Lesson 11: ChiSquare Tests
Lesson 11: ChiSquare TestsObjectives
 Construct a chisquare probability distribution plot in StatKey or Minitab Express
 Determine when a chisquare goodnessoffit test and chisquare test of independence should be conducted
 Compute and interpret expected counts
 Conduct chisquare tests by hand and using Minitab Express
 Calculate and interpret relative risk
In this lesson we will learn how to compare the proportions of more than two independent groups and how to test for a relationship between two categorical variables. In this lesson you will also review risk and odds and learn about relative risk.
Before we begin, let's review the various classifications of variables from Lesson 1:
 Categorical
 Names or labels (i.e., categories) with no logical order or with a logical order but inconsistent differences between groups, also known as qualitative.
 Quantitative
 Numerical values with magnitudes that can be placed in a meaningful order with consistent intervals, also known as numerical.
In this lesson we will be examining methods for analyzing categorical variables.
11.1  Reviews
11.1  ReviewsIn this lesson you may need to use Minitab Express to construct frequency tables or twoway contingency tables. We'll start by reviewing these procedures. You will need to construct probability distribution plots for chisquare distributions. In earlier lessons we constructed probability distribution plots for z, t, and F distributions; the procedure is similar for a chisquare distribution. We will also review conditional probabilities and the term independence in this section.
11.1.1  Frequency Table
11.1.1  Frequency TableThe following example was first presented in Lesson 2.1.1.2.1.
It uses following data set (from College Board):
MinitabExpress – Frequency Tables
To create a frequency table in Minitab Express:
 Open the data set
 On a PC: In the menu bar select STATISTICS > Describe > Tally
 On a Mac: In the menu bar select Statistics > Summary Statistics > Tally
 Double click the variable Region in the box on the left to insert the variable into the Variable box
 Under Statistics, check Counts and Percents
 Click OK
This should result in the following frequency table:
Region  Count  Percent 

ENC  5  9.8039% 
ESC  4  7.8431% 
MA  3  5.8824% 
MTN  8  15.6863% 
NE  6  11.7647% 
PAC  5  9.8039% 
SA  9  17.6471% 
WNC  7  13.7255% 
WSC  4  7.8431% 
N=  51 
Select your operating system below to see a stepbystep guide for this example.
11.1.2  TwoWay Contingency Table
11.1.2  TwoWay Contingency TableRecall from Lesson 2.1.2 that a twoway contingency table is a display of counts for two categorical variables in which the rows represented one variable and the columns represent a second variable. The starting point for analyzing the relationship between two categorical variables is to create a twoway contingency table. When one variable is obviously the explanatory variable, the convention is to use the explanatory variable to define the rows and the response variable to define the columns; this is not a hard and fast rule though.
MinitabExpress – Constructing a TwoWay Contingency Table
 Open the data set:
 On a PC: Select STATISTICS > Cross Tabulation and Chisquare
On a Mac: Select Statistics > Tables > Cross Tabulation and ChiSquare  Select Raw data (categorical variable) from the drop down menu
 Double click the variable Smoke Cigarettes in the box on the left to insert the variable into the Rows box
 Double click the variable Biological Sex in the box on the left to insert the variable into the Columns box
 Click OK
This should result in the twoway table below:
Female  Male  All  

