S. Graves - Geography

Quiz: Chapter 15

Spatial Statistics and Analysis

This is a challenging chapter, especially toward the end. Try not to get too bogged down in the details regarding statistics. This is not a stats course, but I STRONGLY URGE you to have a solid grounding in basic statistical procedures. Many students are scared to death of stats, but take it from a classic landscape geographer, it's not all that bad. The fear of math and statistics is always worse than the reality. Statistics rarely involve anything other than combinations of adding, subtracting multiplying and dividing...so you can do it.

GIS is somewhat useless without spatial statistics and spatial analysis.

Instructions:

Fill in the letter or answer that best answers the questions below. Use TAB or your mouse to advance your cursor into response boxes.

Do not press the Enter key until you are ready to submit your answers. If you do, click the Back button in your browser window. When you are satisfied with your answers, press the Enter key or click on the "Submit" button below.

Make sure you enter your name (first name first and last name last) , code and email address in the appropriate data entry boxes below before you submit your answers.

I encourage students to keep a dated copy of their answers in case there is a problem with the submission of your quiz.

Quiz Table
Answer	Letter	Questions and Answer Options
Answer 1	1.	Which of the following is an adequate definition of sampling?
	a.	gathering data from a small subgroup of a larger group
	b.	testing whether or not a person is in a subgroup of a larger group
	c.	collecting all the information from a data set that one can possibly collect
	d.	trying selected study sites before a field site is chosen
Answer 2	2.	Walter Lipman was once quoted as saying "The tendency of the casual mind is to pick out or stumble on a sample which supports or defines its prejudices, and then make it the representative of a whole class". This is a caution against:
	a.	picking out a sample that is not representative of the whole
	b.	finding a sample that suits your particular needs as a researcher
	c.	finding a sample that undermines the argument made by others
	d.	all of these.
Answer 3	3.	What is the main weakness of a case study, not found in statistical testing?
	a.	case studies are not in-depth enough
	b.	case studies are too difficult to do in terms of time commitments
	c.	case studies results are hard to generalize
	d.	all of these are problems with case studies.
Answer 4	4.	An accurate sample:
	a.	is not much different than the overall population.
	b.	does not differ from the population in a systematic fashion.
	c.	will be much like the next accurate sample, if the sampling procedure is repeated several times over.
	d.	all of these are characteristics of accurate samples.
Answer 5	5.	Which of these IS NOT a main concern of sampling design?
	a.	The reason why you are doing the research in the first place
	b.	Finding a sample that confirms your hypothesis
	c.	The time and money available to you.
	d.	The ability of the researcher to do the research
Answer 6	6	I was asked to extract from a roll of voters in a county a sample of prospective voters. I needed to draw from the population a sample that was 80% white, 15% black and 5% "other". My employer though cautioned me that I had to over sample blacks by 100% because they were twice as likely not to answer the phone. What kind of sample did I need to draw?
	a.	simple random
	b.	systematic
	c.	stratified
	d.	proportionate stratified.
Answer 7	7.	In this same task, I was asked not to draw phone numbers from the same household, in order to avoid having pollsters call the same house twice. Which of these sampling strategies would avoid this problem ONCE I SORTED THE HOUSEHOLDS BY PHONE NUMBER?
	a.	simple random
	b.	systematic
	c.	stratified
	d.	proportionate stratified.
Answer 8	8	One of the special considerations for Geographers is spatial autocorrelation. What is it?
	a.	When things that are next to each other are too much alike.
	b.	When things that are next to each other are too different.
	c.	When things next to each other have no relationship to one another.
	d.	When observations of data that are next to each other tend to be affected by each other, and therefore not truly random.
Answer 9	9.	If I were to take a sample of a city's population from one neighhborhood and find that this neighborhood is differerent than the known total population (in terms of say, income) and found a z-score of -1.96, then I could say that:
	a.	I'm 95% sure that this neighborhood is statistically similar to the rest of the city.
	b.	I'm 5% sure that this neighborhoods is wealthier than the rest of the city.
	c.	I'm 95% sure that this neighborhood is not average in terms of income.
	d.	I'm 95% sure that this neighborhood is poorer than any other in the city.

10. Take a look over the following paragraph in italics. It was produced by a public relations firm that was charged with trying to counter the findings of a recent research project that found many payday lenders located their businesses with a few miles of military bases. It's a classic example of how a neutral looking sample, may in fact be biased by its very design.

Find three examples of potential bias in the way the sample was collected and write them in the response box below. There are at least 10 ten sources of bias.

X, Y & Z Associates developed and conducted a survey to determine the incidence of payday loan usage among active duty members of the military.

A nationally representative sample of 1,002 active duty members of the United States Armed Forces, across all branches of service, was interviewed between January 11-20, 2005. The list was purchased from Equifax, a credit monitoring company. Calls were made from PSB’s telephone call center, located in Denver, Colorado.

During the interviews, respondents were asked to confirm they were indeed active duty members of the armed forces. They were then asked a series of screening questions to determine which branch of the military they served in and whether they had taken out a payday loan in the last five years.

Of the 1002 members of the military surveyed, only 37 active military service people said they had taken out a payday advance loan in the last five years. This is an incidence of 3.69% usage. Among that sub sample, only 32% had a current loan outstanding, an overall incidence rate of 1.18%. Based on the current 1.4 million active duty military population, in real figures, one could make an educated projection that 51,660 members of the military have taken payday loans in the last five years and 16,520 currently have a payday loan outstanding. The margin of error with this sample size is +/-3.1% at the 95% confidence level.

While some media reports have estimated that usage of payday advance loans among military personnel could be as high as 30% to 40%, PSB’s polling has found that figure to be radically overstated. Clearly, only a small segment of the military has used the service in the last five years.

Answer 10

Name

E-mail

Code

When you click the button below, you will be directed to a web page that shows your answers. The instructor will get a copy of this email as well, but you may want to keep a copy for your records. If you are curious about the correct answers, please bring your questions to class.

Geography 300

Quiz: Chapter 15

Spatial Statistics and Analysis

Instructions:

Get Adobe Reader