study guide Qualitative Methods

I attached the study guide

Qualitative Methods:

Lecture:

Content analysis, intercoder reliability, examples

of content analysis your professor has done,

code sheets as a way to systematize content analysis

Chapter 8: reactivity, primary data, secondary data, participant observation, field study, direct and

indirect observation, overt and covert observation, str

uctured and unstructured observation, ethnography,

ethical issues in observation (threat to subjects in comparison to other methods), Institutional Review

Boards (IRBs), the difference between an erosion measure and an accretion measure, potential problem(

s)

with physical trace measures in studying political phenomena

Chapter 9: document analysis

–

qualitative, quantitative or both, content analysis and its procedures,

sampling frame, recording units (what are they, what if they are too small?), intercoder

reliability, running

record vs. episodic record (what each is, examples, advantages and disadvantages), advantages and

disadvantages of archival (written) records

Survey Research

Lecture:

most important lesson for us as consumers of surveys, s

ampling, po

pulation, sample, the logic

of sampling (why it makes sense with the rules of statistics that a sample is a reasonable estimate of the

population), confidence interval and margin of error, confidence level, types of information that questions

generally ask

for (knowledge, opinions, experiences, feelings), common sources of error in survey

research (timing, phrasing of questions, order of questions, interpretation of responses)

, American

Journalism Review study, Bradley effect, intangible problem in sampling

discussed in lecture,

Chapter 10:

survey research vs. interviewing, survey instrument, the importance of pre

–

testing

questionnaires,

response rates, response quality, possible types of bias (leading questions, interviewer

bias, etc.), ways to prevent bi

as in surveys, sample

–

population congruence, open

–

ended vs. close

–

ended

questions (advantages, disadvantages, reasons to use one over the other), types of surveys (face to face,

telephone, internet, etc.), potential problems with questions (leading, narrow

, ambiguous, double barreled,

etc.),

the impact of interviewer characteristics, probing, question wording and ordering effects

Stats

Intro, Distributions, Descriptive Statistics

Lecture: the normal distribution, standardized (Z) scores, the bell curve, pr

operties of the normal

distribution (shape, symmetry, meaning of standard deviation, empirical rule, ability to use standardized

scores), percentiles (what are they, how are they different from a percentage), t Distribution (what is it,

what do we use it f

or?)

;

descriptive statistics,

frequency distributions, percentages as a VERY easily

understood statisti

c, measures of central tendency and the levels of measurement to which they

correspond,

measures of dispersion

Chapter 11: response set, frequency distribution, relative frequency, descriptive statistics, trimmed mean

and outliers, positive and negative skew, measures of central tendency, mode, median, mean, range,

minimum and maximum, inter

–

quartile range, resist

ant measures, measures of dispersion, standard

deviation, variance, types of charts and graphs

Chapter 12: statistical hypothesis, null hypothesis, absolute value, sampling, Type I vs. Type II error, as

standard deviation increases in size what happens to

the standard error of the mean, level of statistical

significance, factors that affect significance, steps for hypothesis testing, significance tests of a mean

(normal distribution vs. small (t) distribution), degrees of freedom in t, finding the t Value

(alpha

–

see

example in Figure 12

–

4), a z

–

score of 1.96 means what, confidence intervals and levels (what are they,

why do we use them

, the general form of confidence interval

)

Measures of Relationships

Lecture: percentage differences as the simplest way

to show relationships, comparing measures of central

tendency, strength of relationships (logic: the extent to which changes in one variable are accompanied

by changes in another

–

no matter what level of measurement, the basic logic is the same

),

Yule’s

Q

and its properties, ultimately what do we want to do?

We want to reduce error! The idea for all of our

measures is, ultimately, to know how much we can reduce error in our estimates of a dependent

variable by knowing the values of an independent variab

le (or multiple independent variables)

, the

basic equation (in words) of the measure of reduction in error, measures for nominal data (lambda, tau),

measures for ordinal data (gamma, somer’s d), measures of relationship for interval level variables (r, r

–

s

quared), steps: start with a graph (three elements of a graph), the regression line (

what does it tell us

about the variables,

think of it as a prediction), parts of the regression line: slope, direction, strength of

relationship, what the slope

(b)

tell

s us,

what the Y intercept with zero tells us,

what

Pearson’s

r and r

–

squared tell us

, rule of thumb about a “strong” value of r

Chapter 13:

levels of measurement and the statistical procedures that go with them, types of relationships

(association, monot

onic and linear correlation), types of correlation, what does a measure of association

tell us, what do cross

–

tabulations show us, nominal measures of association, ordinal measures of

association (what are concordant pairs, discordant pairs, tied pairs), b

ounded measures such as Pearson’s r

vary between

–

1 and 1,

if the categories of an independent variable are across the top of a table (across the

columns) then what should the percentages down each column add up to (100%), the effect of increased

sample si

ze on Chi

–

squared

Multiple variables

Lecture: two kinds of information in multiple correlation/multiple regression (cumulative and partial),

time series analysis, interpreting the strength of a relationship

–

what do

relationship measures tell us,

when ar

e relationship measures particularly useful,

Chapter 14: analyzing multivariate relationships with

nominal and ordinal level data (what can you do? Don’t worry about technicalities

–

just understand that

you can do this with cross

–

tabulation

, how can you

control for a third variable?

),

multiple linear

regression (used with a

dependent variable

of what level of measurement?),

constants (beta

–

y when all

the independent variables have a value of zero), partial regression coefficients, interaction between

variables, homoscedasticity, multicollinearity and assumptions about the error terms in linear models (see

helpful hints tabl

e on p 530), dummy variables,

spurious relationships, standardized regression

coefficients,

ways in which

standardized and unstandardized

regression results are similar and different

,

logistic regression (when do we use this? It has to do with the type of

dependent variable)

Statistical Significance

Lecture (posted on Canvas): how statistical significance differs from strength of relationship; review of

the normal distribution and standard deviation and standard errors, difference between margin of error

and confidence level; Verba and Nie example, examples of different measures of statistical significance