No  120  89  209 
Yes  7  10  17 
All  127  99  226 
Cell Contents: Count 
Select your operating system below to see a stepbystep guide for this example.
11.1.3  Probability Distribution Plots
11.1.3  Probability Distribution PlotsIn previous lessons you have constructed probabilities distribution plots for normal distributions, binomial distributions, and \(t\) distributions. This week you will use the same procedure to construct a probability distribution plot for the chisquare distribution.
MinitabExpress – Constructing a Probability Distribution Plot
Chisquare tests of independence are always righttailed tests. Let's find the area of a chisquare distribution with 1 degree of freedom to the right of \(\chi^2 = 1.75\). In other words, we're looking up the \(p\) value associated with a chisquare test statistic of 1.75.
 On a PC: from the menu select STATISTICS > Distribution Plot
On a Mac: from the menu select Statistics > Probability Distributions > Distribution Plot  Select Display Probability
 For Distribution select ChiSquare
 For Degrees of freedom enter 1
 Select A specified X value
 Select Right tail
 For X value enter 1.75
This should result in the following output:
Select your operating system below to see a stepbystep guide for this example.
Example: Area to the Right of ChiSq = 6.25, df=3
11.1.4  Conditional Probabilities and Independence
11.1.4  Conditional Probabilities and IndependenceIn Lesson 2 you were introduced to conditional probabilities and independent events. These definitions are reviewed below along with some examples.
Recall that if events A and B are independent then \(P(A) = P(A \mid B)\). In other words, whether or not event B occurs does not change the probability of event A occurring.
 Conditional Probability

The probability of one event occurring given that it is known that a second event has occurred. This is communicated using the symbol \(\mid\) which is read as "given."
For example, \(P(A\mid B)\) is read as "Probability of A given B."
 Independent Events
 Unrelated events. The outcome of one event does not impact the outcome of the other event.
Example: Queens & Hearts
If a card is randomly drawn from a standard 52card deck, the probability of the card being a queen is independent from the probability of the card being a heart. If I tell you that a randomly selected card is a queen, that does not change the likelihood of it being a heart, diamond, club, or spade.
Using a conditional probability to prove this:
\(P(Queen) = \dfrac{4}{52}=0.077\)
\(P(Queen \mid Heart) = \dfrac {1}{13} = 0.077\)
Example: Gender and Pass Rate
Data concerning two categorical variables can be displayed in a contingency table.
Pass  Did Not Pass  Total  
Men  6  9  15 
Women  10  15  25 
Total  16  24  40 
If gender and passing are independent, then the probability of passing will not change if a case's gender is known. This could be written as \(P(Pass) = P(Pass \mid Man)\).
\(P(Pass) = \dfrac{16}{40} = 0.4\)
\(P(Pass \mid Man) = \dfrac{6}{15}=0.4\)
In this sample, gender and passing are independent.
11.2  Goodness of Fit Test
11.2  Goodness of Fit TestA chisquare goodnessoffit test can be conducted when there is one categorical variable with more than two levels. If there are exactly two categories, then a one proportion z test may be conducted. The levels of that categorical variable must be mutually exclusive. In other words, each case must fit into one and only one category.
We can test that the proportions are all equal to one another or we can test any specific set of proportions.
If the expected counts, which we'll learn how to compute shortly, are all at least five, then the chisquare distribution may be used to approximate the sampling distribution. If any expected count is less than five, then a randomization test should be conducted.
 When randomly selecting a card from a deck with replacement, are we equally likely to select a heart, diamond, spade, and club?
 According to one research study, in the United States 2% of adults identify as homosexual, 2% as bisexual, and 96% as heterosexual. Are these proportions different in the population of Penn State students?
 A concessions stand sells blue, red, purple, and green freezer pops. They survey a sample of children and ask which of the four colors is their favorite. They want to know if the colors differ in popularity.
In conducting a goodnessoffit test, we compare observed counts to expected counts. Observed counts are the number of cases in the sample in each group. Expected counts are computed given that the null hypothesis is true; this is the number of cases we would expect to see in each cell if the null hypothesis were true.
 Expected Cell Value
 \(E=n (p_i)\)

\(n\) is the total sample size
\(p_i\) is the hypothesized proportion of the "ith" group
The observed and expected values are then used to compute the chisquare (\(\chi^2\)) test statistic.
 ChiSquare (\(\chi^2\)) Test Statistic

