Chapter 1
A First Look at Statistics
and Data Collection
CHAPTER OVERVIEW AND OBJECTIVES
The purpose of this chapter is to establish the frame of reference for the study of statistics. It introduces the student to many of the key terms used throughout the text as well as introducing them to types of data, data sources and methods of data collection. At the completion of this chapter, the student should be able to answer the following questions:
1.What is "statistics" and why is the study of statistics important to a business manager?
2.What is meant by the term "descriptive statistics"? "inferential statistics"?
3.What is a population? A sample? A census?
4.What are the types of data? (discrete/continuous)
5.What are the strengths of data? (nominal, ordinal, interval, ratio)
6.What distinguishes qualitative data from quantitative data?
7.When should one use primary/secondary data as a data source?
8.What is a random sample and when is it appropriate to use a convenience sample?
9.What should one keep in mind when designing and coding a questionnaire?
1
Chapter 1 1
Chapter 1 Glossary
census. A sample that contains the entire population, such as the sample obtained by performing 100% inspection of an incoming batch of material.
closed ended question. A question with a pre-determined number of possible responses.
continuous data. Data that can assume any value over some continuous range.
convenience sample. A sample in which the individuals are selected for their ready availability and presumed resemblance to the population of interest.
data coding. Preparing questionnaire data for computer input.
descriptive statistics. The process of collecting and describing sample data.
discrete data. Data that have limited, specific possible values, characterized by gaps in the possible values.
designed experiment. A source of primary data that examines the effect of changes to a process.
inferential statistics. The process of drawing conclusions about a population based on the results of a statistical sample.
Chapter 1 1
interval data. Data for which both the order of the data and the difference between any two data values is meaningful. However, the value 0 (zero) is not indicative of the absence of whatever is being measured; that is, there is no "zero point".
judgment sample. A sample deliberately selected because of their supposed knowledge of the population.
Likert scale question. A question containing scaled (ranked) responses, generally ranging from "strongly disagree" to "strongly agree."
nominal data. Categorical data that may be numeric codes. This coding is not necessary, since nominal data can also consist of words (such as, MALE and FEMALE).
nonrandom samples. Samples selected in a convenient or deliberate manner with little or no attention paid to randomization.
open ended question. A question requiring a written response.
ordinal data. Data for which the order of the values is meaningful, but not the distance between them.
parameter. A value that describes a population, such as the average height or the percentage of defective components.
population. The set of all possible measurements that are of interest.
primary data. Data from an original (primary) source designed to gain insight into a specific research question.
quota sample. A sample deliberately selected to be a miniature of the population.
ratio data. Data having all the properties of interval data and also containing a definite zero point.
Chapter 1 1
sample. The portion of the population selected for observation.
sample size. The number of items in a sample, generally written as "n".
SASE. Self addressed stamped envelope.
secondary data. Previously recorded data, generally collected as part of a regularly scheduled data collection procedure (such as an annual government report or annual company financial statement).
simple random sample. A sample for which every sample of size n has the same chance of being selected.
statistic. A value that describes a sample, such as the average of 200 sample heights or the percentage of defective components in a sample of 500.
qualitative data. Nominal or ordinal data.
quantitative data. Interval or ratio data.
Chapter 1 1
1.1The following populations are examples of populations that would be of interest to a business manager:
a)Consumers of a certain product.
b)Employees of a company.
c)All electrical components manufactured by a factory.
A business manager would prefer to sample from a population rather than take a census becuase of the cost and time involved.
1.2There is a definite zero point with ratio data.
1.3Inferential statistics
1.4Let the first format be the total number of hours of sleep per night. Let the second format be the 4 boxes.
a)Answers to the first format would be ratio data. Answer to the second format would be ratio data.
b)The second format would be easier for the respondent to answer.
c)The first format would yield more accurate results.
d)The second format would produce a higher response rate.
1.5a)Quantitative, ratio
b)Qualitative, ratio
c)Qualitative, nominal
d)Quantitative, interval
e)Qualitative, ratio
1.6a)Used import automobiles in Japan.
b)A numerical fact is a number of interest about a population. A statistic is an estimate of a parameter of a population. The number 24% is a statistic.
c)Ratio data.
d)Checking every imported car in Japan would be very costly and time consuming. Thus a sample is more practical.
1.7a)Secondary
b)Primary
c)Secondary
d)Primary
1.8An experiment could be designed so that random samples are selected from each shift. Productivity would be measured after every random observation was selected.
