Field Project (Due on 5 May)
Copyright © 2003–2015 by Stan Brown
Copyright © 2003–2015 by Stan Brown
This is your chance to apply what you’ve learned about inferential statistics. You will pick a question to investigate, plan your project, gather sample data, do a hypothesis test to answer the question, and write a report.
The Field Project is worth 100 points, about 17% of your course grade. You can also earn extra credit by presenting your project to the class.
Yes, whether or not you have an approved Project Plan, you can email me or visit the Baker Center, or both.
You should pick a question that matters to you, probably in your professional field of interest, or perhaps a cherished hobby. Finding it hard to get started? The following ideas from past projects may help, but be creative!
Please feel free to ask me about your topic idea before you start gathering data on it.
For Cases 1 to 5, you need a sample of at least 40.
For Cases 6 and 7, plan your sample size so that all the E’s are at least 5.
If those sample sizes aren’t practical, please discuss sample size with me in advance.
Plan your sample — look back at Chapter 1 for survey techniques. You should know by now that “randomly” asking people is not random at all and is not okay.
In fact, asking people out loud is not okay because it makes response bias more likely. Use a written survey form, and have a box or bag for people to put their forms without handing them to you.
If any minors will be given the survey, you must have their parents’ consent. In practice, this will probably be too hard to accomplish.
Yes, TC3’s Director of Institutional Research has very specific rules. Your form must include:
Nothing on the form can identify the respondent personally.
Yes, you do, and I may not be able to accept your project if you don’t.
Attach survey approval(s) to your project report.
Not at all! You can do any form of observational study or experiment. The most interesting results come from experiments.
You learned good techniques in Chapter 1.
Bear in mind ethical issues relating to humans or other animals. See TC3 Guidelines for Research Involving Human Subjects. Please share your plans with me before you start, and we may need to consult the Director of Institutional Research.
Yes, if you have the raw data and do your own analysis “from scratch”. If you’re not sure that you have the data you need, please ask.
Cite the source of your data. For example, give the exact URL rather than a home page; give page numbers if using a book. See the Chicago Manual’s guide or any of the TC3 library’s citation guides.
Yes, English counts 10% of your grade on the project. I urge you to visit a Baker Center writing tutor if your English composition is shaky, whether English is your first language or not.
At least proofread your project. Every semester I get a few papers with glaring mistakes like words left out. Don’t let yours be one of them!
Staple or permanently fasten these items, in order:
If you have survey forms or other bulky data, put them in a separate envelope with your name on it, not sealed and not attached to the other materials.
If you’re using published data, check with me before making a printout of more than a few pages.
Here’s the list, with the number of points for each item.
Make your report as long as necessary to cover these items, but no longer — don’t try to make it more impressive by padding it.
Please follow this order, to ensure that you get full credit for your work.
|Sampling and Analysis (35 points)|
|4||Problem description in a paragraph or two. Reason for your interest? State in English the claim that you will test.|
|2||Data type. Be specific.|
|3||Case number (1 pt) and description (2 pts) from Inferential Statistics: Basic Cases.|
|5||Which population(s) are you trying to learn about? How large are they?|
How large was your sample and how did you decide on that sample size?
How, where, and when did you choose the members of the sample?
Don’t describe your sample as random unless it really is.
If you obtained data from a published source, explain how data were gathered.
|−8||If sample size is below requirement, unless approved by instructor.|
|−20||If a survey was done without needed approvals.|
|4||All significant sources of bias. How could they
be avoided, given more resources?
If you think the answer is “no sources of bias”, say so; don’t feel you must make something up. Refer to Chapter 1 of the textbook to be sure you know what “bias” means; it’s not just any error source.
|4||Summarize your data numerically.
Don’t list individual values at this point. For numeric data, give x̅, s, and n, plus the median if appropriate. For non-numeric data, show sample size and also a table with the frequency and relative frequency of each response.
|5||Summarize your data graphically. Need guidance? Review Chapter 2.|
|Hypothesis Test (46 points)|
|10||(1) Hypotheses in symbols (5 pts) and English
(5 pts, 10 for Cases 6 and 7).