\(\chi^2=\sum \dfrac{(ObservedExpected)^2}{Expected}\)
StatKey has the ability to conduct a randomization test for a goodnessoffit test. There is an example of this in Section 7.1 of the Lock^{5} textbook. If all expected values are at least five, then the sampling distribution can be approximated using a chisquare distribution.
Like the t distribution, the chisquare distribution varies depending on the degrees of freedom. Degrees of freedom for a chisquare goodnessoffit test are equal to the number of groups minus 1. The distribution plot below compares the chisquare distributions with 2, 4, and 6 degrees of freedom.
To find the pvalue we find the area under the chisquare distribution to the right of our test statistic. A chisquare test is always righttailed.
11.2.1  Five Step Hypothesis Testing Procedure
11.2.1  Five Step Hypothesis Testing ProcedureThe examples on the following pages use the five step hypothesis testing procedure outlined below. This is the same procedure that we used to conduct a hypothesis test for a single mean, single proportion, difference in two means, and difference in two proportions.
When conducting a chisquare goodnessoffit test, it makes the most sense to write the hypotheses first. The hypotheses will depend on the research question. The null hypothesis will always contain the equalities and the alternative hypothesis will be that at least one population proportion is not as specified in the null.
In order to use the chisquare distribution to approximate the sampling distribution, all expected counts must be at least five.
 Expected Count
 \(E=np_i\)

Where \(n\) is the total sample size and \(p_i\) is the hypothesized population proportion in the "ith" group.
To check this assumption, compute all expected counts and confirm that each is at least five.
In Step 1 you already computed the expected counts. Use this formula to compute the chisquare test statistic:
 ChiSquare Test Statistic
 \(\chi^2=\sum \dfrac{(OE)^2}{E}\)
 Where \(O\) is the observed count for each cell and \(E\) is the expected count for each cell.
Construct a chisquare distribution with degrees of freedom equal to the number of groups minus one. The pvalue is the area under that distribution to the right of the test statistic that was computed in Step 2. You can find this area by constructing a probability distribution plot in Minitab Express.
Unless otherwise stated, use the standard 0.05 alpha level.
\(p \leq \alpha\) reject the null hypothesis.
\(p > \alpha\) fail to reject the null hypothesis.
Go back to the original research question and address it directly. If you rejected the null hypothesis, then there is evidence that at least one of the population proportions is not as stated in the null hypothesis. If you failed to reject the null hypothesis, then there is not evidence that any of the population proportions are different from what is stated in the null hypothesis.
11.2.1.1  Video: Cupcakes (Equal Proportions)
11.2.1.1  Video: Cupcakes (Equal Proportions)11.2.1.2 Cards (Equal Proportions)
11.2.1.2 Cards (Equal Proportions)Research question
When randomly selecting a card from a deck with replacement, are we equally likely to select a heart, diamond, spade, and club?
I randomly selected a card from a standard deck 40 times with replacement. I pulled 13 hearts, 8 diamonds, 8 spades, and 11 clubs.
Let's use the fivestep hypothesis testing procedure:
\(H_0: p_h=p_d=p_s=p_c=0.25\)
