Study your flashcards anywhere!

Download the official Cram app for free >

  • Shuffle
    Toggle On
    Toggle Off
  • Alphabetize
    Toggle On
    Toggle Off
  • Front First
    Toggle On
    Toggle Off
  • Both Sides
    Toggle On
    Toggle Off
  • Read
    Toggle On
    Toggle Off
Reading...
Front

How to study your flashcards.

Right/Left arrow keys: Navigate between flashcards.right arrow keyleft arrow key

Up/Down arrow keys: Flip the card between the front and back.down keyup key

H key: Show hint (3rd side).h key

A key: Read text to speech.a key

image

Play button

image

Play button

image

Progress

1/58

Click to flip

58 Cards in this Set

  • Front
  • Back
Longitudinal Data
data collected that can be ordered in time

in some designs, the same panel is followed over time

sometimes sample members are rotated or completely replaced (measuring population stats)

frequency of follow up measurements can vary
-before and after designs
-studies can have multiple waves of follow ups (looking at trends, patterns)

cross sectional
Trend Studies
Panel Studies
Cohort Designs
Repeated Cross Sectional Designs/ Trend Studies
i.e. polling

sample drawn from population at time one and data is collected from them

as time passes, ppl leave and enter population

at time 2 a different sample is drawn from this population and data are collected

can't establish causality bc they are different sets of people

hinting at causality bc its longitudinal, but bc its a different set of people, you can't infer causality
Fixed-Sample Panel/Panel Studies
study in which data is collected from the same individuals (the panel) at 2 or more points in time

panel is drawn from pop. at time 1 and data is collected

at time 2 data are collected from the same ppl as in the panel, except for those who cant be located

repeated over and over

useful for testing a causal hypothesis

downside-expense and attrition
Cohort Study
study in which data are collected at 2 or more points in time from individuals in a cohort

Cohort: individuals or groups with a common starting point (college class, employees at UConn who started between 1980 and 1985, Gen X)

similar to trend studies but sample of cohort members have some common starting point

sampling your graduating class, you won't get the same people every time you collect the study, but they will still be PART of that graduating class
Survey Research
Involves the collection of infor from a sample of individuals through their responses to questions

an efficient method for systematically collecting data from a broad spectrum of individuals and social settings

need a uniform set of questions
skill needed for making the questions
Writing Survey Questions
must be asked of many ppl

same survey question must be used with each person, not tailored to the specifics of a convo

must be understood in the same way by ppl who differ in many ways

open and closed-ended questions
>>open ended are usually lengthier, can be limited to a specific length (2 sentences)

avoid confusing phrases
>>use short words and sentences
>>avoid double-barreled questions
(2 questions in one sentence is not okay)

be clear what the question is intended to ask

be sure questions are asked only of respondents that may have information
>>use filter/screening and contingent questions
(if you answered 'no' to Q3, move on to Q9) (N/A)

minimize risk of bias
>>avoid idiosyncratic errors in wording
(universal medicine v. socialized healthcare)

offer the 'don't know' response to avoid forcing people to choose
Designing Questionnaires
Questionnaires:
>>the survey instrument containing the questions in a self administered survey

Interview schedule:
>>the survey instrument containing the questions asked by the interviewer in an in-person or phone survey

make sure the questions are ordered in an appropriate manner (don't ask about sexuality right off the bat, too personal)

let the person know how long the survey will take

for in-person interviews, make sure the person doing the interview is comfortable (when questions are sensitive, if a woman responder, use woman interviewer)

make questionnaire attractive

be careful making motives clear in a survey, don't want them to jump to conclusions about the research
Things to Consider When Creating a Survey (measurement error and non-response bias)
Writing Questions:
ambiguous
biased
specific enough
too long?
too sensitive?

Designing the Instrument
order of questions, logical?
impact of order on answers?
hard to follow?
too thick?
hard to read?
not appealing enough?
any superfluous questions?