1.9a)Primary
b)Secondary
c)Primary
d)Secondary
1.10This sample was obtained from a public opinion poll. Usually, polls use a procedure in which observations are randomly selected. A telephone poll would be an excellent way to obtain a random sample for this poll.
1.11a)To use Table A.13, do the following:
- Start at an arbitrary position.
- Select twenty five-digit numbers by reading either across or down the table.
- For each five-digit number selected, place a decimal between the third and fourth digits and round to the nearest integer. If any numbers are over 500, then continue to select random numbers until twenty numbers are chosen with values between 1 and 500.
b)Generating random numbers with Excel or Minitab is easier, particularly if one wants to generate lots of observations.
1.12Closed-ended questions provide data that are easy to quantify.
1.13Give 5 categories of income and ask the respondent to which category does he/she belong.
1.14Pretesting is important to identify possible problems with the questionnaire, such as time required to complete the questionnaire.
1.15Some comments motivating a person to fill out the questionnaire should be included. A closing statement may thank respondents for their time.
1.16It is easier to edit the values and detect unusual values.
1.17a)This is a leading question. Simply ask, "Do you believe teachers' salaries should be raised?"
b)Ask the question about the "Nightly Business Report" after the respondent lists his/her favorite television programs.
c)It is not clear whether Alarge@ refers to population or area.
Chapter 1 1
1.18a)Convenience sample
b)Examples of qualitative questions:
What training at Michigan Medical School did you find most useful?
What do you enjoy the most about the medical profession?
Examples of quantitative questions:
How many hours do you work per week on the average?
How many years do you intend to continue as a full-time physician?
c)Perhaps one incentive could be the promise of a donation to the physician's specified charity.
1.19a)All citizens of the United States.
b)Ask the general questions first and then the specific questions.
c)The question is a leading question. Simply ask, "Do you think that it is in the best interest of the United States to have Congress pass the trade pact?@
1.20Possible errors:
A value of 60 for number of years employed.
A value of 8 for marital status.
A value of 15 for number of children.
A value of $9,000 for monthly payments for someone with a salary of $34,390.
1.21a)The firms that buy computer chips from the company.
b)A table of random numbers can be used. Also Excel or Minitab can be used to generate random numbers.
c)Satisfaction by firms that buy from the company.
d)The average of the responses can be used as the statistic to estimate the parameter in part (c).
1.22a)Primary
b)Secondary
c)Primary
d)Secondary
1.23a)Japanese companies doing business in China.
b)Random numbers could be used to randomly select a company.
c)A convenience sample could be taken from Japanese companies that are easy to contact. Statistical analyses require that a sample be random.
1.24a)A table showing the results of the survey would be a useful way to display the data. Percentages can be used within the table.
b)Yes, inferential statistics could be used since estimates of parameters of interest can be used in the decision making process.
1.25a)Represents a sample of employees at General Motors, if randomly selected from the population of all employees at General Motors.
b)Represents a sample of students on the university campus.
Chapter 1 1
c)Represents a sample of people listed in the telephone book.
d)Represents the population of all possible ways of choosing two cards from a deck of 52 cards.
e)Represents a sample of electronic parts that are manufactured.
1.26A convenience sample. The respondents are not being selected at random.
1.27 Level of
Type of DataMeasurement
Own HomeQualitativeNominal
Number of promotionsQuantitativeRatio
Medical expensesQuantitativeRatio
Enjoy current jobQualitativeNominal
Expect a promotionQualitativeNominal
Date of birthQuantitativeInterval
Annual salaryQuantitativeRatio
Years marriedQuantitativeRatio
1.28a)Quantitative, ratio
b)Qualitative, ordinal
c)Quantitative, ratio
d)Qualitative, nominal
e)Quantitative, ratio
f)Qualitative, nominal
1.29a)Working class
b)Random numbers could be generated from a statistical software package.
c)1.What is the average amount that you save on a monthly basis?
- Do you feel that your savings are adequate for retirement purposes?
- How much money do you plan to have at retirement?
- How many years will it take to reach your goal of having adequate retirement funds?
- At what age do you plan to retire?
d)Including a ticket to a lottery drawing would be an incentive.
e)A second mailing or a follow-up phone call reminding the participant to fill out the survey would increase the response rate.
1.30Answers will vary since data are being generated randomly.