Probability

Probability

Defining probability:

• For a situation in which several different outcomes are possible, the probability for any specific outcome is
defined as a fraction or a proportion of all the possible outcomes.
• If the possible outcomes are defined as A, B, C, D, and so on then:

𝐧𝐮𝐦𝐛𝐞𝐫 𝐨𝐟 𝐨𝐮𝐭𝐜𝐨𝐦𝐞𝐬 𝐜𝐥𝐚𝐬𝐬𝐢𝐟𝐢𝐞𝐝 𝐚𝐬 𝐀

probability of A =
𝐭𝐨𝐭𝐚𝐥 𝐧𝐮𝐦𝐛𝐞𝐫 𝐨𝐟 𝐩𝐨𝐬𝐬𝐢𝐛𝐥𝐞 𝐨𝐮𝐭𝐜𝐨𝐦𝐞𝐬

• Make use of a notation system that eliminates many words.

• p : used to express probability of a specific outcome e.g. p (king) refers to probability of picking a king card.
• Generally concerned with the probability of obtaining a particular sample from a population.
• Probability can be expressed equally as either decimals or fractions.

Probability values:

• Most often expressed as decimal values.

• Contained in a limited range from 0 to 1 or 0% to 100%.

Random sampling:

• Outcomes obtained need to be obtained by a process of random sampling.

• Random sampling requires that each individual in the population has an equal chance of being selected.
• Samples obtained through this process are known as simple random samples.
• Another necessary requirement is that if more than 1 individual is being selected, the probabilities must stay
constant from one selection to the next → produces what is called independent random sampling.
• Consequences of random sampling requirements:
 Assumes that there is no bias in the selection process.
 Necessary to replace each individual back to the population before making the next selection.
 Above statement is known as sampling with replacement.

Probability and Frequency Distribution

Probability and frequency distribution:

• Probability usually involves a population of scores that can be displayed in a frequency distribution graph.
• Different portions of the graph  represent different portions of the population.
• Make use of probability notation when stating the probability.
• Whenever a population is presented in a frequency distribution graph, it becomes possible to represent prob-
abilities as proportions of the graph.

N = 10 where the scores are: 1 1 2 3 3 4 4 4 5 6

p (X > 4) = ? p (X < ) = ?
 p (X > 4) = 2/10  p (X < 5) = 8/10
 p (X > 4) = 0.2 or 20%  p (X < 5) = 0.8 or 80%

Miss JS Jensen (24688665)

Probability and the Normal Distribution

Probability and the normal distribution:

• Normal distribution:
 Symmetrical.
 Highest frequency in the middle.
 Frequencies taper off as you move toward either extreme.
• Normal shape can also be described by the proportions of area that are contained in each distribution section.
• Statisticians often identify the scores of a normal distribution by using z-scores.
• Possible to define a normal distribution in terms of its proportions i.e. normal and has the correct proportions.

Example: the population distribution of SAT scores is normal with a mean of 500 and a standard deviation of 100.
Given this information about the population and the known proportions for a normal distribution, we can determine
the probabilities associated with specific samples. What is the probability of randomly selecting an individual from
this population who has an SAT score greater than 700?

p (X > 700) = ?
1. The probability question is translated into a proportion question: out of all the possible SAT scores, which
proportion is greater than 700?
2. The set of ‘all possible SAT scores’ is simply the population distribution. The mean is µ = 500 so the score X = 700
is to the right of the mean  shade in the area to the right of 700 → represents the proportion we are trying to find.
3. Identify the exact position of X = 700 by computing a z-score … and find that an SAT score of X = 700 is exactly
two standard deviations above the mean and corresponds to a z-score of z = +2.00.
𝑋− µ 700−500
z= σ z= 100
 z = 2.00
4. The proportion we are trying to determine may now be expressed in terms of its z-score.
p (z > 2.00) = ?  p (X > 700) = p (z > 2.00)  p (X > 700) = 2.28%.

