Business Stats Exam 2

Front 1

systematic sampling

Back 1

This technique is often considered when an ordered listing (e.g., telephone book, list of property owners, list of registered voters, etc...) of every member of the population is available. The process involves selecting every “nth” item using a random starting point.

Front 2

cluster sampling

Back 2

The clusters could be city blocks or some other kind of geographic area. Cluster samples are cheaper than stratified random sampling, but not as accurate

Front 3

stratified random sampling

Back 3

the population is broken out into groups or strata’s, and a random sample is then drawn from each strata

Front 4

convenience sampling

Back 4

(sometimes called accidental sampling) is non-probabilistic. As such, it is not very representative. If an interviewer decided to survey every tenth person that walks out of a store, this is convenience sampling. Subject’s end-up getting selected by being at the right place at the right time.

Front 5

quota sampling

Back 5

is a mix between convenience and stratified sampling. In addition to selecting every 10th that happens to walk by, the researcher fills the sample according to the understanding of the population proportions (similar to the way it is done in stratified sampling). For example, if it were assumed that 15% of the population is minority and a sample of 200 is needed, the analyst would look to include 30 minorities in the sample. This is a slight improvement over straight convenience sampling.

Front 6

snowball sampling

Back 6

When population listings are unavailable (e.g., community elites, individuals that purchase certain goods, individuals of specific opinions, etc...), another non-probabilistic technique known as snowball sampling can be useful. The researcher identifies a participant who meets the criteria for inclusion in a study and than asks the participant to recommend other participants who also meet the criteria.

Front 7

judgment sampling

Back 7

is a non-probabilistic method used when the researcher selects subjects based on their best judgment as to whom or what is most representative of the population and should be included in the sample to facilitate the investigation. Generally, this means building a sample based on individuals considered to be the norm or average and is judged to be typically representative of the population.

Front 8

margin of error (Sampling Error) equation

Back 8

z*(s/sqrt(n))

Front 9

margin of error (sampling error) is...

Back 9

is the amount you are willing to be off with a stated level of confidence;
really only 3 different z-scores you use: 90%, 95%, and 99%
95% is by far industry standard

Front 10

confidence interval equation (SD known)

Back 10

x ± z*(sd/sqrt(n))

Front 11

standard error or mean (SEM) equation

Back 11

s/sqrt(n)

Front 12

confidence interval equation (SD unknown)

Back 12

x ± t * (s/sqrt(n))

t = the t value providing an area in both the upper and lower tails of the t-probability distribution

Front 13

t - distribution

Back 13

a specific t - distribution depends on a parameter known as the degrees of freedom;
degrees of freedom refer to the number of independent pieces of info that go into the computation of s. generally this means n-1 degrees of freedom for s;
a t-distribution with more degrees of freedom has less dispersion. As the degrees of freedom increases, the difference between the t distribution and the standard normal probability distribution becomes smaller and smaller

Front 14

1. A discrete probability distribution:

Back 14

is the distribution of any single random variable

Front 15

2. The central limit theorem applies to

Back 15

sample mean

Front 16

3. when the standard deviation is small we may infer that the

Back 16

curve of the distribution is more peaked

Front 17

how do we determine necessary sample size for a mean (for a desired level of confidence)?

Back 17

use the margin of error (sampling error)

Front 18

4. If the value of the standard normal random variable Z is positive, then the original score is where in relationship to the mean?

Back 18

greater than the mean

Front 19

5. The standard deviation of a probability distribution is a:

Back 19

measure of variability of the distribution

Front 20

6. The standard normal distribution is referred to as the:

Back 20

z distribution

Front 21

7. The standard normal distribution has a mean = _____ and standard deviation = ______.

Back 21

0 and 1

Front 22

8. In a normal distribution, changing the standard deviation:

Back 22

makes the curve more or less spread out

Front 23

9. Selecting a random sample from each identifiable subgroup within a population is called:

Back 23

stratified sampling

Front 24

10. The standard deviation of x is usually called the

Back 24

standard error of the mean

Front 25

11. The sample mean x is the ________ estimate fr the pop mean

Back 25

point

Front 26

12. which of the following statements are correct?

Back 26

an interval estimate is an estimate of the range of possible values for a population parameter

Front 27

13. all confidence intervals are a function of which of the following three things?

Back 27

the data in the sample, the confidence level and the sample size

Front 28

14. as the samples size increases, the t-distribution becomes more similar to the ______ distribution.

Back 28

z

Front 29

19. You want to determine the necessary sample size to obtain a 95 confidence interval on a population proportion that has an error of plus or minus 1%. If you have no prior knowledge about the likely value of the population proportion then you can apply the appropriate formula using and estimated proportion value of _______.

Back 29

0.50