Research Methods Test 2
A researcher divides the city of Flint, Michigan, into 500 blocks of 10 households each. He requires a sample of 400 households. He then selects his sample by choosing 100 blocks and 4 households per block. This is an example of a. systematic sampling b. two-stage area sampling c. probability-proportional-to-size sampling d. one-stage area sampling e. sequential sampling
b.
A researcher is considering drawing a sample of size n from a negatively skewed parent population with a mean, g and variance 02; he should realize that n is _, the sample mean x is __ distributed with mean x and variance 0 2In (assuming the sample represents a small proportion of the population). a. small; normally b. large; normally c. small; binominally d. large; binominally e. none of the above
b.
If a population is defined as all pickup truck owners residing in Tippecanoe County. a. asking 100 owners their attitude toward a new truck style would be an example of example of a consensus. b. asking 80 owners their attitude toward a new gun rack design would be an example of a sample. c. asking all owners their attitude toward a new truck style would be an example of universal testing. d. asking only owners listed in the telephone directory would be an example of a sampling frame. e. None of the above are true.
b.
A researcher divides the population of the purchasers of brand X into heavy and light users. He then chooses a random sample independently from each group. This is an example of _ sampling. a. nonprobability b. random c. judgmental d. stratified e. quota
d.
A researcher wants to be able to assess the amount of sampling error associated with an estimate. Which of the following sampling methods would you recommend he use? a. convenience b. quota c. nonprobability d. probability e. judgment
d.
Snowball samples are an example of a. convenience sampling. b. judgement sampling. c. quota sampling. d. simple random sampling. e. stratified sampling.
d.
Which of the following is needed before a confidence interval can be established for a population mean? a. a confidence level b. a point estimate of the population mean c. an estimate of the sampling error associated with the sample mean d. all of the above e. a and b above
d.
Spectrum Inc., an automobile manufacturer, is interested in determining the satisfaction of the users of one of its brands, Credence. Instead of sampling the entire population of car owners, Spectrum Inc. samples those who owned a Credence. Spectrum is drawing its sample from a: A. sampling unit. B. defined target population. C. designated sampling population. D. subset relevant population. E. full-set relevant population.
B. defined target population.
A sampling frame a. is a list of population elements from which the sample will be drawn. b. is the list of population elements actually included in the sample. c. usually provides biased statistics. d. is a form of probability sampling. e. is a form of nonprobability sampling.
a.
All of the following are probability sampling plans EXCEPT a. quota b. area c. cluster (1. disproportionate stratified e. systematic
a.
In cluster sampling, statistical efficiency a. is often traded off for increased economic efficiency. b. is apparent if one samples 50 population elements instead of 20. c. is the same as the overall efficiency of a sampling plan. d. can be offset by observational efiiciency. e. b and d.
a.
In probability samples, the probabilities of each element must be a. known b. unequal c. equal d. known and equal e. known and unequal
a.
In the previous example, the researcher selects his sample by choosing 50 blocks and 8 households per block. The second stage sampling fraction is a. 8/10 b. 1/8 c. 1/10 d. 1/2 e. indeterminate
a.
Systematic sampling is a form of a. cluster sampling. b. stratified sampling. c. quota sampling. d. convenience sampling. e. simple random sampling.
a.
The sampling distribution of a statistic refers to a. the distribution of all possible sample values of the statistic that could be drawn from the parent population under the specified sampling plan. b. the distribution of the variable in the parent population. c. the distribution of the variable in a particular sample. d. the spread of the variable in the parent population. e. the unbiased nature of most sample statistics.
a.
Which of the following is TRUE? a. The controls for a quota sample are usually chosen on the basis that (1) they are believed to be correlated with the characteristic to be studied, and (2) reasonably up-to-date information on their distribution within the universe is available. b. To validate a quota sample, various characteristics of the sample are compared to known universe values. If the differences in these characteristics are small in the comparison, this constitutes proof that the sample corresponds to the universe. c. Quota sampling is the most common kind of probability sampling used in marketing research. d. Quota sampling is the only nonprobability sampling method that provides some measure of the sampling error associated with the sample estimate. 6. All of the above statements are true.
a.
Which of the following statements is FALSE with regard to stratified sampling? a. A stratified random sample is drawn by dividing the parent population into mutually exclusive and exhaustive subsets and by then taking a random sample of these subsets. b. The basic rule by which we stratify suggests the resulting strata should be as homogeneous as possible. c. When sampling from a population that is highly skewed, a stratified sample generally provides the most representative sample. d. A market researcher needs information about parts of the universe, as well as the total universe. He should therefore plan on using a stratified random sample. e. Stratified samples are more statistically efficient than simple random samples.
a.
