Deck 16: Generalizing a Sample's Findings to its Population and Testing Hypotheses About Percents and Means

Question 1

Ping Golf, Inc. conducts a study in which they use a probability sample of 600 golfers. They examine the data from the sample and calculate that 28 percent of them own Ping golf clubs. The 28 percent is:

A) the average
B) the sample statistic
C) the grand mean
D) the population parameter
E) the grand percentage

Answer 1

B

Question 2

Roman letters, such as s, p, and b, are typically used to represent:

A) sample statistics
B) Greek parameters
C) population parameters
D) Roman statistics
E) population statistics

Answer 2

A

Question 3

The proper command sequence in SPSS to generate a confidence interval for a mean is:

A) ANALYZE; COMPARE MEANS; ONE-SAMPLE t TEST
B) ANALYZE; COMPARE MEANS; CONFIDENCE INTERVAL; MEAN
C) ANALYZE; COMPARE MEANS; CONFIDENCE INTERVAL
D) ANALYZE; CONFIDENCE INTERVAL; COMPARE MEANS
E) ANALYZE; CONFIDENCE INTERVAL; MEAN

Answer 3

A

Question 4

The proper command sequence in SPSS to generate a confidence interval for a percentage is:

A) ANALYZE; COMPARE PERCENT; ONE-SAMPLE t TEST
B) ANALYZE; COMPARE MEANS; CONFIDENCE INTERVAL; PERCENT
C) ANALYZE; COMPARE PERCENT; CONFIDENCE INTERVAL
D) ANALYZE; COMPARE PERCENT; ESTIMATE
E) None of the above: SPSS does not perform percent estimation.

Question 5

A 99 percent confidence interval allows us to say that if we took many, many samples from the population:

A) all of the samples would contain a range that included the population parameter
B) 99 percent of the samples would contain a range that included the population parameter
C) 99 percent of the samples would allow us to estimate the mean within + or - 1 percent
D) 1 percent of the samples would contain a range that included the population parameter
E) 1 percent of the samples would allow us to estimate the mean with + or - 1 percent

Question 6

Which of the following refers to a measure of the variability in the sampling distribution based on what is theoretically believed to occur if we were to take a multitude of independent samples from the same population?

A) multiple variance analysis
B) standard variance
C) standard error
D) confidence error
E) independent error

Question 7

Values computed from information provided by a sample are referred to as:

A) Greek parameters
B) Roman parameters
C) statistics
D) parameters
E) sample parameters

Question 8

________ is the process of using sample information to compute an interval that describes the range of a parameter such as a population mean or the population percentage.

A) Sample estimation
B) Hypothesis testing
C) Parameter estimation
D) Confidence analysis
E) Regression analysis

Question 9

Researchers use ________ to estimate the population parameter within a range because there is always sampling error.

A) range parameters
B) confidence intervals
C) standard percentages
D) regressive parameters
E) parametric intervals

Question 10

Values that are computed from a complete census and are considered precise and valid measures of the population are called:

A) Greek parameters
B) Roman parameters
C) statistics
D) parameters
E) population statistics

Question 11

Statisticians make estimates of population parameters that may be either:

A) means or percentages
B) dollars or cents
C) positive or negative
D) subjective or negative
E) subjective or objective

Question 12

When referring to Greek parameters of the sample data, statisticians use:

A) mu
B) sigma
C) beta
D) phi
E) None of the above; there is no such thing as Greek parameters of sample data.

Question 13

If you have five friends who tell you they all have had a great experience with their purchase of a Chevrolet, and you used this evidence to decide to buy a Chevrolet, you would be using:

A) consumer self-report and evaluation
B) statistics
C) inference
D) independent judgment
E) casual judgment

Question 14

Which of the following refers to a set of procedures in which the sample size and sample statistics are used to make estimates of population parameters?

A) sample logic
B) inference
C) statistical inference
D) estimate inferences
E) population logic

Question 15

All other variables held constant, a 95 percent confidence interval will give us a ________ range, as opposed to a 99 percent confidence interval.

A) wider
B) narrower
C) the same
D) more encompassing
E) The range depends on the sample size.

Question 16

Two types of statistical inference presented in Chapter 16 of your textbook are:

A) parameter estimates and regression
B) hypothesis tests and regression
C) parameter estimates and hypothesis tests
D) parameter estimation and interval tests
E) regression and interval tests

Question 17

When computing standard error, if the sample size increases, then the standard error:

A) decreases
B) increases
C) remains the same
D) becomes null
E) Sample size has nothing to do with standard error.

