Quiz5ch11&12 Flashcards Preview

Marketing Research > Quiz5ch11&12 > Flashcards

Flashcards in Quiz5ch11&12 Deck (34)
Loading flashcards...
1
Q

What is the difference between a sample and a census?

A

A sample is a subset of all members of a population and a census is data from about every member of the population

2
Q

What are the seven steps in devloping a sample plan?

A

1) define target population, 2)choose data collection method, 3) select sample frame, 4)obtain sample, 5) determine sample size, 6) select sample units, 7)Conduct fieldwork

3
Q

What is one thing you should always do when defining populaion of interest?

A

Create screening questions to eliminate individuals who do not belong

4
Q

What is a sampling frame?

A

A list of elements from which units to be sampled are selected

5
Q

What are some examples of sampling frames?

A

Telephone book, customer list, random-digit dialing

6
Q

What are the two major sampling methods?

A

Probability and nonprobability samples

7
Q

What is a probabilty sample?

A

A known, nonzero probabilty of selection

8
Q

What are four types of probability samples?

A

Systematic, stratified, cluster, simple random

9
Q

What is a nonprobability sample?

A

Elements selected in a nonrandom manner (convenience)

10
Q

What are four types of nonprobability samples?

A

Convenience, snowball, judgment, quota

11
Q

What are three advantages of a probabilty sample?

A

Information from a representative cross-section, sampling error can be computed, results are projectable to the total population

12
Q

What are errors that results when the same sample is not perfectly representative of the population?

A

Sampling Error

13
Q

What is the formula for computing the simple random probabilty of selection?

A

Sample size / Population size

14
Q

What sampling uses fixed skip intervals to draw elements from a numbered population?

A

Systematic Sampling

15
Q

What is the formula for a skip interval?

A

Population Size / Sample Size

16
Q

What are the steps in implementing a properly stratified sample?

A

1) Identifiy salient demographic or classification factors to correlate behavior of interest, 2) determine what % of pop. Fall into subgroups, 3) Select separate simple random sample

17
Q

What are the two basic steps in cluster sampling?

A

1) Population of interest is divided into mutually exclusive subsets, 2) a random sample of subsets is selected

18
Q

What is a nonprobabilty sample used to select additional respondents on the basis of referrals?

A

Snowball samples

19
Q

What are general rules about determining sample size?

A

Larger sample-smaller error, larger sample cost more on linerar basis, before trying to determine size of sample you must decide confidence interval

20
Q

What happens as sample size gets smaller?

A

Sampling error increases

21
Q

What do you need to calculate sample size?

A

1) an estimate of standard deviation, 2) acceptable level of sampling error, 3) confidence interval

22
Q

What is the distribution of a large number of sample means or sample proportions will approximate a normal distribution called?

A

Central Limit Theorem

23
Q

What are four important characteristics of a normal distribution?

A

1) bell shaped and one mode, 2) symmetrical about the mean, 3) defined by its mean and standard deviation, 4) area =1

24
Q

What is the formula for z score?

A

(Value of variable - mean of variable) / standard deviation

25
Q

What are the percentages for the area under the curve for z values 1,2,3?

A

68%, 95%, 99%

26
Q

What is the formula to calculate the standard error of the mean (standard deviation)?

A

=Standard deviation / SQRT(n)

27
Q

What are the steps in calculating confidence interval?

A

1) calculate mean, 2)estimate standard deviation, 3) estimate standard error, 4) estimate z value, 5) calculate

28
Q

What are four methods used to deal with the estimate of population standard deviation?

A

1) use results from prior survey, 2) conduct pilot survey, 3) use secondary research, 4) Use judgement

29
Q

What is the formula for calculating sample size based on a proportion or precentage?

A

n = (Z^2 * (P(1-P))) / E^2

30
Q

What is the type of error saying there is a differnce when there is no difference?

A

Type 1 Error

31
Q

What is the probability of not making a Type 2 Error?

A

Statistical Power

32
Q

What is the error of saying there is no difference when there is a difference

A

Type 2 Error

33
Q

What does the statistical power of standard formulas for calculating sample size all assume a power of what?

A

50%

34
Q

What are the z values for a 90%, 95%, and 99% confidence interval?

A

1.645, 1.96, 2.575