exam 2 rm sociology

Sampling

•  the process of deciding what or whom to observe when you cannot observe and analyze everything or everyone

why do we use sampling?

• Heterogenous opinions

• Limited resources 

Probability sampling

•  the process of selecting a sample based on probability, typically involving some random selection mechanism 

Benefits of probability sampling

Representativeness, Generalizable, Cost efficient, Avoids bias, Provides a calculate margin or error

Representativeness

• when a sample has the same distribution of characteristics as the population from which it was selected

Generalizable

• the degree to which you can apple the results of your study to a broader context 

margin or error

• the amount of uncertainty in an estimate 

Benefits of large samples

smaller margins of error, more closely approximates the true population

How to draw a probability sample

• Identify population: the group that we’re interested in generalizing about 

  • Establish a sampling frame: the list of units composing your population

    • Use random selection to sample your elements, or the individual units comprising your sample 

• Types of probability samples 

simple random, cluster, systematic, stratified

simple random sample

Equivalent to pulling names out of a hat

Systematic sample

E.g. every 4th person from the sampling frame

Cluster sample

Define the population, cluster the population, randomly select clusters

Stratified

Divide the population into groups and then sample members from those groups in strategic ways

Nonprobability sampling

a sample that is not drawn using a method of random selection

nonprobability sampling is generalizable or not generalizable

not generalizable

types of nonprobability sampling

convenience, purposive, quota, snowball

convenience sampling

subjects selected due to convenience, only justified if the people passing are your target sample or if its exploratory

Purposive sampling

Units are selected based on which will be the most useful or insightful; often done when looking for people with specific characteristics

Types of cases purposive samples examine

typical, extreme, deviant, contrasting, important

Extreme

the most intense, but not necessarily going against the norm

Important

big players, who are the most important in the phenomena we wish to study

Deviant

anomaly, opposite to the norm

Contrasting

usually a few cases, comparative case studies, the cases differ and create a puzzle when next to one other, different outcomes but identical stimuli

Sampling for range

maximizing respondents ranges of experiences with the phenomena under study

Quota

Units are selected on the basis of pre-specified characteristics, so that the sample will have the same distribution of characteristics assumed to exist in the population being studied

Snowball

Interviewees suggest additional people for interviewing

Key informants

the first point of contact a researcher has with his/her study population

Benefits of Non Representative Samples

rich information, causal mechanisms

when do we stop collecting data

when we reach saturation

Saturation

when additional data fails to yield new insights and simply reinforce what the researcher already knows

Causal mechanisms

the why to relationships

4 types of surveys (from most to least expensive)

Face-to-face interviews

Telephone surveys

Mail surveys (self-administered questionnaires)

Internet based surveys

face-to-face interview

interviewer meets in person with the respondent and asks a series of questions aloud

advantage to face-to-face interviews

high completion rates

disadvantages to face-to-face interviews

interview effects, social desirability bias, cost

interview effects

the characteristics of the interview will affect how the respondent answers the interviews questions

social desirability bias

study participants report positively valued behaviors and attitudes rather than giving truthful responses

telephone surveys

interview administered over telephone

advantages to telephone surveys

higher completion rates, data quality, more cost efficient than face-to-face interviews, fewer interviewer effects

disadvantages to telephone interviews

response rate, less rapport, response fatigue, sampling bias

mail surveys (self administered questionnaire)

Individual independently completes questionnaire themselves

advantages of mail surveys (self administered questionnaire)

low cost, low bias, convenient

disadvantages of mail surveys (self administered questionnaire)

low response and completion rates

internet based survey

respondent answers survey online

advantages of internet based surveys

cheap, easy access and administration, fast

disadvantages of internet based surveys

sampling bias, response rate

ordering

What questions are placed where in the survey

how to order

establish rapport, start with easy questions, controversial questions in the middle, end with demographics, avoid monotony

how to avoid monotony

alternate topics, vary response options, quality controls

monotony

boredom

quality controls in a survey

speeders, straight-liners, attention checks

Order effects

when the order in which questions appear biases the responses

avoid order effects and _____ when ordering

priming

priming effects

when exposure to a particular image, word, or feeling shapes how respondents think and feel in the immediate aftermath

the goal of the format of a survey

Goal is for respondent to have a streamlined, confusion-free experience

how should a survey be formatted

Consistent

Uncluttered

Clear

Intuitive

Composite measures

combines multiple survey items to create a single value that captures a multifaceted concept

Filter question(s)

a question or series of questions associated with conditional response to a prior questions

Split ballot design

one half of the sample receives one module, and one half receives another, determined by random assignment

survey design

the process of creating, formatting, and stylizing the survey; critical for ensuring reliable and valid research

you should keep questions ___, ____, and _____

short, consistent, and easy

Double barreled question

asks about 2 r more ideas or concepts in a single question

avoid questions that require _____

expertise

Negation

contains negative words such as “no” or “without”

double negative

contains 2 negative words

research questions should not include a _______ (negative wording)

negation

avoid ___ wording in surveys questions

unclear

Leading question

influences or guides respondents toward a particular answer, often implying a preferred response

avoid asking _____ questions in surveys (asking people a question with the intention of getting them to answer in a certain way)

leading

avoid asking survey questions that requires ____ time

excess

recall bias

when respondents do not accurately remember a past event or experience or leave out details when reporting about them

avoid asking survey questions that include _____ topics

sensitive

Causality

one variable causes another

3 things needed for causality

Correlation

Time-ordering

Non-spuriousness

Non-spuriousness

unaccounted for other variable cannot be influencing the correlation

Experiment

the researcher manipulates one or more independent variables to determine the effect(s) on a dependent variable

independent variable

cause

dependent variable

effect

experiments seek to determine ____

“Does x cause y”

experiments are good at finding out causal relationships because of these 2 reasons

eliminates all other explanations, isolates the effect of x on y