Question 18

Which of the following refers to a form of logic in which you make generalizations about an entire class based upon what you have observed about a small set of members of that class?

A) sample logic
B) fuzzy logic
C) Greek logic
D) inference
E) logistics

Question 19

If a researcher wishes to have a 95 percent confidence interval, he/she should use a standard error value of:

A) 95
B) 95 percent
C) 1.96
D) 2.58
E) 1.52

Question 20

When computing standard error, if the variance (standard deviation)increases, then the standard error:

A) decreases
B) increases
C) remains the same
D) becomes null
E) Variance has nothing to do with standard error.

Question 21

Inference is a form of logic in which you make a generalization about a small set of members of a class based upon complete knowledge about that entire class.

Question 22

In parameter estimation, a value such as the population mean is estimated based on a sample's mean and its size.

Question 23

In hypothesis testing, given the standard error of the percentage (or mean)remains constant, the closer the hypothesized value is to the sample statistic:

A) the smaller the z
B) the larger the z
C) the more likely the z is to approach 100
D) the z nears 0
E) the z nears 2.58

Question 24

Which of the following may be used to estimate a "best," "pessimistic," and "optimistic," value of market potential?

A) mean, median, mode
B) a confidence interval
C) hypothesis test (three)
D) the theoretical market potential estimator
E) intuitive hypothesis tests (three)

Question 25

Statistical inference is a set of procedures in which the population size and sample statistics are used to make estimates of population parameters.

Question 26

Which of the following is a directional hypothesis?

A) The population percentage is equal to 60%.
B) The population percentage is not equal to 60%.
C) The population mean is equal to $50.
D) The population percentage is less than 70%.
E) The population mean is equal to 100.

Question 27

When you use SPSS to test a hypothesis for a mean, you must do which of the following?

A) run a DESCRIPTIVES program first
B) specify a test value
C) calculate the standard error of the hypothesis
D) plot the data points
E) run a scatterplot

Question 28

Greek letters are used for sample statistics and Roman letters are used for parameters.

Question 29

Which of the following is NOT used to test hypotheses?

A) sample statistic (mean or percentage)
B) Roman parameter
C) standard error of the statistic (mean or percentage)
D) hypothesized parameter value
E) desired level of confidence

Question 30

If the SPSS output for the test of the hypothesis about a mean shows a significance of 0.000, this means:

A) SPSS failed to run the test
B) the significance level shows very little support for the hypothesis
C) the significance level shows much support for the hypothesis
D) more data is needed
E) the wrong test statistic was chosen.

Question 31

In hypothesis testing, the alternative hypothesis:

A) is always implicit
B) is sometimes stated
C) will be supported if the stated hypothesis is not supported
D) will not be supported if the stated hypothesis is supported
E) All of the above are true.

Question 32

The SPSS output for the test of the hypothesis about a mean includes which of the following?

A) the N (which is the sample size)
B) mean
C) standard deviation
D) standard error of the mean
E) all of the above

Question 33

To use SPSS to test a hypothesis for a percentage you would use:

A) ANALYZE; TEST HYPOTHESIS; PERCENTAGE
B) ANALYZE; TRANSFORM; HYPOTHESIS
C) ANALYZE; CONFIDENCE LEVEL; HYPOTHESIS; PERCENTAGE
D) ANALYZE; CONFIDENCE LEVEL; PERCENTAGE
E) None of the above; SPSS does not test hypotheses for percentages.

Question 34

Values that are computed from a complete census, which are considered to be precise and valid measures of the population, are referred to as parameters.

Question 35

If we make a hypothesis and wish to be 95 percent confident in testing the hypothesis, a z score that supports the hypothesis will fall ________ of + or - 1.96:

A) inside
B) outside
C) near, but not inside
D) near, but not alongside
E) within some multiple

Question 36

Which of the following refers to using sample data to accept or reject some exact specification as to what the population parameter is?

A) confidence analysis
B) standard error analysis
C) hypothesis testing
D) specification support/rejection
E) theory testing

Question 37

If we make a greater than directional hypothesis and wish to be 95 percent confident in testing the hypothesis, a z score that supports the hypothesis will be a value ________ +1.64:

A) inside
B) outside
C) greater than
D) less than
E) of 5 percent less than

Question 38

A toy store owner hypothesizes that parents spend less than $100 on toys per visit to her store. This is:

A) a hypothesis
B) a directional hypothesis
C) a negative hypothesis
D) a positive hypothesis
E) an alternative hypothesis

Question 39

Which of the following occurs when someone uses something he or she has observed to see if it agrees with or refutes a belief about that topic?

