MMC 3420 EXAM 2 QUESTIONS & ANSWERS
Distinguish between a population and a sample - Answer -Population is everyone,
sample is just a portion of the population.
Why do we take samples? - Answer -Obtaining a group of people from a population in
such a way as to be representative of that population
What is the key difference between probability sampling and non-probability sampling? -
Answer --Random or Probability Sampling: Everyone in the population has an equal
chance of being included
Follows mathematical guidelines and theory that allow us to calculate with 95%
confidence that the sample results can be generalized
-Not everyone in the population has an equal chance of being chosen (Sampling for a
specific purpose. Only those who fit certain criteria)
What is a sampling frame? - Answer -All messages or people to be surveyed within our
population (List of registered voters in Orange County)
Universe - Answer --Specify boundaries of what's being explored
-Operational definitions: Topic area, time period, sources
What are the types of non-probability samples? - Answer --Available/Convenience
("man on the street"): Mall intercept
-Purposive: Focus group participants (Selected for specific purpose, not to be
representative)
-Quota: Selected to represent quotas of participants that exist in your population
(Previous research said population is 40% male, so you survey 40/100 males)
-Snowball: Identify small sample of population and then use a "pass-along" method
What are the types of probability samples? - Answer --Simple Random Selection
-Random Sampling
-Single or multi stage
-Systematic Random Selection
-Stratified Random Selection
What is Simple Random Selection? - Answer --Ex. List of all home addresses in prime
TV viewership market
-Start at an arbitrary point
-Follow our list of all household addresses in the TV market
-Use a random number generator until we meet our sample size
-Uses a randomly selected start number
Each unit has an equal chance of being selected
With or Without Replacement
, Random Sampling - Answer -Everyone in the sample has an equal chance of being
chosen and responding (Regardless of researcher biases)
Allows researcher to infer beyond the sample to the population by knowledge of
Population parameters (characteristics) Estimating accuracy
Estimating error of measurement
Single or Multi stage - Answer -Cross-sectional (single time)
Longitudinal
-Trend: Randomly selected different people from a --population over time
-Panel: The same randomly selected people from a population over time
Systematic Random Selection - Answer --Uses a sampling interval to select every nth
unit; Widely used
-Subject to periodicity issues: Order of units in a list may introduce bias into selection
Stratified Random Selection - Answer -Chosen units represent a % of the population
drawn at random from that population; Similar to quota sampling, but done so with
measures put into place to make it random
Standard Error - Answer -the standard deviation (or how samples deviate from the
population)
-Regular old random error
-Standard error decreases as sample size increases
Standard Error continued... - Answer --allows you to calculate a confidence interval
around a particular sample mean
-Tells you much error to expect between your sample mean and population mean
-This is determined by Size of the sample, and the Standard Deviation assoc. with the
variable of interest
-A low sampling error indicates less variability or range in the sampling distribution of
scores
Sampling Error - Answer -(margin of error, confidence interval) how representative
sample is to the population
-May be statistically estimated, but only for probability samples
Sampling Error continued... - Answer --The discrepancy between the sample statistic
and the true population
-It is the margin of error - it can be set prior to data collection to help you determine
necessary sample size
Our class is a sample of UCF students (N = 155) of a total -60,000 UCF student
population
What is the central limit theorem? - Answer -For samples of a sufficiently large enough
size, the real distribution of means is almost always about normal. The original variable
can have any distribution
Distinguish between a population and a sample - Answer -Population is everyone,
sample is just a portion of the population.
Why do we take samples? - Answer -Obtaining a group of people from a population in
such a way as to be representative of that population
What is the key difference between probability sampling and non-probability sampling? -
Answer --Random or Probability Sampling: Everyone in the population has an equal
chance of being included
Follows mathematical guidelines and theory that allow us to calculate with 95%
confidence that the sample results can be generalized
-Not everyone in the population has an equal chance of being chosen (Sampling for a
specific purpose. Only those who fit certain criteria)
What is a sampling frame? - Answer -All messages or people to be surveyed within our
population (List of registered voters in Orange County)
Universe - Answer --Specify boundaries of what's being explored
-Operational definitions: Topic area, time period, sources
What are the types of non-probability samples? - Answer --Available/Convenience
("man on the street"): Mall intercept
-Purposive: Focus group participants (Selected for specific purpose, not to be
representative)
-Quota: Selected to represent quotas of participants that exist in your population
(Previous research said population is 40% male, so you survey 40/100 males)
-Snowball: Identify small sample of population and then use a "pass-along" method
What are the types of probability samples? - Answer --Simple Random Selection
-Random Sampling
-Single or multi stage
-Systematic Random Selection
-Stratified Random Selection
What is Simple Random Selection? - Answer --Ex. List of all home addresses in prime
TV viewership market
-Start at an arbitrary point
-Follow our list of all household addresses in the TV market
-Use a random number generator until we meet our sample size
-Uses a randomly selected start number
Each unit has an equal chance of being selected
With or Without Replacement
, Random Sampling - Answer -Everyone in the sample has an equal chance of being
chosen and responding (Regardless of researcher biases)
Allows researcher to infer beyond the sample to the population by knowledge of
Population parameters (characteristics) Estimating accuracy
Estimating error of measurement
Single or Multi stage - Answer -Cross-sectional (single time)
Longitudinal
-Trend: Randomly selected different people from a --population over time
-Panel: The same randomly selected people from a population over time
Systematic Random Selection - Answer --Uses a sampling interval to select every nth
unit; Widely used
-Subject to periodicity issues: Order of units in a list may introduce bias into selection
Stratified Random Selection - Answer -Chosen units represent a % of the population
drawn at random from that population; Similar to quota sampling, but done so with
measures put into place to make it random
Standard Error - Answer -the standard deviation (or how samples deviate from the
population)
-Regular old random error
-Standard error decreases as sample size increases
Standard Error continued... - Answer --allows you to calculate a confidence interval
around a particular sample mean
-Tells you much error to expect between your sample mean and population mean
-This is determined by Size of the sample, and the Standard Deviation assoc. with the
variable of interest
-A low sampling error indicates less variability or range in the sampling distribution of
scores
Sampling Error - Answer -(margin of error, confidence interval) how representative
sample is to the population
-May be statistically estimated, but only for probability samples
Sampling Error continued... - Answer --The discrepancy between the sample statistic
and the true population
-It is the margin of error - it can be set prior to data collection to help you determine
necessary sample size
Our class is a sample of UCF students (N = 155) of a total -60,000 UCF student
population
What is the central limit theorem? - Answer -For samples of a sufficiently large enough
size, the real distribution of means is almost always about normal. The original variable
can have any distribution
