Week 14 Term

1

MAT 2215 Project Guidelines • You are a primary researcher. Select a topic of interest to you or you’ll likely to get very bored. • Form a specific research question or more than one question. Use a sample of at least 25 to

give you enough data to work with. • Include question(s) that generate a range of quantitative data, discrete or continuous. If you

only ask “yes or no” questions you may not have much data to work with • Write your questions so that they are easy for people to understand and to answer. Questions

such as “On average, how many minutes a day do you spend texting?” or “On average how many calories do you consume each day?” are difficult to answer because most people probably either don’t have the answer or would have a tough time finding it.

• The wording in the question shouldn’t be vague or ambiguous. Questions such as “On average, how many healthy meals do you have each day?” or “On average, how many bottles of water do you drink each day?” are not helpful. The meaning of healthy meal may be different for different people and a bottle may mean 12 ounces or it may mean 20 ounces.

• Writing suggestions for statistical reports: a) Avoid definitive language when presenting your findings. Rather than writing “Satisfaction

levels are higher among drivers of hybrid cars”, it’s better to write ““Satisfaction levels seem (or tend to be) higher among drivers of hybrid cars” Remember that your findings are based on a sample and may not be true about the population, even if the sample is representative.

b) Avoid references to yourself. For example, rather than writing “I found …” it’s better to write “The data show…”

c) Avoid giving opinions. Stick to the facts and what the data show. If you choose to offer opinions, do so in a separate section at the end.

What to submit

o Your project should look and read like a small report (2-3 pages). Below are key points to address. Write a small paragraph for each one.

o You don’t need to write in essay form. You can use the question/answer format below. Whatever you choose, make your work reader-friendly

o At the end of the report, include a table with the actual data that you used.

§ Introduction Which question(s) are you trying on answer?

Here you’d describe what you’re interested in measuring or comparing. For example, “I want to know whether owners of hybrid cars are as satisfied with performance as owners of cars that run on gasoline”

Why is the question important? Here you will describe why the results of your work might be important and to whom. For example, you might say, “the results would be important to potential car buyers because …”, or “I’m not sure how the results are important but I’m very curious”, etc.

What is your target population? Here you’ll describe the population (or populations) of interest. Using the cars example, your populations would be (1) all owners of hybrid cars and (2) all owners of gas cars

2

§ Data Collection

What type of sample did you use? (random, stratified, convenience, etc.) Here you describe how you collected data. With the cars example, ideally, you’d have two random and representative samples from across the country. However, given the limited resources, you’re not likely to have that, so describe the sample(s) you’d like to have and then describe the sample(s) you have. For example, you might say, “I asked 37 friends, family members and co-workers. This included 20 drivers of gas cars and 17 drivers of hybrids”

Explain whether your sample is representative

With limited resources, your sample is not likely to be representative. For the purposes of this project that’s OK but do state the shortcomings of your sample. In the cars example, your contacts make a convenience sample which is not representative.

§ Analysis: Statistical Methods & Graphs

What type of statistical analysis did you perform? Here you explain which statistical methods (summary statistics with mean and SD, regression analysis, confidence interval, hypothesis test, etc.) you used to help answer your original question(s) and the results of these methods. For example, if you want to compare satisfaction levels you might want to make a frequency distribution for drivers of hybrids and a frequency distribution for drivers of gas cars. If, in addition, to satisfaction levels you have the age of each driver, you could run two correlations: age vs. satisfaction for hybrids, and age vs correlation for gas. If you also have mileage data or data on cost repairs, you could provide a mean and standard deviation of each set. Depending on the analysis that you performed, you should report the values of the relevant statistics. For example, sample size, mean, correlation coefficient, margin of error, etc.

What type of graph(s) did you use to present your results? Here you include the graphs that summarize your data. The type of graphs that you use will depend on the type of data you collected. If you’re looking for a correlation you might include a scatter plot; if you’re comparing satisfaction levels, you might use two bar graphs; if you’re comparing cost repairs you might use two histograms or two box-whiskers plots. Whatever you use, don’t just paste a graph(s) in your report. The purpose of a graph is to help the reader see your results quickly. So, a) Label your graphs appropriately b) Keep your graphs simple and avoid bells & whistles (distracting features on graphs) c) You must refer to your graph(s) somewhere in your narrative. Usually, this is done at

the point where you explain your findings. You might say, for example, “… as Graph 1 shows, younger drivers of hybrid cars tend to be more satisfied people tend to be more satisfied with their hybrid car then older drivers of hybrid cars…”

3

§ Discussion & Conclusions

This is where you provide a summary of your findings. It would be OK to remind the reader the purpose of the study and any limitations of the study. For example, you might say “this study aimed at comparing the satisfaction levels of … Because a convenience sample was used, the findings from the study may not generalize to the population…” Then describe the key findings. For this project, it’s OK to use a bullet list for the findings. For example, you might say “The data show that: a) Drivers of hybrid cars reported being about equally satisfied as drivers of gas cars b) Young drivers of hybrid cars reported being much more satisfied than older drivers of

hybrid cars c) Etc.

Paste your data at the end of the report using a table with clear column titles. For example:

Hybrid Driver Satisfaction1 City Mileage2 Gas Driver Satisfaction City Mileage 1 4 31 1 4 24 2 1 34 2 5 25 3 4 37 3 5 24 4 1 33 4 2 33 5 5 43 5 4 34 6 4 41 6 5 34 7 5 38 7 2 24 8 3 39 8 5 31 9 3 39 9 3 23 … … … 10 2 25 … … … 11 1 28 … … … … … … … … … … … …

1Drivers used a 1 to 5 Likert scale to report their level of satisfaction with their car. (1=very unsatisfied, 5 = very satisfied) 2Drivers estimated the number of miles their car gets on a single gallon of gas, in the city