\(H_a:\) at least one \(p_i\) is not as specified in the null
We can use the null hypothesis to check the assumption that all expected counts are at least 5.
\(Expected\;count=n (p_i)\)
All \(p_i\) are 0.25. \(40(0.25)=10\), thus this assumption is met and we can approximate the sampling distribution using the chisquare distribution.
\(\chi^2=\sum \dfrac{(ObservedExpected)^2}{Expected} \)
All expected values are 10. Our observed values were 13, 8, 8, and 11.
\(\chi^2=\dfrac{(1310)^2}{10}+\dfrac{(810)^2}{10}+\dfrac{(810)^2}{10}+\dfrac{(1110)^2}{10}\)
\(\chi^2=\dfrac{9}{10}+\dfrac{4}{10}+\dfrac{4}{10}+\dfrac{1}{10}\)
\(\chi^2=1.8\)
Our sampling distribution will be a chisquare distribution.
\(df=k1=41=3\)
We can find the pvalue by constructing a chisquare distribution with 3 degrees of freedom to find the area to the right of \(\chi^2=1.8\)
The pvalue is 0.614935
\(p>0.05\) therefore we fail to reject the null hypothesis.
There is not evidence that the proportion of hearts, diamonds, spades, and clubs that are randomly drawn from this deck are different.
11.2.1.3  Roulette Wheel (Different Proportions)
11.2.1.3  Roulette Wheel (Different Proportions)Research Question
An American roulette wheel contains 38 slots: 18 red, 18 black, and 2 green. A casino has purchased a new wheel and they want to know if there is any evidence that the wheel is unfair. They spin the wheel 100 times and it lands on red 44 times, black 49 times, and green 7 times.
If the wheel is fair then \(p_{red}=\dfrac{18}{38}\), \(p_{black}=\dfrac{18}{38}\), and \(p_{green}=\dfrac{2}{38}\).
All of these proportions combined equal 1.
\(H_0: p_{red}=\dfrac{18}{38},\;p_{black}=\dfrac{18}{38}\;and\;p_{green}=\dfrac{2}{38}\)
\(H_a: At\;least\;one\;p_i\;is \;not\;as\;specified\;in\;the\;null\)
In order to conduct a chisquare goodness of fit test all expected values must be at least 5.
For both red and black: \(Expected \;count=100(\dfrac{18}{38})=47.368\)
For green: \(Expected\;count=100(\dfrac{2}{38})=5.263\)
All expected counts are at least 5 so we can conduct a chisquare goodness of fit test.
\(\chi^2=\sum \dfrac{(ObservedExpected)^2}{Expected} \)
In the first step we computed the expected values for red and black to be 47.368 and for green to be 5.263.
\(\chi^2= \dfrac{(4447.368)^2}{47.368}+\dfrac{(4947.368)^2}{47.368}+\dfrac{(75.263)^2}{5.263} \)
\(\chi^2=0.239+0.056+0.573=0.868\)
Our sampling distribution will be a chisquare distribution.
\(df=k1=31=2\)
We can find the pvalue by constructing a chisquare distribution with 2 degrees of freedom to find the area to the right of \(\chi^2=0.868\)
The pvalue is 0.647912
\(p>0.05\) therefore we should fail to reject the null hypothesis.
There is not evidence that this roulette wheel is unfair.
11.2.2  Minitab Express: GoodnessofFit Test
11.2.2  Minitab Express: GoodnessofFit TestResearch Question:
When randomly selecting a card from a deck with replacement, are we equally likely to select a heart, diamond, spade, and club?
I randomly selected a card from a standard deck 40 times with replacement. I pulled 13 hearts, 8 diamonds, 8 spades, and 11 clubs.
MinitabExpress – Conducting a ChiSquare GoodnessofFit Test
Summarized Data, Equal Proportions
To perform a chisquare goodnessoffit test in Minitab Express using summarized data we first need to enter the data into the worksheet. Below you can see that we have one column with the names of each group and one column with the observed counts for each group.
Suit  Count  