Administering the Survey
interviewer influence?
certain kinds of people less likely to respond?
incentives to get more people to respond?
In-Depth Interviews
relies on open ended questions

expect respondents to answer in their own words

engage researchers more actively with subjects than standard survey research

develop a comprehensive picture of
Multiperson Interviews: Group Interviews and Focus groups
usually small numbers of people

Group interview:
one interviewer directs the interview with at least 2 respondents
useful when group is a unit of analysis (family, class, etc)
-removes inhibitions of ppl regardless of relationship and help them recall info


Focus groups:
uses group interaction on a topic used as pilot studies
>test particular questions on people, looking how people respond, how should we reword our questions to get responses that pertain to our study

ppl in focus group are not related, small number of people with similar characteristics (opposite of group interviews, who are connected in some way)

PROBLEMS
some people are inhibited to answer fully bc they are not comfortable with some topics

OR

sometimes more conversation can lead to more recollection of material through spur of topic memories
Interview Schedule
UNLIKE a questionnaire, these are questions proctored by another person. When the questions are done individually, its a questionnaire, when other person is involved, its an interview schedule

In-depth interviews follow a preplanned outline of topics

semi structured interviews

unstructured interviews

regardless of the degree of structure, interviewers must adapt throughout the interview

they can be flexible, cater questions to person to get them to respond better.
allows for follow up questions to be asked
Selection of Sample for an Interview
Random sample is rarely used, it takes too long, rude to ask random people to interview for a long time (weeks, hours)

should target more to be more effective

(i.e. want sample from junior class, send something out to ENTIRE class to get ppl to respond back and say yes, more likely to get a larger sample, otherwise you'd have to random sample over and over to get a sizable sample)
encourage participation through
incentives (financial or other)
assuring anonymity and confidentiality

might use a gatekeeper (someone that the sample frame will trust who endorses the researcher)

i.e. sample a sorority>>gatekeeper would be the president
Behavior in the Interview
Establish rapport with subjects by
considering how they will react to interview arrangements
developing an approach that will not violate their standards of social behavior (don't sit too close to people, uncomfortable)
treat people with respect

during the interview
maintain appropriate distance
maintain eye contact
go at an appropriate pace to allow the interviewee to reflect and elaborate, do not rush them

don't give too much affirmation on certain topics, it might sway people to keep on certain topics
Asking Questions and Recording Answers
Must plain main questions around (minimum) an outline of the interview topic
>questions should be short and to the point
>more details can be elicited through probing follow up questrions such as "tell me more"
>let ppl know youre listening

use tape recorders to capture the conversation so you can go back and listen
transcribe for later review and quotations
Interviewer Effect
change in participants behavior or responses as a resulty of the use of a particular interviewer

good idea to match interviewers to interviewees if possible on salient characteristics, esp on senstive topics
Participant and non participant observation
refers to several roles a researcher can adopt

Covert observer
overt observer
covert participant
overt participant

overt is when its out in the open, you aren't trying to hide it from people. Not trying to shout out to the world though either

covert is when you don't want people to know your role in participating/observing

close line between covert and necessity for informed consent

covert is used in more sensitive topics
Covert Observation
researcher observes others without participating in social interaction and does NOT identify herself as a researcher

really do not want anyone to know you are researching them

use of gatekeeper in order to gain some access

-Has little effect on social processes
Overt Observation
Researcher announces her role as a researcher

May cause reactive effects
---individuals may alter their behavior

don't want to use in sensitive cases

LOOK ONLINE
Entering the Field
Participant observation may require researcher to learn in advance how participants dress and behave so they can fit in

may require a gatekeeper
Developing and Maintaining Relationships in Participant Observation
develop a plausible explanation for your presence

Maintain the support of key individuals in groups under study

Unobtrusive and unassuming
---do not draw attention to yourself

be open but not fake

ask sensitive questions only of those who trust you have earned

reading:two trains of thought
-----review as much lit as possible to sensitize you to what you should look for
-----do not review too much past research or it may interfere in what you are able/willing to see in your own research
>>deciding whether it is a good idea to research or if this research will influence what you focus on so much so that you miss other things.
Sampling
No definitive appropriate sample size

Best to get some Theoretical Saturation
---when new observations or interviews begin to look much like those already sampled
(when u take notes, and they begin to look the same as your past notes, same trends etc.)

