Experimental/Quasi-experimental Design
Sampling
Survey Research
Field Research/Qualitative
Interpreting Data
100
Define randomization
Random assignment is placing participants into experimental or control groups based on a random process
100
Define representativeness
each member of the population has a known and equal (non-zero) chance of being selected into the sample
100
Explain a response rate.
this is the number of people participating in a survey divided by those selected to be in the sample. extra points if you mention that those selected in the sample can be broken down more depending on IF they're reachable and chose not to respond vs. those who are not reachable :)
100
Define ethnography
a report on social life that focuses on detailed and accurate description of a concept as observed in the way it is actually conducted instead of explanation
100
explain the difference between univariate, bivariate, and multivariate.
uni=1, bi=2, and multi=3 or more variables included in the analysis
200
True or False? Longitudinal designs measuring dependent variables over several years are the only way to discuss causal relationships - they can never be determined from a cross-sectional design.
False - they can be determined from a cross sectional design, assuming we measure the DV first, institute some kind of stimulus or IV, then remeasure the DV (all of which COULD take place during 1 setting). We could also tentatively discuss causal relationships between variables that cannot change (like race) and some other variable without having to measure it over the long term.
200
What is the standard error formula and what do each of the letters represent?
standard error (S) = square root of {P (percentage of support) times Q (percentage of nonsupport) divided by n (total number of cases in our sample)}
200
Explain the difference between closed and open ended questions
open ended are those without answer options to choose from whereas closed ended questions have a finite list of options that the participant can select from.
200
True or False: field research interviews are more structured than survey interviews
False - they are less structured (set up more like a causal conversation to gather any information provided)
200
Explain the 4 levels of measurement
nominal (categories), ordinal (ordered ranking), interval (ordered with set differences from 1 value to the next), and ratio (interval with a true zero measure)
300
Explain a double blind research design
a double blind design is where the neither the participant NOR the experimenter know which condition the participant is in (experimental or control groups)
300
What is a sampling frame?
the list of all elements in our population that we use to draw our sample from
300
What are warning mailings?
a letter or post-card sent out just before mailing a survey to a respondent so they are aware of the survey coming in the next short while and will be prepared for it (and more apt to respond).
300
Explain the difference between observer as participant and participant as observer
observer-as-participant is known as a researcher and interacts with the subjects, but does not participate in the activities (ex. a ride-along with police), whereas a participant-as-observer is known as a researcher but also engages in the behaviors and activities along with the subjects (ex. using substances along with the participants to develop rapport when studying drug use/abuse)
300
Which statistic is used to measure proportionate reduction in error for ordinal variables?
gamma
400
What are the 3 major pairs in a classical experimental design?
independent and dependent variables, pretesting and posttesting, experimental and control groups
400
Explain the difference between a population and a sample, and a parameter and a statistic
The population is the entire group of elements we want to discuss findings for whereas a sample is a small (hopefully representative) subset of the population. A parameter is a value for a given variable in the population, whereas a statistic is the value for the variable found in the sample
400
When creating a closed-ended question, option choices should be WHAT and WHAT to ensure high quality measurement?
exhaustive and mutually exclusive
400
Explain how a field researcher might record observations.
note-taking (through 1st 'sketchy notes', then more details notes after observation is complete); recording through video, audio, or photographic records; structured observations (forms that check off specific things to look for in the environment)
400
What do parametric tests tell us?
statistical significance of relationships between variables
500
Pick 3 of the threats to internal validity and explain why it could be a problem.
12 could be discussed: history, maturation, testing, instrumentation, statistical regression, selection biases, experimental mortality, causal time order, diffusion/imitation of treatments, compensatory treatment, compensatory rivalry, demoralization
500
Explain the relationship between a confidence interval and a confidence level
a confidence level shows how sure we are that the statistics fall within the confidence interval, representing the parameter
500
Explain how a contingency and a matrix question work.
Contingency questions are only relevant to participants who answer a prior question in a certain way (meaning some participants will not view ALL of the questions in a survey). Matrix questions are those that use the same set of answer options (usually across the top) so that multiple questions (down the side) can be answered with 1 explanation of the answer options.
500
Explain the 5 roles of the observer in order of participation to observation
full participant, complete participant, participant as observer, observer as participant, complete observer
500
explain the difference between descriptive and inferential statistics
descriptive are simply describing the data at hand, whereas inferential is attempting to show that the sample data should hold true for the larger populations
M
e
n
u