Up to the Final Exam

Question 1

populations in sampling

Answer 1

a population is any well-defined set of units of analysis
does not have to necessarily refer to people
a population might be all the adults living in a geographical such as a country or state, or working in an organization. But it could equally well be a set of countries, corporations, government agencies, events, magazine articles, or years
what is important is that the population be carefully and fully defined and that it be relevant to the research question

Question 2

the 7-1 table

Answer 2

represents the population of adult American citizens

since it is impossible to interview everyone, a more practical approach is to select just a “few” members of a population for further investigation. This is where sampling comes in

a sample is any subset of units collected in some manner from a population

the sample size and how its members are chosen determine at the quality (that is the accuracy and reliability) of inferences about the whole population to be drawn.

Question 3

example of samples

Answer 3

political poll
takes a section of the population to determine whether we will stay in or drop out of a race

the census
tying to identify, count, and measure certain characteristics of every individual living in the country at a given time

Question 4

Bureau of Census and political scientists: how are they alike

Answer 4

very few political scientists are able to marshal such vast resources in pursuing their own research interests. Even the bureau of Census, with all its thousands of workers and millions of dollars, found it impossible to ask everyone of its questions of each person it located. Instead it developed a short questionnaire for most people and a longer one for a select few.

like political scientists and many other researchers, the bureau found it necessary to employ a SAMPLE

Question 5

the ability to draw general conclusions based on analysis of only a few cases is

Answer 5

Generalizability

Question 6

three questions about representation in generalizability

Answer 6

a) just what constitutes a representative sample?
b) how might one select the particular cases that would constitute such a sample?
c) how many cases must we select before a sample may be said to be representative?

Question 7

population v case v sample

Answer 7

a population is any group of people, organizations, objects or events about which we want to draw conclusions

a case is any member of such population

a sample is any subgroup of a population of cases that is identified for analysis

Question 8

who is nate silver?

Answer 8

guru of sampling
former ESPN employee
538: website that does perditions on elections

famous in political circles accurately predicting the last previous election almost to a T

Question 9

What are the basic steps involved in sampling?

Answer 9

in each instance, our procedure is to identify a subgroup of a larger population; to study that subgroup, or sample, in some detail and to generalize our results to the population as a whole.

Question 10

what is a representative sample

Answer 10

a sample is which every major attribute of the larger population from which the sample is drawn is present in roughly the proportioner frequency with which those attributes occur in that larger population

Question 11

a truly representative sample is….

Answer 11

a microcosm - a smaller, but accurate model- off the larger population from which it is taken.

Question 12

this _________ is what we mean by generalizability

Answer 12

extension of findings

to then extent that a sample is truly representative, conclusions based on a study of that sample may be safely regarded as applying to the original population.

Question 13

circle example

Answer 13

suppose we want to study patterns of membership in political groups among adults in the US

3 circles divided into 6 segments

suppose we want to study patterns of membership in political groups among adults in the US

circle 1: represents pop in question. member s of the population have been classified according to the number of political groups (such as parties and interest groups) they belong to. every adult is assumed to belong to at least one and not more than sex groups and these six levels of membership are equally distributed throughout the population.

circle 2: the shaded area in 2 illustrates one possible sample of the size we have specified but one that is clearly atypical of the population.

2 is not representative because it does not reflect the distribution of this population attribute (often called parameter) roughly in propotion to this actual incidence

circle 3:
the most representative sample have a small bit of each segment. a little bit of everything

Question 14

what is the biggest challenges in research?

Answer 14

funds and resources

Question 15

Literary Digest magazine in the 1930’s example**

finish me

Answer 15

LD was a periodical compilation of newspaper editorials and other opinion pieces that enjoyed a wide readership int he early years of this century. Beginning in 1920, the magazine conducted a larger-scale, nationwide straw poll in which postcard ballots were sent to more than a million persons asking them to state their candidate reference in the coming presidential election.

Question 16

How could the LD poll list miss?

what went wrong?

Answer 16

predicting a 60/40 landslide victory for Repub

Landon lost to FDR in the way he thought he was going to win.
Forced to cease publication

made the assumption that everyone had telephones, when that was not the case. only the rich did, so they were not getting an accurate sample!

Biased sample. two sources: telephone directories and automobile registration lists. most people didn’t have phones or automobiles, but the conservatives and republicans did because they had lots of money. more affluent.

Question 17

how can we avoid this problem?

Answer 17

circle example

use circle 3 to avoid LD’s problem

Question 18

Types of sampling

incomplete and inappropriate

Answer 18

Researchers make a basic distinction among Types of Samples according to how the data are collected.

political scientists make inferences about the whole population from what they know about a smaller group.

if a sampling frame is incomplete and inappropriate, Sample Bias will occur. In such cases the sample will be unrepresentative of the population of interest and inaccurate conclusions about the population may be drawn.

Sample bias may also be caused by a biased selection of elements even if the sampling frame is a complete accurate list of the elements in the population.

Question 19

Snow example*

Answer 19

*

Question 20

Probability sample

Answer 20

a sample for which each element in the total population has a known probability of being included in the sample. this knowledge allows a researcher to calculate how accurately the sample reflects the population from which it is drawn.

Question 21

non probability sample

Answer 21

each element in the population has an unknown probability of being selected. Not knowing the probabilities of inclusion rules out the use of statistical theory to make inferences.

*whenever possible, probability samples are preferred to non probability samples**

Question 22

Simple random sample

Answer 22

each element and combination of elements has an equal chance of being selected

a list of all of the elements in the population must be available, and a method of selecting those elements must be used that ensures that each element has an equal chance of being selected.

Question 23

Conditions of simple random sample

Answer 23

the sample must be chosen in such a manner that each and every individual or case int he entire population has an equal opportunity to be selected for analysis

the sample must be chosen in such a manner that each and every possible combination of n cases, where n is simple the number of cases in the Sample, has an equal opportunity to be selected for analysis.

Question 24

lottery process simple random sampling

Answer 24

often simple random samples are employed in studying populations that re too large to permit such a physical lottery procedure. Writing out the names of several hundred thousand cases, entering them in a hopper, and drawing out several thousand, would be a very cumbersome process.

the number of the particular cases to be included in the sample are then identified using a random number table

computer

Question 25

List the types of Samples

Answer 25

Simple random
systematic 
stratified (both proportionate and disproportionate)
cluster
telephone

Question 26

Systematic random sampling after simple random

Answer 26

the drawing of a simple random sample may be no simple matter. in addition to other problems, the technique involves a great deal of clerical work especially when it is employed on a large scale.

for this reason, random sampling procedures are often modified to enhance their manageability. One common variation in systematic random sample.

used when we wish to study a relatively large population who’s members are individually listed in some central location, such as a telephone book a student directory a list of registered voters, an index or a table of contents an agenda or a membership roster.