Statistics paper:

Week 2 will be our first discussion board related to the final project/paper so I will take a moment now to provide some more detail on the paper and my expectations. The paper should be a minimum of 7 pages in APA format, submitted in Word. I know this post contains a lot of information but please make sure you read through this announcement as it will help guide you, starting with selecting a topic.

The goal is to come up with a topic and question that you want to research further with the use of statistics. Statistical tests work best if you have an array of data points that you are comparing to something else (a concrete number or another set of data points). Please do not try to determine whether you can prove that your job/department requires additional staffing – I have found that students who use this popular topic struggle with their papers. This paper can be related to your job but does not have to be!

To be successful in this paper, I strongly suggest you use a t-test or a correlation/regression analysis. You will learn what these mean throughout the class but it may be helpful to look these up on a high level to get a better understanding of how to frame your papers.

Option 1 (typically a t-test) – compare data to a concrete data point: Perhaps you hate the snow in Massachusetts and your significant other promised you that you could move south if you could statistically prove that the snow totals in Massachusetts are usually more than 15 inches. You could gather data from the last 20 years and perform a test to compare those totals to 15 inches. That is an example of how you can use a concrete number (15) to compare to your data.

Option 2 (typically a t-test) – compare data to another set of data to determine which is more/less: Perhaps you are choosing between two towns in Massachusetts and you want to choose the one that statistically gets the least amount of snow. You could gather data from the last 20 years and perform a test to compare both sets of data points. That is an example of how you can use another set of data as a comparison.

Option 3 (typically a correlation analysis) – compare data to another set of data to see if there is a relationship (i.e., as one increases does the other increase/decrease?): Perhaps you want to try to predict the snowfall potential in December and you believe that there is a statistically significant relationship between the preceding July’s average temperature and the snow totals in December. You could gather data from the last 20 years and perform a test to see if there is a relationship between these two sets of data.

A few tips:

1. Data collection: You will want to choose a topic that will allow you to gather data relatively easily – it is probably going to be difficult if your question is something like “Is there intelligent life somewhere other than Earth?”. Try to pick a topic that will allow you to either collect data or research it online.

2. Data sources: You are welcome to use any reputable resources for data or perform experiments or surveys. The Modesto Junior College Library and Learning Center has a great list of data sources at the following link: https://libguides.mjc.edu/statisticsguide/Topic For example:

a. If you want to know if a certain health issue occurs within a specific subpopulation, you might use a source like the World Health Organization or the National Center for Health Statistics.

b. If you ask “Are men or women more likely to help someone who has dropped their belongings?”, you might carry out an experiment of dropping your things in front of a variety of people and making note who helps you. I find carrying out an experiment is much more difficult for students but you are welcome to do this if you choose.

3. Other similar studies to your research: there may be studies that have already been performed on the topic you are interested in. Please do not submit a summary of another study’s findings as your paper. You are welcome to reference other studies to compare to your research, but your research must be unique and calculated by you.

4. Tests/techniques: You are required to use at least one statistic test, but you are encouraged to use more than one if relevant. As we go through the material each week you will be introduced to more ways to test data such as hypothesis testing, confidence intervals, and correlation/regression. Please do not draw conclusions based on simple numbers (e.g., “There were 1,000 cases of illness X in women last year and 1,200 of illness X in men last year so this illness is more prevalent in men.” – this is not proven statistically if you do not use a statistical test or technique to show if the results are significant! Note that you might find that your results are NOT significant, and that is an acceptable outcome as well)

5. Graphs: Sometimes it is helpful to begin with or incorporate a graph or other visual representation of your data. Please feel free to include graphics and/or charts that help summarize and show your process and findings, as these types of visual representations of data are helpful in a statistical analysis. Keep in mind that you will still need 7 pages of text in addition to your graphics.

6. APA Format – I have attached APA guidelines which may be helpful.

Here is the rubric by which you will be graded (and the content you should include!):

Points Available

Topic

Clearly defined and stated

10

Data Collection

Sources cited and data displayed clearly

10

Hypothesis

Clearly defined null and alternate hypothesis(es)

15

Statistical Test/Analysis

Relevant statistical tests performed and interpreted

30

Conclusion

Conclusions drawn from tests

15

Overall flow/grammar

Spelling and grammar

10

APA Style

At least 7 pages and follows APA guidelines

10

Good luck!

​
​

7 pages long, APA formatting. Using t test comparing aircraft fatalities from 1980-2000 vs 2001-2019.