According to the proportions shown in the above graph, all normal distributions will have 2.25% of the scores in the
tail beyond z = +2.00 regardless of µ or σ  for p (X > 700) = 2.28%.

Unit Normal Table

Unit normal table:

• Complete listing of the full range of z-scores and proportions is on SUNLearn in the unit normal table.
• The graph of the normal distribution shows proportions for only a few select z-score values.
• Column A: lists z-score values corresponding to different positions in a normal distribution.
 Imagine a vertical line through the distribution → exact location of the line describes a z-score value.
 The vertical line also separates the distribution into 2 sections: the body and the tail.
 The body is the larger section and the tail is the smaller section.

Miss JS Jensen (24688665)

 • Column B: the proportion of the data in the body i.e. the larger section.
• Column C: the proportion of the data in the tail i.e. the smaller section.
• Column D: identifies the proportion of the distribution that is located between the mean and the z-score.

Keep in mind in order to make full use of the unit normal table:

• The body always corresponds to the larger part of the distribution whether it is on the right or the left.
• The tail is always the smaller section whether it is on the right or the left.
• Normal distribution is symmetrical  proportions on the right side are same as the corresponding left ones.
• For a negative z-score, the tail is on the left and the body on the right → vice versa for positive z-scores.
• Unit normal table does not provide negative z-score values → find corresponding proportion of positive z.
• z-score values are always positive even though their signs (+ and –) change from one side to the other.
• Column C  always lists the proportion in the tail whether it is the right tail or left tail.

Probabilities, Proportions and z-Scores

Probabilities, proportions and z-scores:

• Unit normal table lists relationships between z-score locations and proportions in a normal distribution.
• Probability equivalent to proportion  use a unit normal table → look up probabilities in normal distributions.
• Unit normal tables can be used for finding proportions/probabilities for specific z-score values.

Finding proportions/probabilities for specific z-score values (page 154):

Example taken from page 154 of the textbook:

What proportion of the normal distribution corresponds to z-score values greater than z = 1.00?

- Sketch the distribution and shade in the area you are trying to determine. In this case, the shaded portion is the tail
of the distribution beyond z = 1.00.
- To find this shaded area → look for z = 1.00 in column A to find the appropriate row in the unit normal table.
- Scan across the row to column C (tail) to find the proportion.
- Using the table in Appendix B, you should find that the answer is 0.1587.
- Notice that this problem could have been phrased as a probability question such as ‘for a normal distribution,
what is the probability of selecting a z-score value greater than z = 11.00?’. The answer is p (z > 1.00) = 0.1587 (or
15.87%).

Miss JS Jensen (24688665)

Finding the z-score location that corresponds to specific proportions (page 155):

Example taken from page 155 of the textbook:

For a normal distribution, what z-score separates the top 10% from the remainder of the distribution?

- To answer this question, we have sketched a normal distribution and drawn a vertical line that separates the high-
est 10% (approximately) from the rest. The problem is to locate the exact position of this line.
- For this distribution, we know that the tail contains 0.1000 (10%) and the body contains 0.9000 (90%).
- To find the z-score value, you simply locate the row in the unit normal table that has 0.1000 in column C or 0.9000
in column B. For example, you can scan down the values in column C (tail) until you find a proportion of 0.1000.
- Note that you probably will not find the exact proportion, but you can use the closest value listed in the table. For
this example, a proportion of 0.1000 is not listed in column C but you can use 0.1003, which is listed.
- Once you have found the correct proportion in the table, simply read across the row to find the corresponding z-
score value in column A. For this example, the z-score that separates the extreme 10% in the tail is z 1.28.
- At this point you must be careful because the table does not differentiate between the right hand tail and the left-
hand tail of the distribution. The final answer could be either z = 11.28, which separates 10% in the right-hand tail,
or z = 21.28, which separates 10% in the left-hand tail.
- For this problem we want the right-hand tail (the highest 10%), so the z-score value is z = 11.28.

• General rule is to sketch a distribution.