A) intuitive hypothesis testing
B) statistical hypothesis testing
C) the alternative hypothesis
D) common hypothesis testing
E) hypothetical observation testing

Question 40

An important sample statistic used in making estimates about population parameters is sample size.

Question 41

SPSS will not calculate the confidence interval for a percentage.

Question 42

Parameter estimation is used to compare the sample statistic with what is believed to be the population value prior to undertaking the study.

Question 43

To calculate the mean and to generate a confidence interval estimate of the population mean, the sequence of menu commands in the SPSS for Windows is ANALYZE; COMPARE MEANS; ONE SAMPLE t TEST.

Question 44

If you were investigating the percentage of high school students who eat at the local Taco Bell, the percentage, because it is derived from the sample, is the sample statistic.

Question 45

The standard error will increase if sample size is increased.

Question 46

SPSS will not calculate the confidence interval for a mean.

Question 47

If you were investigating the average number of golf balls purchased by golfers in a month, your sample statistic would be a mean.

Question 48

To estimate a parameter, you need to know the sample statistic, which is either a mean or percentage derived from sample data.

Question 49

Detailed confidence intervals are typically reported to clients.

Question 50

Intuitive hypothesis testing is quite complicated and uses linear algebra to test mathematically intuitive algorithms.

Question 51

The z value associated with a 99 percent confidence interval is either a + or - 2.58.

Question 52

Two types of statistical inference are parameter estimates and hypothesis tests.

Question 53

The standard error will increase if the variation (standard deviation)is increased.

Question 54

There are five steps involved in computing confidence intervals for a mean or a percentage. One of those steps is to determine the sample statistic.

Question 55

One option that researchers have when it comes to reporting confidence intervals to clients or readers
of their marketing research reports is called the General Case.

Question 56

When an exact specification of a population parameter is made based on prior knowledge, assumptions, or intuition, this is a hypothesis.

Question 57

A practical use of confidence intervals is that they may be used to forecast likely, pessimistic, and optimistic forecasts.

Question 58

The z value associated with a 95 percent confidence interval is either a + or - 1.96.

Question 59

When the sample size is the same and variability is the same, a 99 percent confidence interval will be narrower than a 95 percent confidence interval.

Question 60

Hypothesis testing is a statistical procedure used to accept or reject the hypothesis, based on sample information.

Question 61

If we are testing the hypothesis that a population mean is equal to 50 percent, then the alternative hypothesis must be that the population mean is not equal to 50 percent.

Question 62

Suppose we wish to test the hypothesis that an internship program allows its interns to earn $2,750 per semester and let us assume that this hypothesis is, in fact, true. Which of the following best illustrates the logic of hypothesis testing at the 99% confidence level?

A) From among 100 samples, 99 percent of them would generate means that fall outside + or - 2.58 z scores.
B) From among 99 samples, 100 percent of them would generate means that fall within + or - 2.58 z scores.
C) From among 100 samples, 1 percent of them would generate means that fall outside + or - 2.58 z scores.
D) From among 100 samples, 1 percent of them would generate means that fall outside + or - 1.96 z scores.
E) From among 95 samples, 1 percent of them would generate means that fall outside + or - 1.96 z scores.

Question 63

Laurie Fulkerson was in charge of marketing research for Jeff Rogers, a candidate for governor. No research had been undertaken as it was very early in the campaign. The consensus was that Jeff and his opponent, Elvis Dean, were very close in voter preference. Laurie was negotiating with the research firm that was going to do a survey. She told the project director that she wanted to estimate the percentage of registered voters who would vote for each candidate if the election were to be held at the time of the survey. She also said she wanted to be very confident (99 percent)that the estimate was very accurate.between + or - 1 percent. The project director quickly estimated that the sample size would be very large and told Laurie that she estimated the study would cost more than $200,000. Shocked at the price, Laurie asked the project director to cut the sample size down to one-fourth of what she estimated. What should the project director tell Laurie?

