To calculate the standard error for confidence intervals or for a hypothesis test using the correct formula.
To check Validity Conditions for Inference on the Difference of Two Proportions
To apply Theory-Based methods for Confidence Intervals for Two Proportions.
Draw appropriate conclusions from Theory-based techniques for Two Proportions
As usual, we start by loading our two packages: mosaic
and ggformula
. To load a package, you use the
library()
function, wrapped around the name of a package.
I’ve put the code to load one package into the chunk below. Add the
other package you need.
library(mosaic)
library(ggformula)
# put in the other package that you need here
We’ll load the example data, GSS_clean.csv
from this
Url: https://raw.githubusercontent.com/IJohnson-math/Math138/main/GSS_clean.csv
and use the read.csv()
function.
#load data
<- read.csv("https://raw.githubusercontent.com/IJohnson-math/Math138/main/GSS_clean.csv") GSS
We also need to do a little data cleaning to ensure this will work properly for the lab,
<- filter(GSS, should_marijuana_be_made_legal != "")
GSS <- filter(GSS, self_emp_or_works_for_somebody != "") GSS
Our research question is whether there is a difference in the
proportion of people who said marijuana should be made legal in the two
groups of people that are self employed or work for somebody else. We
think of the self_emp_or_works_for_somebody
as the
explanatory variable and should_marijuana_be_made_legal
as
the response variable.
Our null hypothesis is the proportion of people that believe marijuana should be made legal is the same in the self employed group as it is in the work for someone else group. In other words, there is no association between thinking marijuana should be legal and whether a person works for someone else or is self employed.
Let \(\pi_{selfEmp}\) be the proportion of Self Employed people that think marijuana should be legal and \(\pi_{someoneElse}\) be the proportion of people that work for someone else that think marijuana should be legal.
Our null and alternative hypotheses are
\[H_0 : \pi_{selfEmp} - \pi_{someoneElse} = 0\] \[H_a : \pi_{SelfEmp} - \pi_{SomeoneElse} \neq 0\]
Let’s start by creating a bar chart to visualize the data. We want to graph the two groups, self employed or work for somebody else, and see in each bar those that believe marijuana should be legal and those that don’t.
Here is the most basic bar chart of counts.
gf_bar( ~self_emp_or_works_for_somebody, fill= ~ should_marijuana_be_made_legal, data=GSS)
Here is a bar chart of counts that doesn’t have the counts stacked and instead has them positioned side-by-side.
gf_bar( ~self_emp_or_works_for_somebody, fill= ~ should_marijuana_be_made_legal, data=GSS, position ="dodge" )
Here is a segmented bar graph. Notice that the command has changed to
gf_props
instead of gf_bar
.
gf_props( ~self_emp_or_works_for_somebody, fill= ~ should_marijuana_be_made_legal, data=GSS, position ="fill" )
We create a 2-way table with the command tally
to
determine the proportion of self employed people that believe marijuana
should be made legal and the proportion of people that work for someone
else that believe marijuana should be made legal.
Important note: the order of the variables matters!!
It should be tally( response_var ~ explanatory_var)
. Be
careful or your proportions will be incorrect
The first code chunk is a table of counts and the second is a table of proportions.
tally(should_marijuana_be_made_legal ~ self_emp_or_works_for_somebody, data=GSS)
## self_emp_or_works_for_somebody
## should_marijuana_be_made_legal Self-employed Someone else
## Legal 98 809
## Not legal 44 442
We can calculate the sample size for each group: \(n_1\) is the number of people that are self-employed and \(n_2\) the number of people that work for someone else.
= 98+44
n1 n1
## [1] 142
= 809+442
n2 n2
## [1] 1251
tally(should_marijuana_be_made_legal ~ self_emp_or_works_for_somebody, data=GSS, format="proportion")
## self_emp_or_works_for_somebody
## should_marijuana_be_made_legal Self-employed Someone else
## Legal 0.6901408 0.6466827
## Not legal 0.3098592 0.3533173
Validity Conditions: The theory-based test and interval for the difference in two proportions (called a two-sample z-test or interval) work well when there are at least 10 observations in each of the four cells of the 2 × 2 table.
If we look at the tally
of counts, we see that the
values in the 2 x 2 table are 98, 44, 809, 442, all of which are greater
than 10. So our validity conditions are satisfied.
Let’s start by finding our observed statistic.
<- 0.6901408 - 0.6466827
p_diff p_diff
## [1] 0.0434581
For two proportions, in a hypothesis test the standard error of the null distribution is given by
\[ SE=\sqrt{\frac{\hat{p}(1-\hat{p})}{n_1}+\frac{\hat{p}(1-\hat{p})}{n_2}} \] where \(\hat{p}\) is the pooled proportion.
Using R as a calculator the pooled proportion is
<- (98+809)/1393
phat phat
## [1] 0.6511127
The Standard error is
<- sqrt(phat*(1-phat)/(n1) + phat*(1-phat)/(n2))
SE SE
## [1] 0.04220592
Next, we can calculate the standardized statistic using the formula
\[ z = \frac{\hat{p}_1 - \hat{p}_2 - 0}{\sqrt{\frac{\hat{p}(1-\hat{p})}{n_1}+\frac{\hat{p}(1-\hat{p})}{n_2}}} = \frac{\hat{p}_{diff} - 0}{\sqrt{\frac{\hat{p}(1-\hat{p})}{n_1}+\frac{\hat{p}(1-\hat{p})}{n_2}}}\]
<- (0.6901408 - 0.6466827)/SE
z z
## [1] 1.029668
What does this standardized statistic suggest regarding our hypothesis test?
