Transcript
Chapter 2: Data Collection
Chapter Objectives
When you finish this chapter you should be able to
use basic terminology for describing data and samples.
explain the difference between numerical and attribute data.
recognize levels of measurement in data and ways of coding data.
Recognize a Likert scale and know how to use it.
explain the difference between time series and cross-sectional data.
use the correct terminology for samples and populations.
explain common sampling methods and how to use them.
find everyday print or electronic data sources.
describe basic elements of survey design, survey types and sources of error.
Quiz Yourself
True/False Questions
In a sample of 300 students in the College of Arts and Sciences, 51 of them, or 17% are Economics majors. The 17% is an example of statistical inference as opposed to descriptive statistics.
Estimating characteristics of the population is the main goal of descriptive statistics.
Interval data may be treated as ordinal or nominal.
Ordinal data may be treated as interval but not as nominal.
Multiple Choice Questions
Which of the following is most likely a parameter as opposed to a statistic?
the average score of the first five students completing an assignment.
the proportion of females registered to vote in a county.
the average height of people randomly selected from a data base.
None. They are all statistics.
Use this set of responses to indicate the level or type of data described in the following FOUR phrases or statements. Responses may be used more than once.
A = Nominal
B = Ordinal
C = Discrete
D = Continuous
The numbers members of sports teams wear on the back of their jerseys.
Responses to the question: Do you now earn more or less than you did five years ago?
Unemployment is 5.5% of the population.
Shoe sizes such as 7B, 10D and 12EEE
To monitor security, IU Police are surveying the number of students in the Atwater Parking Garage each 30 minutes for a 24-hour period, to help determine when patrols would be most helpful to students. The number of students in the lot is an example of
a categorical random variable.
a continuous random variable.
a discrete random variable.
a statistic.
Which of the following represents an ordinal level of measurement or data?
A. Fish, Cats, Dogs
B. 1:00, 2:00, 3:00
C. Nominal, ordinal, interval
D. Basketball Jersey numbers 23, 33, 43
A recent study of breast cancer revealed that 13% of the women in the sample used antibiotics more than 500 days in their lifetime. Further, 79% of these “heavy antibiotics users” developed breast cancer. According to the American Cancer Society, one in twelve women will develop breast cancer at some time in her life. Of the numbers mentioned, which are parameters?
A. 13% and 79%
B. 13% and one in twelve
C. 79% and 500 days
D. 79% and one in twelve
Which of the following statements is true?
A. Complete populations are often inaccessible.
B. We use samples because we want absolute accuracy.
C. We always have plenty of time to do a study.
D. As a population becomes large, it is usually better to obtain statistical information from the entire population.
Sufficiently large coverage error will result in which of the following?
A. Non-response bias.
B. Probability sampling.
C. Statistics about the actual population rather than the target population.
D. Inability to perform inferential statistics.
A survey was mailed to a total of 400 students; 100 were randomly selected from each of the freshman, sophomore, junior and senior classes last semester. What sampling method was used?
A. Cluster sample
B. Stratified sample
C. Systematic sample
D. Simple random sample
A statistician working for the Department of the Interior is constructing a systematic random sample of trees in a forest preserve. All trees have been identified and numbered with a non-invasive number attached to the bark. The statistician wishes to estimate the mean height of the trees in the forest preserve, and has calculated an appropriate sample size of n=43 trees. The 587th of the 731 trees in the preserve was selected as the first observation in the sample. What is the tree number of the second observation in the sample?
A. 544
B. 588
C. 604
D/ 630
A poll is planned to determine what proportion of all IU students favor an increase in fees to support a new track and field stadium. A questionnaire will be published in the IDS and the first 1000 completed questionnaires will be analyzed. Which of the following errors will be associated with the sample?
A. Sampling error and error due to selection bias.
B. Errors due to interviewer bias and selection bias.
C. Only non-sampling error.
D. Sampling error only.
What is the difference between probability sampling and simple random sampling?
A. There is no difference.
B. Probability sampling is a type of simple random sample.
C. Simple random sampling is a type of probability sampling.
D. In probability sampling each item has an identical chance of being chosen.
What is the difference between probability sampling and simple random sampling?
A. There is no difference.
B. Probability sampling is a type of simple random sample.
C. Simple random sampling is a type of probability sampling.
D. In probability sampling each item has an identical chance of being chosen.
Solved Problems from Text
2.4 a. continuous numerical –measures time which is continuous
b. discrete numerical—a count of whole peanuts
c. attribute—a description of what the applicant does for a living, a quality
d. continuous numerical—measures time
e. attribute—naming is identifying a characteristic
d. discrete numerical—the implication is that sizes are whole, no half sizes (which is probably untrue) but it is true that gaps exist between the sizes regardless of whether they are whole or half.
2.8 a. ratio—zero represents the complete absence of employees and is therefore meaningful
b. ratio—as a, but zero means complete lack of returns.
c. interval—intervals have important meaning and zero is just another point on the scale.
d. nominal—name is what nominal means.
e. nominal—simply names the month
f. nominal—purpose is simply for identification, it has no numerical meaning.
2.12 a. time series—a value from a single variable recorded over equally spaced units of time, ie days.
b. cross-sectional—values from 500 variables, collected at a single point in time.
c. time series-- a value from a single variable recorded over equally spaced units of time, ie quarters
d. cross-sectional-- values from 10 variables, collected at a single point in time.
2.16 a. parameter or statistic, depending on how the population is defined
b. parameter or statistic, depending on how the population is defined
c. statistic, unless the 50 investors were all the investors of interest.
d. statistic, unless the 50 brokers were all the brokers of interest.
2.24 a. Response bias, if questioning is public answers may be inflated
b. Self-selection bias, coverage error, because you are asking only those persons who are going to church, a large untapped segment of the population is missing..
c. coverage error, self-selection bias, because you are asking only McDonald’s customers and ignoring person who do not eat at McDonald’s at all.
Quiz Yourself Answers
True/False
Multiple Choice
1
F
1
B
6
C
11
B
2
F
2
A
7
C
12
A
3
T
3
B
8
A
13
C
4
F
4
D
9
D
14
D
5
B
10
A
15
A
