Info Management Exam 2 Review

Front 1

Formal Political Theory

Back 1

The analysis of rational choices and their aggregate consequences in non-market contexts

Front 2

Reasons for Formal Politcal Theory

Back 2

• Deductive method of theory
• Guarantees theories are consistent
• Guarantees the logical validity
• Must be descriptive or real world
• Must be testable
• Method of theory construction

Front 3

Some definitions in formal political theory

Back 3

• Some propositions are fundamental to the system
• Assumptions or axioms; fundamental prop which form
 Assumptions: empirical conjectures
 axioms
• Both are testable; from these proposition, other propositions are deduced
• The use of logic, or the set of rules which preserve the truth of an argument, guarantees that an argument is consistent
 Argument; set of propositions or statements
 Consistency: assume that all the statements and propositions can be true simultaneously
• Relationship of logical derivation or deduction, in argument is a relationship. Deductively valid argument requires that the premise is true

Front 4

Assumptions of formal political theory

Back 4

• Rationality
 Transitivity A > B, B>C = A > c
 Indiv able to rank alternatives best to worst
 Rankings have property of transitivity
 Actors are thought to choose according to what is best for them
 Rationality doesn't preclude altruistic behavior
 Desirability and likelihood
• Preferences and expected value
 Obviously, any successful theory must take into account not only the utilities of the possible outcomes but their
 FPT uses
• Self Interest and other value constraints
 Other conjectures are added which narrow the types of interests actors are presumed to hole by
 Helps with deductive simplicity
 Others
• Maximize votes, max budgets

Front 5

Game Theory

Back 5

• Prisoners Dilemma: whether to defect or to cooperate
1. Assumptions:
• actors involved are rational, engaging in purposive behavior
• Actors will pursue their self-interests
• The actors will not know what the others said
• The game is played only once
2. Conclusions
• In best interest of both prisoners to talk
• Robert Axelrod's book The Evolution of cooperation
3. Problems with Game
• Not future, only play once
• Assumption that there is no comm b/w nation-state
• Assumption that people make decision based on individual, short-term interests
• Ex. Prisoners Dilemmas

Front 6

The collective good problem, including Dr. Hardt’s wonderful sheep

Back 6

o Sometimes theses are called public goods
o They have the following properties:
• Non-exclusivity - if the good is provided, you can't exclude anyone form enjoying it…ex. Public radio
• Indivisibility - use a collective good by one doesn't diminish the amount of good left over for the others..ex national defense
• Free-rider problem - someone who enjoys the benefits of a collective good but doesn't contribute to it ie the fight to save the spotted owl
 Mancur Olson
• 3 reasons:
• Material Benefits: discounts, etc..
• Solitary Benefits: Vets, PTA, etc…
• Purposive benefits: cause,

Front 7

The problem of collective goods

Back 7

• Sheepville: country and makes only one major product which is sheep/wool. Raise their product on a piece of land the size of a football field. Residents can raise as much as they want.
• Consequence of too many: diseases, trampled, and starved (death)
• Government steps in: before disaster hits the fan. Regulated, tax incentive for not raising, fine, bail, tax, subsidize
• 1 major product

• What typically happens bf event can reach this stage? Gove steps in …. What can gov. do to fix situation?
 Regulate #of sheep
 Tax incentives
 Fine bailout

Front 8

Decision tree analysis as an example of formal theory

Back 8

a diagrammatic method of expressing action choices available to decision maker, and options are determined by chance

Front 9

4 major aspects of decision tree analysis

Back 9

1. Decision tree branches; probabilities
2. Decisions events (boxes)
3. Chance events (circles)
4. The expected value of the payoffs for each branch

Front 10

5 key concepts for the foundation of decision analysis

Back 10

1. Decision alternatives: to be or not to be
2. Outcomes: each decision alt is associated w/ a set of poss outcomes or future states (probabilities)
3. Outcome probabilities : each outcome has a probability of occurrence for each decision alt
4. Associated payoff value: each outcome, for each decision alt, has a numerical payoff value
5. Expect payoff value: each decision alt, is assoc w/ an expected payoff value, which is the sum of the products of the payoff values and their respective probabilities