• Locate the mean with a vertical line and shade in the portion you are trying to find.
• Will help determine which columns to use in the unit normal table.
• Can help avoid making careless mistakes when using the table.

Probabilities and Proportions for a Normal Distribution

Probabilities and proportions for a normal distribution:

• A problem may ask for a specific probability or proportion of a normal distribution.

• Before the answer can be found in the unit normal table → need to transform X values into z-scores.
•  to solve this problem an additional step needs to be added to the process.
• To answer the question about a score (X value) from a normal distribution:
1. Transform the X values into z-scores.
2. Use the unit normal table to look up the proportions corresponding to the z-score values.

Example taken from page 157 of the textbook:

It is known that IQ scores form a normal dsitrbution with µ = 100 and σ = 15. Given this information, what is the pro-
bability (can also ask for the proportion) of random selecting an individual with an IQ score of less than 120?
𝑋− µ 120−15
z = σ z = 100
 z = 20/15  z = 1.33

- an IQ score of X = 120 corresponds to z-score z = 1.33  IQ scores < 120 corresponds to z-scores > 1.33.
- find the z-score in the unit normal table and find the answer in column B.
- find that 1.33 corresponds to a proportion of p = 0.9082.

p (X < 120) = p (z < 1.33) = 0.9082 (or 90.82%).

Miss JS Jensen (24688665)

Finding Proportions or Probabilities Located Between 2 Scores

Finding proportions or probabilities located between 2 scores:

• Process of the probability of selecting a score that is located between 2 specific values.
• Often easiest to solve using the information in column D of the unit normal table.

Example taken from page 158 of the textbook:

The highway department conducted a study measuring driving speeds on a local section of interstate highway.
They found an average speed of 58 miles per hour with a standard deviation of σ =10. The distribution was approx-
imately normal. Given this information, what proportion of the cars are traveling between 55 and 65 miles per hour?
Using probability notation, we can express the problem as:

p (55 < X < 65) = ?

- determine the z-score corresponding to the X value at each end of the interval.

𝑋− µ 55−58
For X = 55: z = z =  z = - 3 / 10  z = - 0.30
σ 10

𝑋− µ 65−58
For X = 65: z = z =  z = 7 / 10  z = 0.70
σ 10

- the first area is the proportion between the mean and z = -0.30 and the second z = +0.70.
- find that these proportions are 0.1179 and 0.2580 respectively.
- the total proportion is found by adding these two sections to get 0.3759.

p (55 < X < 65) = p (-0.30 < z < +0.70) = 0.1179 + 0.2580 = 0.3759 (or 37.59%).

Finding Scores Corresponding to Specific Proportions or Probabilities

Finding scores corresponding to specific proportions or probabilities:

• Possible to reverse the 2 step process discussed above  move in a counter-clockwise direction.
• Find the score (X value) corresponding to a specific proportion in the distribution.
• Begin with a specific proportion, use the unit normal table  find corresponding z-score then transform it into
an X value.

Example taken from page 160 of the textbook:

The distribution of commuting times for American workers is normal with a mean of µ = 24.3 minutes and a stan-
dard deviation of σ =10 minutes. For this example, we will find the range of values that defines the middle 90% of
the distribution.

- the 90% shown in the middle (0.9000) can be split in half with 0.4500 on each side of the mean.
- 0.4500 in the unit normal table → find that that exact proportion is not listed but will find 0.4495 and 0.4505.
- find a z-score of z = 1.65  z-score at right boundary is z = +1.65 and left boundary is z = -1.65.

Miss JS Jensen (24688665)

 for the distribution of commuting times, 1 standard deviation is:

1.65 σ = 1.65(10)  16.5 points.

- the score at the right boundary is  X = 24.3 + 16.5 = 40.8 and X = 24.3 - 16.5 = 7.8 at the left boundary.
- the middle 90% of the distribution corresponds to values between 7.8 and 40.8.
 90% of American commuters spend between 7.8 and 40.8 minutes commuting to work each day.
 10% of commuters spend either more or less time commuting.

Miss JS Jensen (24688665)