WEEK 4 PROJECT

Week 4 Project

Week 4 Project

STAT 3001

Week 4 Project: Confidence Intervals and Hypothesis Testing

Descriptive Statistics

1) After you have opened the Statdisk program, go to Datasets and then Elementary Stats, 9th Edition.

Open the file named Cotinine. The data will appear in the Sample Editor.

There are three different sets of data values labeled Smokers, ETS, and No ETS. The data was collected for a study on the effects of second-hand smoke. The sample data consists of the measured serum cotinine levels in three different groups of people. The NOETS group lists the cotinine levels for subjects who are nonsmokers and have no exposure to environmental tobacco smoke at home or work. The ETS group lists cotinine levels for subjects who are nonsmokers exposed to tobacco smoke at home or work. The SMOKERS group lists cotinine levels for subjects who report tobacco use. Serum cotinine is a metabolite of nicotine, meaning that cotinine is produced when nicotine is absorbed by the body. Higher levels of cotinine correspond to higher levels of exposure to smoke that contains nicotine.

2) Generate descriptive statistics for each data set (Smokers, ETS, and No ETS). Copy the descriptive statistics from Statdisk v 11.1.0 and paste it here. Be sure to change the font color to red.

Smoker

ETS

No ETS

Descriptive Statistics

Descriptive Statistics

Descriptive Statistics

Column 1 Smokers

 

Column 2 ETS

 

Column 3 No ETS

 

 

 

 

 

 

 

Sample Size, n: 40

 

Sample Size, n: 40

 

Sample Size, n: 40

 

Mean: 172.475

 

Mean: 60.575

 

Mean: 16.35

 

Median: 170

 

Median: 1.5

 

Median: 0

 

Midrange: 245.5

 

Midrange: 275.5

 

Midrange: 154.5

 

RMS: 208.9748

 

RMS: 149.1973

 

RMS: 63.87488

 

Variance, s^2: 14279.85

Variance, s^2: 19067.17

Variance, s^2: 3910.438

St Dev, s: 119.4983

St Dev, s: 138.0839

St Dev, s: 62.5335

 

Mean Abs Dev: 94.775

Mean Abs Dev: 87.62

Mean Abs Dev: 29.6975

Range: 491

 

Range: 551

 

Range: 309

 

Coeff. Of Var. 69.28%

Coeff. Of Var. 227.96%

Coeff. Of Var. 382.47%

 

 

 

 

 

 

Minimum: 0

 

Minimum: 0

 

Minimum: 0

 

1st Quartile: 86.5

 

1st Quartile: 1

 

1st Quartile: 0

 

2nd Quartile: 170

 

2nd Quartile: 1.5

 

2nd Quartile: 0

 

3rd Quartile: 251.5

 

3rd Quartile: 32

 

3rd Quartile: 0

 

Maximum: 491

 

Maximum: 551

 

Maximum: 309

 

 

 

 

 

 

 

Sum: 6899

 

Sum: 2423

 

Sum: 654

 

Sum Sq: 1.746819e+6

Sum Sq: 890393

 

Sum Sq: 163200

 

3) Compare the boxplots for each data set. Put all three boxplots on one graph. Once you have all three boxplots on one graph, copy and paste that graph here.

Confidence Levels

4) Create a 95% confidence interval about the mean for each data set. Copy the confidence intervals here.

Smoker

 

 

 

 

 

95% Confident the population mean is within the range:

134.2576 < mean<210.6924

 

 

 

ETS

 

 

 

 

 

95% Confident the population mean is within the range:

16.41367 < mean<104.7363

 

 

 

NO ETS

 

 

 

 

 

95% Confident the population mean is within the range:

-3.649164 < mean < 36.34916

 

 

 

5) Compare the confidence intervals for each data set. Put all three confidence intervals on one graph using the tools in your word processor (example below). Statdisk cannot do that for you. Enter your graph and turn the font red. For this process, I just use the dashes and write a scale below the axes. Here is an example, but it is not based on the data you are analyzing:

No ETS -3.65 -----36.35

Smoking 16.41 -----------------------104.74

ETS 134.26 -------------------210.70

______________________________________________________________________________ -3 20 40 60 80 100 120 140 160 180 200 220

6) Based on the confidence intervals, does there appear to be enough evidence to ...