With proportionate stratified sampling, the number of elements selected from each stratum a. is in direct proportion to the relative number of elements in each stratum in the population. b. is in direct proportion to the relative variability of the characteristic in each stratum in the population. c. varies directly with the heterogeneity of each stratum. d. varies directly with the homogeneity of each stratum. e. is none of the above.
a.
See book
a. o_2 = (o 2/n)[(N-n)/(N-1)]
A __ allows trends in the data to be evaluated as the data are collected. a. nonprobability sample b. sequential sample 0. fixed sample d. probability sample c. all of the above
b.
A complete canvas of a population is called a. a sample. b. a census. c. a sampling frame. d. a universe. e. none of the above.
b.
A market researcher divides the Dallas metropolitan area into blocks having roughly equal populations. He then selects a random sample of blocks and sends interviewers to each block. The interviewers are instructed to interview every eighth dwelling unit. This is __ sampling. a. one-stage cluster b. two-stage cluster c. quota d. two-stage systematic e. stratified
b.
A problem with nonprobability samples is a. they are always more representative than nonprobability samples. b. one cannot assess the amount of sampling error likely to occur. c. there is no problem with using nonprobability samples d. a and b are both true statements. e. none of the above statements are true.
b.
One feature that distinguishes stratified samples from quota samples is that with stratified sampling, elements are selected _, while with quota sampling they are selected a. statistically; deterministically b. probabilistically; judgmentally c. judgmentally; probabilistically d. probabilistically; randomly e. empirically; deterministically
b.
One of the major problems with the mail surveys is nonresponse bias. In an effort to minimize this, a researcher decides to limit the sample to his personal friends from whom he is sure to get replies. This is an example of a a. probability sample. b. nonprobability sample. c. random sample. d. sequential sample. e. quota sample.
b.
Probability-proportional-to-size sampling requires that a. a certain proportion of second-stage units be selected from each first-stage unit. b. a fixed number of second-stage units be selected from each first-stage unit selected. c. a variable ratio of first-stage units be selected. d. a variable proportion of first-stage units be selected. e. none of the above.
b.
The essential difference between a confidence interval constructed employing a stratified sample and that constructed employing a simple random sampling is that with the stratified sample a. a point estimate of the population mean is not required. b. determination of the estimate of the mean and standard error of estimate involves the use of weights. c. the strata results are appropriately weighted by a method of curvilinear estimation. d. inferences are drawn on the basis of the known population mean and variance. e. the researcher does not need to specify the degree of confidence.
b.
The first step that must be executed in drawing a sample is to a. determine the sample size. b. define the population. c. select a sampling procedure. d. select the sample elements. e. identify the sampling frame.
b.
The interesting thing about confidence intervals established employing stratified sampling is that a. the within-strata variability does not enter into the calculation of the standard error of estimate. b. the between-strata variability does not enter into the calculation of the standard error of estimate. c. neither the within-strata variability nor the between-strata variability enters into the calculation of the standard error of estimate. d. the confidence interval is wider than that established employing simple random sampling. e. the sample mean provides a biased estimate of the population mean.
b.
Under which of the following conditions can the population mean be estimated without error employing a stratified sample? a. when the population is partitioned so that the number of elements in each stratum is proportional to size of the sample to be taken from each stratum. b. when the population is partitioned so that the elements in each stratum are equal c. when the population is partitioned in such a way that the variances within strata are equal d. when the population is partitioned so that the elements in each stratum are normally distributed e. none of the above
b.
Which of the following is the best criterion to use to partition owners of sports cars? a. income b. age c. gender d. occupation e. education
b.
You know that consumers with high incomes show a much greater variability in their consumption of a given good than those with low incomes. What type of sample would produce the most efficient estimate of the average consumption per individual? a. simple random sample b. disproportionate stratified sample c. proportionate stratified sample (1. quota sample e. systematic sample
b.
A researcher collects a small sample and finds that he does not have sufficient evidence to draw a conclusion. He then makes additional observations and finds this evidence inconclusive. He continues making observations until a conclusion can be drawn. He is using a. random sampling b. fixed sampling c. sequential sampling d. the principle of diminishing utility e. nonprobability sampling
c.
A(n) _ statistic is one whose __ value fi-om all possible distinguishable samples under the sampling plan is equal to the parameter it estimates. a. biased; median b. efficient; average c. unbiased; mean d. sample; biased e. unbiased; extreme
c.
All of the following are reasons why a researcher might choose a sample over a census for making inferences about population, except a. a census may not be plausible. b. samples are less expensive. c. a census has less nonsampling error. d. a census is less accurate. c. All of the above are reasons to choose a sample over a census.
c.
The distinguishing feature of nonprobability sampling plans is that a. they allow an assessment of sampling error b. they are not as representative of the population as probability samples c. they involve personal judgement somewhere in the selection of sample elements d. there is only one basic type of nonprobability sample while there are several types of probability samples e. they tend to be more expensive than probability samples
c.