The standardized statistic is not greater than 2 or less than -2, so we don’t have enough evidence to reject the null hypothesis. It looks like the difference in proportions of people that believe that marijuana should be legal between those that are self-employed and those that work for someone else could be zero.
Next we calculate the theory based \(p\)-value using prop.test
and
adjust the code below to calculate a 90% confidence interval for our
difference in proportions. Note: In the code below we will omit the
default continuity correction (using the option
correct= FALSE
because the counts in all four cells of the
two-way table are large. The continuity correction becomes important if
one of the cell counts is small, especially if a count is less than or
equal to 5.
#inference for two proportions
prop.test(should_marijuana_be_made_legal ~ self_emp_or_works_for_somebody, data = GSS, success = "Legal", alternative = "two.sided", conf.level = 0.95, correct=FALSE)
##
## 2-sample test for equality of proportions without continuity correction
##
## data: tally(should_marijuana_be_made_legal ~ self_emp_or_works_for_somebody)
## X-squared = 1.0602, df = 1, p-value = 0.3032
## alternative hypothesis: two.sided
## 95 percent confidence interval:
## -0.03708184 0.12399822
## sample estimates:
## prop 1 prop 2
## 0.6901408 0.6466827
Interpret the confidence interval: We are 90% confident that the difference in proportion of people that believe that marijuana should be legal between those that are self-employed and those that work for someone else is between -0.03 and 0.11. Since this confidence interval contains 0, we are not confident that these two proportions are significantly different.
#USE THIS COMMAND if you only have the counts and not the data
# c(98, 809) are the success counts for the two groups: self employed or works for someone else
# c(142, 1251) are the sample size counts for the two groups
prop.test(c(98, 809), c(142, 1251), alternative = "two.sided", conf.level = 0.95, correct=FALSE)
##
## 2-sample test for equality of proportions without continuity correction
##
## data: c out of c98 out of 142809 out of 1251
## X-squared = 1.0602, df = 1, p-value = 0.3032
## alternative hypothesis: two.sided
## 95 percent confidence interval:
## -0.03708184 0.12399822
## sample estimates:
## prop 1 prop 2
## 0.6901408 0.6466827
Does the \(p\)-value from
prop.test
support the conclusion made with the standardized
statistic?
To do find confidence intervals for a difference of proportions, we start by computing the standard error. Recall that the formula is different based on whether we’re doing a confidence interval or a hypothesis test. This is because a hypothesis test has a hypothesized value for the observed statistic and confidence intervals do not.
For two proportions, the standard error for a confidence interval is given by \[ SE = \sqrt{\frac{\hat{p}_1(1-\hat{p}_1)}{n_1} + \frac{\hat{p}_2(1-\hat{p}_2)}{n_2}} \] This formula is slightly different than the formula used above because we have no hypothesized value for the difference in proportions, so we use our observed proportions \(\hat{p}_1\) and \(\hat{p}_2\) instead of the pooled proportion. Thus, our margin of error (MOE) is given by
<- 0.6901408
p1<- 0.6466827
p2<- 2*sqrt(p1*(1-p1)/(98+44) + p1*(1-p2)/(809+442))
MOE MOE
## [1] 0.08248336
So the endpoints of our 2SD confidence interval are
<- p1-p2
p_diff
<- p_diff - MOE
left left
## [1] -0.03902526
<- p_diff +MOE
right right
## [1] 0.1259415
Does this align with a 95% confidence interval calculated using
prop.test
? Yes!
We are 95% confident that the difference in population proportions (\(\pi_{SelfEmp} − \pi_{SomeoneElse}\)) is between -0.04 and 0.13.
To investigate whether giving chest-compression-only (CC)
instructions rather than standard cardiopulmonary resuscitation (CPR)
instructions to the witness of a heart attack will improve the victim’s
chance of surviving, researchers Hupfl et al. (The Lancet, 2010)
combined the results from three randomized experiments. In each
experiment, the emergency services dispatcher randomly assigned either
CC or CPR instructions to the bystander who was at the site where a
person had just experienced a heart attack. The data they collected is
located at
https://raw.githubusercontent.com/IJohnson-math/Math138/main/CPR.csv
.
CPR
. How many observational units are
there? What are the names and types of the variables? Which variable is
the explanatory variable? Which is the response?Observational units and number:
Variables and type:
Define (in words) the parameters of interest of this study. Also, assign symbols to the parameters.
State the appropriate null and alternative hypotheses in words and symbols to address the research question of whether instructions for chest-compressions-only rather than standard cardiopulmonary resuscitation (CPR) instructions to the witness of a heart attack will improve the victim’s chance of surviving. (Hint: Remember that hypotheses are always about population parameters, and think about whether the alternative should be one- or two-sided before you see the data.)
\[H_o: \]
\[H_a: \]
Make a graph to visualize the proportion of survivors in the two treatment groups.
Create two-way tables with one showing the counts and another containing the proportions of survivors in each of the treatment groups.
Use R as a calculator to find and display
Check the Validity Conditions for a Two Proportion Inference Test. Explain what you are checking, any numerical values you are comparing, and whether or not the conditions have been met.
Use the proper command in R to calculate the theory based \(p\)-value for the hypothesis test. Are you using the continuity correction? Why or why not?
Calculate a theory-based 90% confidence interval and interpret the resulting interval in the context of the study.
Based on your findings, state a complete conclusion about the study. Be sure to address significance (p-value and standardized statistic), estimation (confidence interval), causation, and generalization.
Significance with context:
Estimation with interpretation:
Causation:
Generalization: