Click here to get an A+ paper at a Discount

- Obtaining data for the Final Project:

Many students in the past have designed surverys to collect the data they need for their final projects. It is also possible to analyze a data set collected by others. In either case, the key to obtaining data suitable for this course is to make sure that

There is a mixture of categorical and numerical variables.

Some of the variables are potentially related.

I am attaching an example of a survey as an example of the type and amount of data needed for your final projects. Your assignment is to design a survey or locate a data set for your final project. I will review your survey or a description of the data you intend to analyze and provide feedback before you actually hand out the surveys. This review is intended to prevent unnecessary problems once you begin to move forward with the data analysis and writing components of the project.

Collecting and organizing your data:

As you learned in class, random samples are unbiased and allow one to study the sample and learn about the population. When it come to your final project here are some recommendations about how to approach sampling.

Use a sample size between 30 and 50. Although the professional pollsters usually use a sample size of approximately 1000, this is not reasonable in an elementary statistics course.

Consider using a convenience sample. Although professional pollsters use random samples to avoid bias, it is not reasonable to perform a random sample of a large population in an elementary statistics course. I will expect that you include an honest assessment of your sampling methods in your final paper. The key here is to demonstrate your knowledge of appropriate sampling strategies and the disadvantages of convenience samples. The only realistic way for you to collect a random sample is to study a very small population such as Saint Joseph College students or perhaps several dorms on campus. In this case you can use a list of all the students and use StatCrunch to select a random sample.

Once you have collected your data, you can enter it in StatCrunch. Each question is a variable and will be stored in a separate column. Each person’s data will tehn be stored in a single row as shown in the diagram below.**CLICK HERE TO GET THIS PAPER WRITTEN**

A rough draft of your paper is due by Friday, April 11. Although the rough draft is not mandatory, it will allow students to obtain feedback before submitting the final version of the paper for a grade. In the past students have asked question concerning the format of the paper, I am attaching a published paper that incorporates statistics. Note the paper is well written and integrates statistics and graphs into the narrative. Here is a summary of what needs to be included in your final paper.

THINGS THAT SHOULD BE IN THE PAPER

Section 1:Â Background: Provide background information on your topic. This may involve giving a historical perspective and/or summarizing the results of previous research.Section 2:Â Data collection: Create a survey (include a mix of numerical and categorical variables) or design an experiment to answer your fundamental question. If conducting a poll, discuss whether or not you have used a random sample. If you have not used a random sample, clearly state the limitations of your poll. If you are conducting an experiment, be sure to clearly state how randomization was used.

Section 3:Â Data analysis: Apply the data analysis techniques learned this semester. Here is a brief summary of some of these techniques. You will not need to use all of these techniques. The key is to choose the techniques which best address your fundamental question.

Descriptive Statistics:

Numerical variables: Create stemplots, histograms, boxplots, compute five number summaries, mean and standard deviation.

Categorical variables: Create pie charts and bar charts. Also, categorical variables may be used to make comparisons. For example, in my caffeinated beverage survey, I could use the students resident/commuter status to form two groups; residents and commuters. I could then compare the caffeine consumption of the two groups using side by side boxplots. In StatCrunch, consider using Data/Split Columns when you need to separate numerical varaibles by category.

Inferential StatisticsNumerical variables:

Confidence intervals for means and significance tests including 1-sample t-test, 2-sample t-test and ANOVA.

Categorical variables: Confidence interval for proportions and significance tests including 1-proportion and the chi-square test for independence. Again, categorical variables may be used to make comparisons. For example, in my caffeinated beverage survey, I could use the students resident/commuter status to form two groups; residents and commuters. I could then compare the caffeine consumption of the two groups using a 2-sample t-test. This would complete the comparison initiated with the side-by-side box plots.Section 4:Â Conclusion: A paragraph (page at most) summarizing your project (regrets, key findings,…)

- Uploaded additional materials: