 2 green, and 1 yellow. Ella spins it 80 times. She gets red 26, blue 22, green 21, and yellow 11. (a) Use the uniform model to find the probability of each color. (b) How many of each color does the model predict in 80 spins? (c) Compare the model to Ella's counts. Does the model fit well? (d) Ella says red is "due", because it came up less than predicted. Is she right?
a devpromptu thing