StatCrunch logo (home)

Report Properties
Thumbnail:

from Flickr
Created: Jan 22, 2016
Share: yes
Views: 1803
Tags:
 
Results in this report
 
Data sets in this report
None
 
Need help?
To copy selected text, right click to Copy or choose the Copy option under your browser's Edit menu. Text copied in this manner can be pasted directly into most documents with formatting maintained.
To copy selected graphs, right click on the graph to Copy. When pasting into a document, make sure to paste the graph content rather than a link to the graph. For example, to paste in MS Word choose Edit > Paste Special, and select the Device Independent Bitmap option.
You can now also Mail results and reports. The email may contain a simple link to the StatCrunch site or the complete output with data and graphics attached. In addition to being a great way to deliver output to someone else, this is also a great way to save your own hard copy. To try it out, simply click on the Mail link.
Week 2 Report--Cholesterol Levels
Mail   Print   Twitter   Facebook

Introduction:

Cholesterol levels can be an important indicator of heart disease and potential risk for heart disease, especially if the total cholesterol is 200 mg/dL or greater.  Cholesterol is measured via a blood test and is typically ordered as part of a lipid panel.  Hypertension, or high blood pressure is another risk factor for heart disease and heart attack (www.webmd.com; retrieved Jan. 21, 2016).  The goal of this study is to evaluate the relationship between cholesterol levels and primary hypertension (defined as systolic of greater than 140 and a diastolic greater than 90) and patients that are normotensive (blood pressure below 120/80) (www.webmd.com;  retrieved Jan. 21, 2016).

Statistical Methods:


Data was collected using a random sample collection.  133 primary hypertensive patients and 41 normotensive patients were used in the study with each having a total cholesterol level collected and reported.

This chart shows the cholesterol levels for patients with primary hypertension (PH).  The data is distributed showing the number and percentage of PH patients falling into the following cholesterol levels--100-150, 150-200, 200-250, and 250-300.

Result 1: 0103 Cholesterol Levels-PH Pie Chart With Data   [Info]
Right click to copy


This chart shows the same data as above, but for those patients that are Normotensive (NT).  The data is distributed in the same way.

Result 2: 0103 Cholesterol Levels-NT Pie Chart With Data   [Info]
Right click to copy

The following side by side bar graphs (with PH on the left and NT on the right), illustrate the counts of each that fall into each cholesterol level listed.  This graph illustrates much more clearly that there was quite a difference in the sample size (n=133 for PH and n=41 for NT).

Result 3: 0103 Cholesterol Levels Bar Plot With Data Side by Side   [Info]
Right click to copy

The following boxplot shows the distribution of cholesterol levels between the PH and NT groups, and from this boxplot, it is much easier to see the relationship between elevated levels of cholesterol and and elevated number of patients with PH as compared to NT.

Result 7: 0103 Boxplot PH and NT patients   [Info]
Right click to copy

Statistical Analysis:

The following charts provide the statistical data.  The first chart shows that the sample mean cholesterol level for PH patients is 214.71 and for NT patients, is 193.20.  The p-value for both is <0.0001, which confirms that the hypothesis that there is a relationship between elevated cholesterol levels and primary hypertension.

Result 4: 0103 Cholesterol Levels One Sample T   [Info]
Hypothesis test results:
μ : Mean of variable
H0 : μ = 0
HA : μ ≠ 0
VariableSample MeanStd. Err.DFT-StatP-value
PH214.706772.571483913283.495281<0.0001
NT193.195125.16073594037.435577<0.0001

The 95% confidence interval chart illustrates that we can estimate with 95% confidence that the average cholesterol for PH patients is between 209.62 mg/dL and 219.79 mg/dL and for NT patients is between 182.76 mg/dL and 203.63 mg/dL.

Result 6: 0103 Cholesterol Levels One Sample T   [Info]
95% confidence interval results:
μ : Mean of variable
VariableSample MeanStd. Err.DFL. LimitU. Limit
PH214.706772.5714839132209.62012219.79342
NT193.195125.160735940182.76489203.62536

Results/Conclusions/Future Considerations:

Based on the results, it appears that there is a correlation between patients that high elevated cholesterol levels and primary hypertension when compared to those with cholesterol ranges in normal levels that are normotensive.  One caution when looking at the results, is that the sample sizes are variable--for the PH patients, a total of 133 were sampled, while only 41 NT patients were sampled.  In the future, it may be better to to have n-values that are closer together.

Result 5: 0103 Cholesterol Levels Summary Stats   [Info]
Summary statistics:
ColumnnMeanVarianceStd. dev.Std. err.MedianRangeMinMax
PH133214.70677879.4663929.6557992.5714839214167117284
NT41193.195121091.96133.0448335.1607359194165121286

HTML link:
<A href="https://www.statcrunch.com/5.0/viewreport.php?reportid=55871">Week 2 Report--Cholesterol Levels</A>

Comments
Want to comment? Subscribe
Already a member? Sign in.
By nku.katie.waters
Jan 25, 2016

Hi Karen,
Good work! A few things:
1. Nice job at breaking your report into sections (Intro, Methods, etc.). This is the format that you want to continue to use in future reports.

2. Since the variable, cholesterol levels, is interval/ratio, the pie charts and bar charts are not appropriate. Pie charts are typically used for nominal data, and bar charts are typically used for ordinal data. Boxplots and histograms are appropriate ways of displaying interval/ratio data, so your boxplots in Result 7 are fine.

3. It's important to keep in mind that pie charts, bar charts, boxplots, and histograms are all descriptive statistics which means that they apply to the sample only. So consider using phrases like "In this sample" or "the results seem to suggest" in your statements about descriptive statistics. For this problem (and majority of this course), we are more interested in inferential statistics (like hypothesis tests and confidence intervals) as they allow us to make generalizations about the entire population (not just the sample).

4. For the hypothesis test, the null and alternative hypotheses should be set to 200 and the alternative should be "greater than" instead of "not equal". This is because the alternative hypothesis is that the average cholesterol level in the PH group is greater than 200.

5. When interpreting the hypothesis, it is not necessary to discuss the sample mean since it applies to the sample only. A correct interpretation of the hypothesis test would be, "Because the p-value is less than 0.05, we have strong evidence showing that the average cholesterol level for the PH group is greater than 200."

6. Great job interpreting the confidence intervals!

7. Be careful using the word "correlation". We typically reserve this word for when we discuss regression analysis later in the semester. Perhaps a better word here would be "relationship".

Please take a look at the solutions and let me know if you have any questions!
By michellewilliamson34
Jan 23, 2016

Very nice report. Like the way you displayed in in many different ways for descriptive using pie and graphs. Good interpretation f the CI also with parameters

Always Learning