Front 11

Values of a decision tree analysis

Back 11

• validity rest squarely on the accuracy w/ which one can estimate the probabilities of various outcomes and their respective payoff values
• Decision analysis is thus more of an art than a science.
• Also, the amount of risk a decision -maker is willing to bare is a major factor
• Decision analysis can also serve a heuristic function by making us think, helping us with hypothesis construction , and exploratory analysis

Front 12

NYT vs. Tasini, the other examples of internet law

Back 12

New York Times Co. v. Tasini, 533 U.S. 483 (2001), is a leading decision by the United States Supreme Court on the issue of copyright in the contents of a newspaper database. It held that The New York Times, in licensing back issues of the newspaper for inclusion in electronic databases such as LexisNexis, could not license the works of free-lance journalists contained in the newspapers.

Front 13

Types of polls

Back 13

o Tracking poll: conducted as one of a series of polls, used to monitor a small # of aspects of public opinion
o Instant reaction poll: get group reactions immediately
o Call-in poll: this is where tv viewers are invited to call in on a 900 # and register thir opinions on a specific issue
o Exit poll: a poll taken on election day after voters have voted--major media orgs--one of the most important sources of insight into what was on the voters minds when they voted

Front 14

Normal Polls: usually used

Back 14

1. The nationwide poll: usually done in a long interview at home where an effort made to contact about 1200 voters
2. Media telephone poll: usually to specific event or happening ; most always done over the telephone at night and they usually get he same size samples as the nationwide poll, but cannot ask as many questions

Front 15

The Key (polling)

Back 15

o The key: person interviewed must be a random sample that is repres. Of the entire popul.
o There are some key terms used here:
• Random: any given person must have an equal chance of being selected
• Sample

Front 16

Refusal response rate

Back 16

about 1/5 to 1/3 of all respondents will refuse---some groups refuse more often

Front 17

Sampling error

Back 17

the discrepancy between an observed and a true value that arises purely by chance or happenstance

Front 18

Confidence intervals

Back 18

the range of likely values associated with a given probability or confidence level

Front 19

Sampling fraction

Back 19

the size of the desired sample (n) divided by the size of the population (N) or (n/N)

Front 20

Population parameters

Back 20

characteristics of the population. Ex….The % of the population that approves of the way that the president is handling his job.

Front 21

Sampling statistics

Back 21

The % of the sample that approves of the way that the president is handling his job. Is unbiased if its expected value is equal to the corresponding population parameter. The more sampling you do the closer you get to the populations ideas.

Front 22

Normal distribution of sampling error

Back 22

Distribution defined by a mathematical formula and the graph of which has a symmetrical, bell shape; in which the mean, mode and median coincide; and in which a fixed proportion of observations lies between the mean and any distance from the mean measured in terms of the standard deviation.

Front 23

Inverse square root law and its implications for doing survey research

Back 23

o Ex. Sample of a given size has a margin of error of 6% we can reduce this margin of error by increasing the sample size, but it takes a sample size four times as large to cut it in half.
o E2/E1= square root of N1/N2
o Implications (trouble) in order to go up in a sample size you have to be willing to pay the cost. Use subgroup (African American) women, white women) 3 to 4 ½ the margin or error will go up.

Front 24

Margin or error

Back 24

Magnitude of the confidence interval

Front 25

Types of survey questions

Back 25

o Open ended:
1. respondent is not provided with any answers from which to choose
2. allow respondents to explain answers more fully and create the opportunity for researchers to find answer choices they had not anticipated.
o Close ended;
1. provides respondents with a list of responses from which to choose
2. easy to code and analyze, but they limit answers to those predetermined by the researcher

Front 26

Types of surveys

Back 26

1. rating scales: 1-7
2. thermometer: how do you feel
3. agree/disagree format
4. indices:
5. Michigan studies
a. National studies
b. Pre and post election panels
c. Occasional extended panels (four years)
d. Party id was important in the 80’s
6. ANES: American national election studies