1  Hearts  13 
2  Diamonds  8 
3  Spades  8 
4  Clubs  11 
 On a PC: Select STATISTICS > ChiSquare GoodnessofFit
On a Mac: Select Statistics > Tables > ChiSquare GoodnessofFit  From the dropdown box select Summarized data in a column
 Doubleclick Count to enter it into the Observed Counts box
 Doubleclick Suit to enter it into the Category names box
 Click OK
This should result in the following output:
Category  Observed  Test Proportion 
Expected  Contribution to ChiSq 

Hearts  13  0.250000  10  0.90 
Diamonds  8  0.250000  10  0.40 
Spades  8  0.250000  10  0.40 
Clubs  11  0.250000  10  0.10 
N  DF  ChiSq  PValue 

40  3  1.80  0.6149 
Select your operating system below to see a stepbystep guide for this example.
All expected values are at least 5 so we can use the chisquare distribution to approximate the sampling distribution. Our results are \(\chi^2 (3) = 1.80\). \(p = 0.6149\). Because our pvalue is greater than the standard alpha level of 0.05, we fail to reject the null hypothesis. There is not evidence that the proportions are different in the population.
The example above tested equal population proportions. Minitab Express also has the ability to conduct a chisquare goodnessoffit test when the hypothesized population proportions are not all equal. To do this, you can choose to test specified proportions or to use proportions based on historical counts:
11.2.2.1  Video Example: Tulips (Summarized Data, Equal Proportions)
11.2.2.1  Video Example: Tulips (Summarized Data, Equal Proportions)A company selling tulip bulbs claims they have equal proportions of white, pink, and purple bulbs and that they fill customer orders by randomly selecting bulbs from the population of all of their bulbs.
You ordered 30 bulbs and received 16 white, 8 pink, and 6 purple.
Is there evidence the bulbs you received were not randomly selected from a population with an equal proportion of each color?
Minitab Express is used to conduct a hypothesis test to address this research question.
11.2.2.2  Video Example: Roulette (Summarized Data, Different Proportions)
11.2.2.2  Video Example: Roulette (Summarized Data, Different Proportions)11.3  ChiSquare Test of Independence
11.3  ChiSquare Test of IndependenceThe chisquare (\(\chi^2\)) test of independence is used to test for a relationship between two categorical variables. Recall that if two categorical variables are independent, then \(P(A) = P(A \mid B)\). The chisquare test of independence uses this fact to compute expected values for the cells in a twoway contingency table under the assumption that the two variables are independent (i.e., the null hypothesis is true).
Even if two variables are independent in the population, samples will vary due to random sampling variation. The chisquare test is used to determine if there is evidence that the two variables are not independent in the population using the same hypothesis testing logic that we used with one mean, one proportion, etc.
Again, we will be using the five step hypothesis testing procedure:
The assumptions are that the sample is randomly drawn from the population and that all expected values are at least 5 (we will see what expected values are later).
Our hypotheses are:
\(H_0:\) There is not a relationship between the two variables in the population (they are independent)
\(H_a:\) There is a relationship between the two variables in the population (they are dependent)
Note: When you're writing the hypotheses for a given scenario, use the names of the variables, not the generic "two variables."
 ChiSquare Test Statistic
 \(\chi^2=\sum \dfrac{(ObservedExpected)^2}{Expected}\)
 Expected Cell Value
 \(E=\dfrac{row\;total \; \times \; column\;total}{n}\)
The pvalue can be found using Minitab Express. Look up the area to the right of your chisquare test statistic on a chisquare distribution with the correct degrees of freedom. Chisquare tests are always righttailed tests.
 Degrees of Freedom: ChiSquare Test of Independence
 \(df=(number\;of\;rows1)(number\;of\;columns1)\)
If \(p \leq \alpha\) reject the null hypothesis.
If \(p>\alpha\) fail to reject the null hypothesis.
Write a conclusion in terms of the original research question.
11.3.1  Example: Gender and Online Learning
11.3.1  Example: Gender and Online LearningA sample of 314 Penn State students was asked if they have ever taken an online course. Their genders were also recorded. The contingency table below was constructed. Use a chisquare test of independence to determine if there is a relationship between gender and whether or not someone has taken an online course.
Have you taken an online course?  

Yes  No  
Men  43  63 
Women  95  113 
\(H_0:\) There is not a relationship between gender and whether or not someone has taken an online course (they are independent)
\(H_a:\) There is a relationship between gender and whether or not someone has taken an online course (they are dependent)
Looking ahead to our calculations of the expected values, we can see that all expected values are at least 5. This means that the sampling distribution can be approximated using the \(\chi^2\) distribution.
In order to compute the chisquare test statistic we must know the observed and expected values for each cell. We are given the observed values in the table above. We must compute the expected values. The table below includes the row and column totals.
Have you taken an online course?  

