Stats Exam 1 Assignment #2

Front 1

Population and Parameters

Back 1

The key numbers used in mathematical models to represent reality. Pop. Parameters are used in a model for a population

Front 2

Simple Random Samples (SRS)

Back 2

the standard against which we measure other sampling methods, and teh sampling method on which the theory of working with sample data is based.

each combination of people has an equal chance of being selected (randomization of selection of the sample group)

Front 3

Stratified Sampling

Back 3

sometimes a population is first sliced into homogenous groups (subgroups) called <b>strata</b> <i>before</i> the sample is selected. <b>Then</b> Simple random sampling is used (SRS) within each stratum before the results are combined.

Front 4

Cluster and Multi-Stage Sampling

Back 4

Splitting the population into representative clusters can make sampling more practical.

Then select one or a few clusters at random and perform a census within each cluster.

(eg assessing the reading level of a text book based on the length of the sentences)

Front 5

Systematic Sampling

Back 5

surveying every nth person from a sampling frame. The systematic selection must begin from a randomly selected individual

Front 6

Observational Studies

Back 6

researches don't assign choices; they simply observe them.

Good for discovering trends and possible relationships

Front 7

Randomized Comparative

Back 7

-

Front 8

Experiments

Back 8

requires random assignment of subjects to treatment (randomization).

Study the relationship between 2 or more variables.

Must identify at least one <b>explanatory</b> variable called a FACTOR to manipulate, and at least one <b> Response Variable</b> to measure.

Experiments actively and deliberately manipulate factors to control the details of the possible treatments and assigns the subjects to those treatments at random.

Front 9

Difference between Cluster Sampling & Stratified Sampling?

Back 9

Stratify to ensure that sample represents <i>different groups</i> in the given population (eg Male & Female are the strata while the population is a college campus) Strata are <b>Homogenous</b> but differ from each other (all male stratum vs. all female stratum)

Clusters are <b>Heterogenous</b> and resemeble the overall population. Use clusters to make sampling more practical/affordable (eg figuring out what freshman think about caf. food, to hard to track each down from your SRS so you decide to randomly select 2 dorms out of the 10 freshman dorms and then perform a census on those 2 dorms)

Front 10

Multistage Sample

Back 10

Sampling Scheme that combines several methods

Front 11

The best way to protect a survey from unanticipated measurement errors is to perform a ____

Back 11

Pilot

A trial run of the survey you eventually plan to give to a larger group. Analyze the results from the small group to see if you can use them

Front 12

Bad Survey Sampling Methods because they have no well-defined sampling frame

Back 12

Convenience Survey

Voluntary Response Survey

Front 13

Undercoverage

Back 13

some portion of the population is not sampled at all or has a smaller representation in the sample than it has in the population (potential source of bias)

Front 14

Sources of Bias

Back 14

Undercoverage

Non-response

Response (interviewer effect)

Question Wording

Front 15

Non response Bias

Back 15

problem is that those who dont respond may differ from those who do and they may differ on just the variables pertinent to the survey

Front 16

Response Bias

Back 16

refers to anything in the survey design that influences response. (not the opposite of nonresponse). The tailoring of responses to please the interviewer.

Front 17

Garbage in garbage out--there is no way to recover from a poorly designed/biased survey

Back 17

true

Front 18

3 Elements of Good Experimental Design

Back 18

1. Control or Comparison Group

2.Randomization (Random allocation of treatments to subjects & reduces effect of lurking variables)

3. Replication (Asssigning more than 1 subject to a treatment (sample size larger than 1))

Front 19

Retrospective Study

Back 19

identify subjects with target variables and collect data from the past

Front 20

Prospective Study

Back 20

Identifying subjects in advance and collecting data as events unfold

Front 21

Subjects/Participants

Experimental Units

Back 21

Humans who are experimented on

rats days petre dishes of bacteria

Front 22

Treatments

Back 22

the specific values that the experimenter chooses for a factor are called the levels of the factor.

the combination of specific levels from all the factors