When using area sampling, a researcher should seek a balance between statistical efficiency which suggests __ and economic considerations which suggest a. the need for accuracy; that benefits he considered b. a small number of areas be used; few areas be used c. a large number of areas he used; few areas he used d. that a small sample size can be used to predict accurately; possible losses due to faulty predictions can be costly e. that a large number of areas be used; a large number of observations within each area be collected
c.
Which of the following is NOT a nonprobability sample? a. quota b. convenience c. cluster d. judgment e. All of the above are nonprobability samples.
c.
Which of the following is true? a. When sampling from a population that is highly skewed, a cluster sample generally provides the most representative sample. b. Stratified sampling is a very useful sampling technique. It allows closer examination of the characteristics of particular subgroups. It also lowers the chance of error by ensuring that subgroups are adequately represented in the sample. However, this method generally produces less precise estimates of population values than simple random samples. c. In calculating the standard error of the mean when using a stratified sample, one finds that differences among strata means do not enter the estimate. d. When using proportional allocation in stratified random sampling, one samples from each of the strata in proportion to their respective variabilities. e. All of the above statements are true.
c.
Which of the following statements is correct? a. A low incidence means that a high percentage of the population of interest qualifies for inclusion in the study. b. A low incidence means that a low percentage of the population of interest is expected to respond to the study. c. A low incidence means that a low percentage of the population of interest qualifies for inclusion in the study. d. A low incidence means that a low percentage of the target population will participate in the study. e. None of the above are true statements.
c.
Which of the following statements is true? a. In cluster sampling, the parent population is broken into mutually exclusive and exhaustive subgroups and a simple random sample is selected from each subgroup. b. Cluster samples are most statistically efficient when the clusters are internally homogeneous. c. In order to yield high statistical efficiency, each cluster in a cluster sample should include as many diverse universe values as possible. d. One of the main advantages of cluster sampling over simple random sampling is its greater statistical efficiency for the same size sample. c. All of the above statements are false.
c.
A researcher wants to study the effects of social class on consumption behavior. He establishes three different categories of social class in terms of amount of income. He then assigns each field worker a specified number of interviews with people in each income category although the interviewers are allowed to select whom they interview. This is an example of a. random sampling. b. systematic sampling. c. area sampling. d. quota sampling. e. probability sampling.
d.
A researcher, in attempting to sample the population of Cleveland, decides to select five households for each block designated for inclusion in the sample. The blocks themselves are to be selected with variable probabilities depending on their size. This is an example of a. one-stage area sampling. b. quota sampling. c. simple two-stage area sampling. d. probability-proportional-to-size area sampling. e. systematic sampling.
d.
Judgment samples are distinguished by the fact that a. the proportion of those possessing a given characteristic are balanced against the proportion of the population possessing the characteristic. b. those entering the sample happen to be where the study is being done when it is being done. c. each field worker is allowed to judge whether any given respondent would be a good representative of the population. d. the sample elements are handpicked by the investigator because it is expected they can offer the contributions sought. e. they are superior to other forms of sampling for descriptive studies.
d.
Quota samples are typically selected so that the distribution of a characteristic or characteristics in the sample matches the distribution of these same characteristic(s) in the population. With respect to the representativeness of quota samples, it can thus be said that a. a quota sample may be very far off the mark with respect to other important characteristics likely to influence the result but that were not used as controls when selecting the sample. b. the fact that the distribution of the characteristic in the sample parallels the distribution in the population does not guarantee that the sample is representative. c. the quota assigned each field worker should exactly mirror the distribution of the control characteristic(s) in the population. d. a and b are true. e. a, b, and c are true.
d.
Suppose a researcher has constructed a 95 percent confidence interval using the formula, x - o s u 2T+ z o _. The interval means that x x a. there is a 95 percent probability that the population mean is within the established interval. b. 95 percent of the sample values lie within the established interval. c. 95 percent of the population values lie within the established intervals. d. 95 percent of the intervals constructed employing this procedure would contain the true population mean and this interval is hopefully one of those 95 out of 100. e. none of the above.
d.
The distinguishing feature of probability samples is that a. they involve personal judgment somewhere in the selection of sample elements. b. each population element has an equal chance of being included in the sample. c. each population element has an equal chance of being included in the sample and every combination of 11 population elements is a sample possibility. d. each population element has a known chance of being included in the sample. e. they will always be more representative of the population than nonprobability samples.
d.
The feature that distinguishes stratified sampling from other forms of probability sampling is that a. only stratified samples involve partitioning the parent population into mutually exclusive and exhaustive subsets. b. only stratified samples allow the a priori determination of the probability that any population element will be included in the sample. c. only stratified samples make the probability of selection of any population element equal. d. only stratified samples involve partitioning the parent population into mutually exclusive and exhaustive subsets and selecting a simple random sample from each subset. e. none of the above.
d.