Yes  No  
Men  43  63  106 
Women  95  113  208 
138  176  314 
\(E=\dfrac{row\;total \times column\;total}{n}\) 
\(E_{Men,\;Yes}=\dfrac{106\times138}{314}=46.586\) 
\(E_{Men,\;No}=\dfrac{106\times176}{314}=59.414\) 
\(E_{Women,\;Yes}=\dfrac{208\times138}{314}=91.414\) 
\(E_{Women,\;No}=\dfrac{208 \times 176}{314}=116.586\) 
Note that all expected values are at least 5, thus this assumption of the \(\chi^2\) test of independence has been met.
Observed and expected counts are often presented together in a contingency table. In the table below, expected values are presented in parentheses.
Have you taken an online course?  

Yes  No  
Men  43 (46.586)  63 (59.414)  106 
Women  95 (91.414)  113 (116.586)  208 
138  176  314 
\(\chi^2=\sum \dfrac{(OE)^2}{E} \)
\(\chi^2=\dfrac{(4346.586)^2}{46.586}+\dfrac{(6359.414)^2}{59.414}+\dfrac{(9591.414)^2}{91.414}+\dfrac{(113116.586)^2}{116.586}=0.276+0.216+0.141+0.110=0.743\)
The chisquare test statistic is 0.743
\(df=(number\;of\;rows1)(number\;of\;columns1)=(21)(21)=1\)
We can determine the pvalue by constructing a chisquare distribution plot with 1 degree of freedom and finding the area to the right of 0.743.
\(p = 0.388702\)
\(p>\alpha\), therefore we fail to reject the null hypothesis.
There is not evidence that gender and whether or not an individual has completed an online course are related.
Note that we cannot say for sure that these two categorical variables are independent, we can only say that we do not have evidence that they are dependent.
11.3.2  Minitab Express: Test of Independence
11.3.2  Minitab Express: Test of IndependenceIf you have a data file with the responses for individual cases then you have "raw data" and can follow the directions below. If you have a table filled with data, then you have "summarized data." There is an example of conducting a chisquare test of independence using summarized data on a later page. After data entry the procedure is the same for both data entry methods.
MinitabExpress – Chisquare Test Using Raw Data
Research question: Is there a relationship between where a student sits in class and whether they have ever cheated?
Null hypothesis: Seat location and cheating are not related in the population.
Alternative hypothesis: Seat location and cheating are related in the population.
To perform a chisquare test of independence in Minitab Express using raw data:
 Open Minitab data set
 On a PC: Select STATISTICS > Cross Tabulation and ChiSquare
On a Mac: Select Statistics > Tables > Cross Tabulation and ChiSquare  Doubleclick on the variable Seating to insert it into the Rows box
 Doubleclick on the variable Ever Cheat to insert it into the Columns box
 Click the Display tab and check the boxes Chisquare test for association and Expected cell counts
 Click OK
This should result in the following output:
Cell contents grouped by Back, Front, Middle, All; First row: count, Next row: expected count
No  Yes  All  

Back  24  8  32 
24.21  7.79  
Front  38  8  46 
34.81  11.19  
Middle  109  39  148 
111.98  36.02  
All  1714  55  226 
ChiSquare  DF  PValue  

Pearson  1.54  2  0.4633 
Likelihood Ratio  1.63  2  0.4435 
Select your operating system below to see a stepbystep guide for this example.
All expected values are at least 5 so we can use the Pearson chisquare test statistic. Our results are \(\chi^2 (2) = 1.54\). \(p = 0.4633\). Because our \(p\) value is greater than the standard alpha level of 0.05, we fail to reject the null hypothesis. There is not evidence of a relationship in the population between seat location and whether a student has cheated.
11.3.2.1  Video Example: Dog & Cat Ownership (Raw Data)
11.3.2.1  Video Example: Dog & Cat Ownership (Raw Data)This example uses the dataset:
\(H_0:\) There is not a relationship between dog ownership and cat ownership in the population of all World Campus STAT 200 students
\(H_a:\) There is a relationship between dog ownership and cat ownership in the population of all World Campus STAT 200 students
Cell contents grouped by No, Yes, Missing; First row: count, Next row: expected count
No  Yes  All  