Front 27

Designing survey questions and all the problems therein
5 factors to consider

Back 27

1. Cost
2. Completion rates: If response rate is low then the likelihood that the sample is representative of the population decreases
3. Sample-population congruence: if those that fail to participate in the survey share common characteristics, then the sample will be wrought with sampling error and
4. Questionnaire length: lengthy questionnaire can lead respondents to lose focus or speed through questions in order to finish sooner, resulting in low response quality. Different survey methods can use different length surveys with the maximum number of questions ranging between three or four dozen depending on complexity and format.
5. Data Processing Issues: Although advancements in personal computing have made the collection, preparation, and analysis of survey data easier, a good deal of time and effort are still required to process data. Surveys using more elaborate designs, large numbers of questions, or many open-ended questions will require more time and effort.

Front 28

Problems with designing survey questions

Back 28

1. question must be comprehensible – 7th grade should be able to answer it
2. must be asked fairly and not be leading
3. unbalanced v balanced format
4. dealing with ambiguity – but this time in the question
a. ex. Time question; how many years have you been in school?
5. Presence/absence of alternatives
6. invoking authority
7. double barreled
8. lengthy questions
9. negatives; I disagree that cand. Should not
10. social desireability

Front 29

Bad questions

Back 29

1. do you think teens caught with cigarettes should be fined in order to prevent them from smoking.
a. Leading question
2. the pace at our company is H
a. negative about the company and assuming the pace
3. I can always talk to my supervisor about problem
a. Always in the question
4. what brand of computer do you own? Ibm, mac, dell
a. no alternatives

Front 30

Types of sampling

Back 30

o Random Sampling (probability sampling) disproportionate
1. samples for which each element in the population has a known probability of inclusion in the sample. These sampling methods are the first choice because they produce the most representative samples:
a. simple random
b. systematic sample
c. stratified sample
d. cluster sample
o Non-random sampling (non-probability sampling)
1. a sample which each element in the pop has an unknown probability of being selected - increase the amount of sampling error involved bc they are not done w/ the same precision
a. purposive sample

Front 31

Complicating factors in public opinion

Back 31

a. Ambivalence: public often wants 2 things that contradict
b. Direction
c. Intensity
d. Saliency: the importance of significance of the object--how much of a persons life is it
e. The amount of info--we are relatively uninformed: public doesn’t know a whole lot about specific details of government
f. Volatility: public attitudes towards a given policy by the government casn vary over time and quite dramatically--when knowledge base is detaled, there is a rare change in opinions
g. Latency: become active after some stimulus -- think of it as somehwhat hidden, but ready to be mobilized--attitudes and opinions can be latent---an inattentive public
h. Degree of integration: do you have a set of consistent attitudes and opinions that fall into some sort of ideology?
• Are they well integrated
• Each of the opinions and attitudes is related to the others and is internally consistency

Front 32

Content analysis

Back 32

o Measuring the amount of times someone (Pres.) references a certain issue (economy) in a speech (such as the State of the Union.
o Steps for content analysis
1. select material germane to the research subject (the sampling frame) and then sampling the material to be analyzed from that sampling frame.
2. define the categories of content that are going to be measured – the topics of interest within the content.
3. choose the recording unit – how to divide the content into standard units for analysis (e.g., a single word, paragraph, page)
4. decide the numeric values that will be used to code each category in each recording unit

Front 33

Conducting elite interviewing

Back 33

Generalizing to a larger population from an anonymous sample verses learning about a select group of individuals specifically because of who they are

Front 34

Research design

Back 34

o a plan of action for executing a research project that specifies the theory to be tested, the unit of analysis (like individual, organization, or country), the necessary observable data and how they will be collected, and the procedures that will be used to examine the data
o Many things affect the choice of research design, including the purpose of the research (exploratory, descriptive, explanatory) and practical limitations (resources like time, money, and skill or ethical concerns

Front 35

Episodic v. running record

Back 35

o Episodic: records that are not part of an ongoing, systematic record-keeping program but are produced and preserved in a more casual, personal, and accidental manner
o Running: produced by organizations, carefully stored and accessed; available for long periods of time

Front 36

Statistical analysis

Back 36

Collection, organization, and interpretation of data

Front 37

Univariate statistics

Back 37

statistical analysis with one variable

Front 38

Multivariate statistics

Back 38

statistical analysis with more than one variable

Front 39

Data can be organized in 2 basic categories

Back 39

1. Descriptive stats: measurement of population characteristics (parameters) – within the category of descriptive statistics can be two fundamental groups of measures:
2. Inferential stats: manipulation of the data to make generalizations about a population or to reach judgments about causation.