H0: p = 0.1, 10% of car trips are made in a Subaru
H1: p > 0.1, more than 10% of car trips are made in a Subaru
(Case 6 and 7 hypotheses are in English only, 10 points.)
|2||(2) Appropriate α. Justify your choice.|
|5||(RC) Requirements check.|
|10||(3/4) TI-83 screen name, inputs, and outputs for computing test statistic and p-value.|
|5||(5) Decision rule in statistical language.|
|8||(6) Conclusion in English.
Make sure you actually answer the question or decide on the claim that you identified at the beginning of your report.
|6||Compute a confidence interval (2 pts), at an appropriate confidence level for the α from your HT (1 pt), interpreted in English (3 pts). (Cases 6 and 7 get these points free.)|
|+6||“Difficulty bonus” for Case 3, 4, or 5.|
|General (19 points)|
|10||Standard English spelling, grammar, etc.|
|5||Professional appearance (neat handwriting won’t count against you; scratch-outs will)|
|4||Original data sheets or published data attached at
end, or survey forms in envelope
Attach your original data sheets, no matter how messy they might be. Never rewrite data sheets to make them look neater. If your survey was on slips of paper, put them in an unsealed envelope of appropriate size.
If you obtained data from a published source, cite the source and include a copy, as explained above.
|—||Attach your Project Plan showing instructor’s approval, and Institutional Research’s approval for your survey form if any.|
|−5||If report is not stapled or otherwise permanently fastened. (A cover isn’t needed.)|
|Optional: Presentation to Class (extra credit, up to 15 points)|
To qualify for extra credit, you must have written approval for your Project Plan by the deadline, and you must follow the approved plan.
|1||Purpose of study clearly stated: What claim was being tested?|
|2||Data type and case number|
|1||Choice of α and the reason|
|2||Hypotheses clearly stated in symbols (except words for Cases 6 and 7)|
|1||Population clearly described and size given|
|2||Sample and sampling technique clearly described; largest possible bias (if any) briefly mentioned|
|1||p-value clearly given; no need for calculation details|
|2||Conclusion clearly stated in English; neutral language if appropriate|
|1||Audience questions fielded clearly; or questions were solicited but none asked|
|2||Presenter well prepared, no stumbling or floundering|
If you have an email or signature from me saying that I approve your Field Project Plan, you are eligible to present your project to the class on project day, 5 May. You earn extra credit for your presentation if you followed your approved plan.
The points to cover are listed at the end of the grading rubric. You earn the indicated points for each topic you cover.
You may use any notes or visual aids that you want, but “talk & chalk” is just fine. Officially there’s a limit of five minutes, but most students cover all their points in less time.
Hang on to your project report until after your presentation, in case you need to refer to it.
Your overall course grade is points earned divided by 592 possible points, as a percentage. Extra credit, including up to 15 for the project presentation, is added to the top number of that fraction. (However, extra credit can’t be used to raise a failing grade.)
One word: planning.
At least a couple of weeks before the project is due on 5 May, come up with a one-sentence idea for your project. Drop me an email or discuss it with me in person to make sure it’s workable.
Then, let me have your completed Project Plan. That covers the major points of the assignment, so when you have an approved Project Plan you’ll know you’re on the right track. Of course, you can always ask me specific questions at any time.
Please see the Project Plan Questionnaire below. Either print that page, fill in the answers, and hand it to me, or (preferably) email me the seven numbered answers — just the answers, not the questions.
It’s not due, because it’s not a graded assignment. But there’s a deadline — if you intend to present your project to the class for extra credit, you must have had your plan approved by the start of class on 28 Apr.
Think of it as a negotiation. Those almost always go in several rounds, as both parties adjust their expectations till there’s a meeting of the minds. But it’s not an instant process.
To avoid misunderstandings, approvals are all in email or in writing. Your Plan is approved when you have my signature on your questionnaire, or an email from me saying “I approve your Project Plan.”
Well, there’s no direct penalty, but you’re closed out of earning extra credit, and you’re at risk of making a major mistake, like using the wrong case for hypotheses and calculations, or even picking a project topic that’s not acceptable.
Whether you have a project plan or not, you’re free to ask me questions in email or in person.
Caution: The Project Plan isn’t required, but if you do a survey then you must get your survey form approved before you start taking your survey.
Instructions: Print out this form and write in your answers, write just your numbered answers on separate paper, or email me just the numbered answers (in plain text, not an attachment).
Deadline: See When is the Project Plan due, above.