BaNock  183  69  252 
176.02  75.98  
Yes  183  89  272 
189.98  82.02  
Missing  1  0  
All  366  158  524 
Assumption: All expected counts are at least 5. The expected counts here are 176.02, 75.98, 189.98, and 82.02, so this assumption has been met.
ChiSquare  DF  PValue  

Pearson  1.77  1  0.1833 
Likelihood Ratio  1.77  1  0.1828 
Since the assumption was met in step 1, we can use the Pearson chisquare test statistic.
\(Pearson\;\chi^2 = 1.77\)
\(p = 0.1833\)
Our p value is greater than the standard 0.05 alpha level, so we fail to reject the null hypothesis.
There is not evidence of a relationship between dog ownership and cat ownership in the population of all World Campus STAT 200 students.
11.3.2.2  Video Example: Coffee and Tea (Summarized Data)
11.3.2.2  Video Example: Coffee and Tea (Summarized Data)11.3.3  Relative Risk
11.3.3  Relative RiskA chisquare test of independence will give you information concerning whether or not a relationship between two categorical variables in the population is likely. As was the case with the single sample and two sample hypothesis tests that you learned earlier this semester, with a large sample size statistical power is high and the probability of rejecting the null hypothesis is high, even if the relationship is relatively weak. In addition to examining statistical significance by looking at the p value, we can also examine practical significance by computing the relative risk.
In Lesson 2 you learned that risk is often used to describe the probability of an event occurring. Risk can also be used to compare the probabilities in two different groups. First, we'll review risk, then you'll be introduced to the concept of relative risk.
The risk of an outcome can be expressed as a fraction or as the percent of a group that experiences the outcome.
 Risk
 \(Risk=\dfrac{n\;with\;outcome}{total\;n}\)
Examples of Risk
Asthma
60 out of 1000 teens have asthma. The risk is \(\frac{60}{1000}=.06\). This means that 6% of all teens experience asthma.
Flu
45 out of 100 children get the flu each year. The risk is \(\frac{45}{100}=.45\) or 45%
 Relative Risk
 Relative risk compares the risk of a particular outcome in two different groups.
 Relative Risk
 \(Relative\ Risk=\dfrac{Risk\ in\ Group\ 1}{Risk\ in\ Group\ 2}\)
Thus, relative risk gives the risk for group 1 as a multiple of the risk for group 2.
Examples of Relative Risk
Asthma
Suppose, 7% of teenage girls have asthma compared to 5% of teenage boys with asthma. Using the females as group 1, what is the relative risk for girls compared to boys?
 \(Relative\;risk=\dfrac{7 \%}{5 \%}=1.4\)
The risk of asthma is 1.4 times as great for teenage girls as it is for teenage boys.
Flu
Suppose that the risk of a child getting the flu this year is .45 and the risk of an adult getting the flu this year is .10. What is the relative risk of children compared to adults?
 \(Relative\;risk=\dfrac{.45}{.10}=4.5\)
Children are 4.5 times more likely than adults to get the flu this year.
Caution
Watch out for relative risk statistics where no baseline information is given about the actual risk. For instance, it doesn't mean much to say that beer drinkers have twice the risk of stomach cancer as nondrinkers unless we know the actual risks. The risk of stomach cancer might actually be very low, even for beer drinkers. For example, 2 in a million is twice the size of 1 in a million but is would still be a very low risk. This is known as the baseline with which other risks are compared.
11.4  Lesson 11 Summary
11.4  Lesson 11 SummaryObjectives
 Construct a chisquare probability distribution plot in StatKey or Minitab Express.
 Determine when a chisquare goodnessoffit test and chisquare test of independence should be conducted.
 Compute and interpret expected counts.
 Conduct chisquare tests by hand and using Minitab Express.
 Calculate and interpret relative risk.
In this lesson you added two new statistical procedures to your repertoire: the chisquare goodnessoffit test for one categorical variable and the chisquare test of independence for two variables. In the remaining weeks we will continue to learn more procedures.