The following list contains the six-step procedure for drawing a sample. Select the proper order for this procedure. A. Collect the data from the designated elements. B. Identify the sampling frame. C. Define the population. D. Determine the sample size. E. Select a sampling procedure. F. Select the sample elements. a. C,B,D,E,F,A b. B,C,D,E,F,A c. F,D,B,E,C,A d. C,B,E,D,F,A e. D,C,B,E,F,A
d.
Which of the following regarding probability samples is FALSE? a. Evaluation of the precision of a sample result is important to most researchers and therefore they seek to use probability samples whenever possible. b. In general, probability sampling requires personnel who are more skilled and experienced than nonprobability sampling. c. A probability sample will usually cost substantially more than a nonprobability sample of the same size on a per observation basis. d. One real advantage that probability samples have over nonprobability samples is their greater economic efficiency. e. Nonprobability samples can be more accurate than probability samples.
d.
Which of the following types of probability samples does not require a complete list of population elements by name in order to draw a sample? a. stratified sample b. systematic sample c. simple random sample d. cluster sample e. quota sample
d.
For simple random sampling the mean of all possible sample means is equal to the population mean when which of the following conditions is true? Choose the best possible answer. a. Sampling is with replacement. b. Sampling is with or without replacement. c. Sampling is from an infinite population. d. Sampling is from a finite population. e. Sampling is with or without replacement and from an infinite or finite population.
e.
In order to define a population, the researcher needs to specify a. the geographic boundaries of the population. b. which elements are not of interest. c. the time period for the study. d. a and b. e. a, b, and c.
e.
Samples are often used to infer something about a population rather than canvassing the population itself because a. they are typically cheaper than complete counts. b. they are typically faster than complete counts. c. they are typically more accurate than complete counts. d. a and b. e. a, b, and c.
e.
Sampling plans fall under the following basic categories: a. probability b. area 0. nonprobability d. a and b e. a and c
e.
The advantages of stratified over simple random sampling include a. stratified sampling ensures adequate representation from each stratum of interest. b. stratified sampling can produce more precise sample statistics. 0. stratified sampling allows for the comparison of variables between strata. d. b and c. e. all of the above.
e.
The criteria that need to be balanced when choosing a disproportionate stratified sample are a. strata size. b. strata heterogeneity. c. strata variances on the variable of interest. d. a and b. e. a, b, and c.
e.
The distinguishing feature of a simple random sample is (are) a. each population element has a known chance of being selected. b. each population element has an equal chance of being selected. c. every combination of n population elements is a sample possibility. d. a and b. e. a, b, and c.
e.
The distinguishing feature of cluster samples in comparison to other probability sampling plans is that a. cluster samples provide each population element with an equal chance of being included in the sample. b. cluster samples provide each population element with a known chance of being included in the sample. c. cluster samples involve the partitioning of the parent population into mutually exclusive and exhaustive subsets. d. cluster samples involve the partitioning of the parent population into mutually exclusive and exhaustive subsets and the probabilistic selection of elements from each subset. e. none of the above.
e.
The steps necessary to perform a cluster sample of Dove Bar Milk Chocolate consumers are a. perform a census of two systematically chosen strata. b. divide the consumers into separate groups on the basis of age categories (e.g.,<18, 19-24, 25-34, 35-44, 45-64, 65+). c. perform simple random sampling in all the strata. d. randomly choose two age categories and interview all or some of those consumers. e. b and d.
e.
The weaknesses of a quota sample include a. the sample could be skewed with respect to a noncontrol characteristic that is likely to affect the observed variable. b. interviewers are allowed to choose their subjects and interview locations. c. it is difficult to verify whether the sample chosen is representative of the population. d. b and c. e. a, b, and c.
e.
Which of the following is NOT TRUE for simple random sampling? a. Each element has a known chance of being selected. b. Each element has a non-zero chance of being selected. c. Each element has an equal chance of being selected. d. Each combination of n population elements has an equal chance of being selected. e. All of the above are true statements.
e.
Which of the following is not a problem with using a phone book for a sampling frame? a. unlisted phone numbers b. not all homes have phones c. double counting homes with multiple phone numbers d. the phone book is always outdated e. all of the above are true
e.
Which of the following statements is true? a. A sampling distribution is the probability distribution of elements from a particular sample. b. The standard error of the mean is the standard deviation of the sampling distribution of sample means. c. The sampling distribution of the mean is another name for the frequency distribution that results from a single sample. d. The sampling distribution of the mean is another name for the distribution of means corresponding to the samples that could be drawn under the specified sampling plan. e. Both b and d are true.
e.
While giving guidelines to her field interviewing staff, a researcher states, "I want you to interview every tenth customer entering a K mart store at the following key sites." This is an example of __ sampling. a. judgement b. quota c. sequential d. simple random e. convenience.
e.