Front 40

Descriptive stats: measurements

Back 40

1. measures of central tendencies (location)
a. point of central tendency: where does the data seem to fall?
b. Center of mass
c. May be inadequate if you have clustered data
d. Mean, median, mode, skewness, robustness, kurtosis
2. measures of dispersion
a. variation
b. spread
c. minimum, maximum, range, IQR, deviation, variance, standard deviation

Front 41

Mean (X bar)

Back 41

1. “average” or x: sum of all #’s, divided by the # of values
2. interval and ration (sometimes applied to ordinal scales)

Front 42

Mode

Back 42

1. categorize the #’s to contain all the #’s (10-20, 20-30)
2. take the bin w/the most observations and find it’s mean
3. if more than one option; take the mean of the mean
4. nominal and categorical ordinal data

Front 43

Range

Back 43

1. Maximum – Minimum = single number
2. for interval and ratio level scales

Front 44

Median

Back 44

1. middle point or value that separates the groups into 2 equal groups
2. odd # of observations = take the middle #
3. even # of observations = take the 2 middle and divide by 2
4. ordinal, interval and ratio data

Front 45

Interquartile Range (IQR)

Back 45

1. divide Range (max – min) by 4, then
2. add that value to your minimum and subtract it from your maximum value
3. lastly subtract these 2 value’s and this is you IQR

Front 46

Variance

Back 46

1. Quantity that summarizes each point’s deviation from the mean without direction.
2. the sum of the all the values of (x-x(mean)) squared and then
3. divided the sum of all those values by (n – 1) for sample and (n) for population
4. the standard deviation squared

Front 47

Standard Deviation

Back 47

the square root of the variance

Front 48

Nominal

Back 48

1. Variable values are unordered names or labels. (examples: ethnicity, gender, country of origin)
2. ct: mode or modal category (frequency category with the most observations)
3. disp: proposition of the observed values that are in the modal category

Front 49

Ordinal

Back 49

1. Variable values are labels having an implicit but unspecified or measured order. Numbers may be assigned to categories to show ordering or ranking, but strictly speaking, arithmetical operations (for example, addition) are inappropriate. (Example: scale of ideology)
2. ct: median and mode can both be used
3. disp: range and interquartile range (iqr) – most preferred measure of dispersion for ordinal variables is the iqr – can anyone remember why?? Because it is less susceptible to outliers than the range

Front 50

Interval

Back 50

1. numbers are assigned to objects such that interval differences are constant across the scale, but there is no true or meaningful zero point (examples: temperature, intelligence scores)
2. ct: mode, median, and now the mean (subject to extremes, remember?)
3. disp: range, iqr, and now the mean deviation or MAD, variance, SD

Front 51

Ratio

Back 51

1. in addition to having the properties of interval variables, these scales have a meaningful zero value. (Examples: income, percentage of the population with a high school education)
2. ct: mode, median, and now the mean (subject to extremes, remember?)
3. disp: range, iqr, and now the mean deviation or MAD, variance, SD

Front 52

Understand different types of graphs and when to use them

Back 52

o Presentation graphs
1. end products, summarize information or data to be published or otherwise publicly disseminated
2. Bar Graph
3. Pie Diagrams
o Exploratory graphs
1. all in one devices that simultaneously display various aspects of data, such as central tendency, variation and shape.
2. Histograms
3. Dot Plots
4. Box plot
5. Time Series Plot

Front 53

Probabilities (relative frequency approach)

Back 53

o The likelihood that, in the long run, an observation will have a specific characteristic or an event will occur.
o Probabilities must lie between
1. 0 and 1.0, inclusive
2. for a random process, such as a coin toss or responses to a poll question, the sum of the probabilities must equal 1.0 (this just means something must happen)