Use Purposive Sampling
---use of a sample frame related to the research topic
Observation Techniques
methods of collecting data by observing people, most typically in their natural settings
Participant Observation
observations performed by observers who take part in the activities they observe
Nonparticipant Observation
observation made by an observer who remains as aloof as possible from those observed
Controlled Observations
observations that involve clear decisions about what is to be observed
Thin Description
bare bone descriptions of acts
Thick Description
reports about behavior that provide a sense of things like the intentions motives and meanings behind the behavior
Complete Participant Role
being, or pretending to be a genuine participant in a situation one observes
Participant as Observer Role
being primarily a participant while admitting an observer status
Observer as Participant Role
being primarily a self professed observer, while occasionally participating in the situation
Complete Observer Role
Being an observer of a situation without becoming a part of it
Theoretical Saturation
the point where new interviewees or settings look a lot like interviewees or settings one has observed before
Purposive Sampling
a nonprobability sampling procedure that involves selecting elements based on the researchers judgement about which elements will facilitate his or her investigation
Account
a plausible and appealing explanation of the research that the researcher gives to prospective participants
Gatekeepers
Someone who can get a researcher into a setting
Visual Sociology
an approach to studying society and culture that employs images as a data source
Video Ethnography
the video recording of participants and the reviewing of the resulting footage for insight into social life
Informants
participants in a study situation who are interviewed for an in-depth understanding of the situation
Grounded Theory
theory derived from data in he course of a particular study
Demand Characteristics
characteristics that the observed take on simply as a result of being observed
Available data
data that are easily accessible to the researcher
Existing Statistics
summaries of data collected by large organizations
Secondary Data
research data that have been collected by someone else
Primary Data
data that the same researcher collects and uses
Ecological Fallacy
The fallacy of making inferences about certain types of individuals for information about groups that might not be exclusively composed of the individuals
Physical Traces
physical evidence left by humans in the course of their everyday lives
Erosion Measures
indicators of a populations activities created by its selective wear on its physical environment
Accretion Measures
indicators of a populations activities created by its deposits of materials
Unobtrusive Measures
indicators of interactions events or behaviors whose creation does not affect the actual data collected
Personal Records
records of private lives such as biographies letters diaries and essays
Experiments
most appropriate for answering research questions about the effect of a treatment or variavle whose values can be manipulated by the researcher

have at least 3 things
-at least 2 groups (unless pretest,posttest)
-variation in the independent variable BEFORE assessment of change in the dependent variable
-random asignment of the 2 or more comparison groups

medical research, testing out drug effects on others

can't systematically choose people for experiments
Experimental and Comparison Groups
Experimental group>>subjects who receive some treatment

Comparison Group>>group that you compare to experiment group
(receive no treatment, different treatment, or a placebo)

They differ in terms of one or more independent variables whose effects are being tested
Before and After Designs
absence of a comparison group
all cases exposed to treatment
comparison is done by comparing the pre-treatment with the post-test measures
Pretest post test
Look it up
Randomization
individuals selected to participate are randomly assigned to either control or experimental groups

makes the comparison group provide a good estimate of the outcomes that would have occurred if subjects who were exposed to the stimulus treatment had not been exposed

researcher cant determine the effects of a treatment if the comparison group differs from the experimental way

Double Blind Studies:
neither those conducting the research nor participants know who is in the control and who is in the experimental group.
>doesn't sway participants to behave differently depending on whether or not they believe they got the stimulus/treatment
>>>especially important in psychological and medical treatments
Matching
another technique used to compare experimental and comparison groups

may involve pairing persons on the basis of similarities in gender are race etc

problem that is very difficult to exactly match 2 persons or groups

may be used in research to examine the effects of stimuli in the real world.

Audit studies>>housing audits to figure out if there is a lending bias...make everyone the same sans one difference to study
Content Analysis Benefits
can be relatively inexpensive (not always)

allows research team to return to the units of observation repeatedly to garner info (older files more difficult)

less obtrusive, people are less likely to be disturbed, fewer ethical concerns, not really needing IRB approval when human beings are not involved (content analysis of the NY times)
Steps to a Content Analysis
ID a population of docs or other sources (appropriate to research question)

Determine units of analysis
units about whats collected
-individual issues of a newspaper or book could be more complicated like the protest reported in a newspaper article (instead of coding for every comedy show, code for every comedy show on monday nights)

determine units of observation

select a sample of units
(random sample or use stratified samples (local and national papers)

design coding procedures for variables to be measured
-reliable
-carefully constructed
-frequently looks like a survey
--researcher filling in a form rather than giving it to different participants