A) that it is impossible to do an accurate study with a sample size this small
B) that the sample size can be cut, but it means that the confidence interval that will be calculated around the estimate will be much smaller - far smaller than + or - 1 percent
C) that the sample size can be cut, but it means that the confidence interval that will be calculated around the estimate will be much larger but still under the + or - 1 percent
D) that the sample size can be cut, but it means that the confidence interval that will be calculated around the estimate will be much larger - far greater than the + or - 1 percent
E) that it is fine to cut the sample size and the only impact will be reducing the cost of the study

Question 64

Suppose we wish to test the hypothesis that an internship program allows its interns to earn $2,750 per semester and let us assume that this hypothesis is, in fact, true. Which of the following best illustrates the logic of hypothesis testing at the 95% confidence level?

A) From among 100 samples, 95 percent of them would generate means that fell within + or - 1.96 z scores.
B) From among 99 samples, 100 percent of them would generate means that fell within + or - 2.58 z scores.
C) From among 100 samples, 95 percent of them would generate means that fell within + or - 2.58 z scores.
D) From among 100 samples, 99 percent of them would generate means that fell within + or - 1.96 z scores.
E) From among 95 samples, 100 percent of them would generate means that fell within + or - 1.96 z scores.

Question 65

The alternative hypothesis refers to doing just about anything to make a decision other than hypothesis testing.

Question 66

A toy store owner hypothesizes, at the 95 percent level of confidence, that parents spend less than $100 on toys per visit to her store. A sample is taken and the hypothesis test shows a z value of -1.65. We:

A) support the hypothesis
B) do not support the hypothesis
C) support the alternative hypothesis
D) immediately take another sample
E) try to take another sample in another toy store

Question 67

If we are testing a hypothesis that a sample mean does not differ from a hypothesized value at the 99 percent level of confidence, and we arrive at a z value of 2.59, we would fail to support the hypothesis.

Question 68

Directional hypotheses contain statements of "more than" or "less than."

Question 69

Failure to support the hypothesis essentially tells the hypothesizer that the assumptions about the population are in error and that they must be revised in light of the evidence from the sample.

Question 70

A hypothesis test gives you the probability of support for your hypothesis based on our sample evidence and sample size.

Question 71

When we use SPSS to test a hypothesis for a percentage, we use the F value instead of the z value.

Question 72

Forming a directional hypothesis means that we are still seeking a direction for the hypothesis.

Question 73

You test a directional hypothesis that consumers spend less than $12 a person to go to the movies. Your z value turns out to be -1.64. Your data supports your hypothesis.

Question 74

John Daniel is the brand manager for Healthy-Pick Cookies, Inc. John had been through the new product development process and has developed a new cookie that has very good taste and texture, yet has no calories, fat, sodium, or cholesterol. Concept tests on the product have been excellent and now John is ready to test market the new cookie and has selected Phoenix, Arizona, as his first test city. But before he starts the test market, John wants to conduct a research project that will help him forecast sales, so that he can better prepare production to supply the test market. The project leads to a probability sample in which household members are randomly called and are given a thorough description of the new cookie, including the price and choices of flavors. The key question respondents are asked is how likely it is that they will actually buy the new cookie, and this is measured on a 7-point intensity continuum scale ranging from 1 being Very Unlikely to 7 being Very Likely. Respondents were also asked how many boxes of cookies they would expect to buy in a month. At the end of the study, the researchers tell John that 8 percent of the households contacted stated that they were Very Likely to buy the new cookie. In order to get an estimate of the sales potential in the test market John could:

A) take 8 percent of all the households in Phoenix and multiply this times the population parameter
B) use the 8 percent as the "best estimate" times the number of households in the market and multiply this number by the average number of boxes expected to be purchased
C) use the 8 percent as the "best estimate" times the number of households in the market and divide this number by the average number of boxes expected to be purchased
D) use the 8 percent as the "best estimate" times the number of households in the market and multiply this number by the average number of boxes expected to be purchased. In addition to this use, the upper and lower limits of a hypothesis test to calculate a pessimistic and optimistic estimate respectively
E) use the 8 percent as the "best estimate" times the number of households in the market and multiply this number by the average number of boxes expected to be purchased. In addition to this use, the upper and lower limits of a confidence interval around the "best estimate" to calculate optimistic and pessimistic estimates respectively

Question 75

The logic of hypothesis testing is that if we assume that the hypothesized value is the population parameter, then, at the 95 percent level of confidence, if we took 100 separate samples, the z value for 95 of those samples would fall between + or -1.96.