exam 2 Flashcards
know it all
What is non-probability sampling?
Sample selected using non-random sampling. All forms of sampling not conducted according to the canons of probability sampling.
What is characteristic of probability sampling?
Sample findings are generalizable to the population they’re drawn from.
What is a sampling frame?
A list of elements in the population. the source material or device from which a sample is drawn.
What are the 4 types of probability sampling?
Simple random sampling
Stratified random sampling
Multi-stage cluster sampling
Systematic sampling
What are the characteristics and process of simple random sampling?
Everyone has an equal chance of being selected
- Define the population
- Devise a sampling frame
- Decide on sample size
- List population and assign them consecutive numbers
- Pick numbers at random
What are the characteristics and process of stratified random sampling?
-Units are randomly selected from a population that has already been divided into sub-groups
-Assure subgroups are represented proportionately
-Requires that relevant criteria for stratification be known in advance
What are the characteristics and process of multi-stage cluster sampling?
-Used for large populations
-Involves 2+ stages
-Divide population into groups
-Randomly sample a number of groups
-From each of the groups sample a number of people
-Stratified cluster sampling can ensure clusters are representative of entire population
What are some pros and cons of multi-stage cluster sampling?
+ Solves problem of inadequate sampling frame
+ Cost effective
- Technical issues
- Clusters are not all the same size
What is heterogeneity of a population?
An issue pertaining to sample size. Populations with higher heterogeneity lead to more varied samples, and thus require a larger sample size.
What is characteristic of absolute sample size?
-More reliable and generalizable
-More important than relativity
-Sampling error decreases as sample size increases
How much does sampling error decrease with each increase in standard sample size?
1/2->1/3->1/4->1/5
What are the standard sample sizes?
-100
-400
-900
-1600
-2500
What is characteristic of relative sample size?
-Sample size is representative of population size
-Less reliable than absolute samples because different sample sizes have different confidence intervals
What is non-response?
An issue with sample size
- If there is a common issue to the non-responders that brings them to differ significantly from participants
How does “kind of analysis” pose an issue with sample size?
-Sample size may need to vary to accommodate different kinds of analysis
- Smaller groups may need to be oversampled in order to draw meaningful comparisons
What is time sampling?
-Aspect of structured observation
-Observer records whatever is happening at a given time in a given place
What is characteristic of time sampling?
-Observation periods may have to be randomly selected
-Observation times can be randomly scattered across a period of time
What is place sampling?
-Random sampling at specific places
-Goes hand in hand with time sampling and behaviour sampling
What is behaviour sampling?
Researcher may want to observe every nth (5th, 7th, 11th, whatever) interaction between people at a given time and place
What is sampling error?
Errors of estimation that occur due to discrepancies between sample and population.
-Impossible to eliminate completely, but can be mitigated by large sample size and random sampling
How is sampling error measured?
Standard error of the mean
- 95% of sample lies within +/-1.96 standard errors of the mean
What are the characteristics and process of snowball sampling?
-Similar to convenience sampling
-Unlikely to be representative of population
-Mostly qualitative
1. Researcher makes initial contact with small group of people
2. Researcher uses said group of people to reach more people
What are the 4 types of non-probability sampling?
Snowball sampling
Quota sampling
Purposive sampling
Convenience sampling
What are the characteristics and process of convenience sampling?
Researcher chooses sample simply because they are easily accessible
- High response rate
- Low generalizability
- Useful for test/pilot studies
- Ethically questionable
What are the characteristics and process of quota sampling?
- Does not use random sampling to fill quotas
- Individuals are selected on the basis of quotas/criteria
- Quotas are filled until they are proportional to the population
What some pros and cons of quota sampling?
+ Cheap
+ Can be conducted quickly
+ Good for pilot tests/exploratory research
- Not likely to actually be representative because researchers choose participants
- Judgement about eligibility may incorrect
- Amount by which a characteristic differs from the population cannot be determined because it is inappropriate to calculate the standard error of the mean
What is purposive sampling?
Cases are selected on the basis of their ability to provide information relevant to the topic of study
What are the characteristics of unstructured interviews?
- Interviewer free to explore any topic
- Memory aid used at most
- Informal questioning/content/phrasing
- Sequencing of questions can vary from interview to interview
- Conversational
What are the limits to generalization?
- Findings can only be generalized to the population they were taken from
- Findings may be time specific, rendered obsolete with time
What are the differences between quantitative and qualitative research?
- Numbers v. Words
- Means of social analysis - POV of researcher v. POV of participant
- Through what lens is the research structured, who provides point of orientation? - Researcher is distant v. Researcher is close
- How much contact do researchers have with their subjects? - Theory and concepts tested in research v. Theory and concepts developed through research
- Chronology of theory/concepts & data - Structured v. Unstructured
- Maximize validity/reliability v. know subjects intimately and allow theory to develop - Generalizable knowledge v. Contextual understanding
- Understanding the whole of society v. understanding the individuals being studied - Hard, reliable data v. Deep, rich data
- Robust/Precise/Unambiguous v. Rich/Detailed/Nuanced - Macro v. Micro
- Scale of researched group - Behaviour v. Meaning
- Focus on behaviour v. meaning of behaviour - Artificial settings v. Natural settings
- Where is research conducted?
What is a direct question?
Answered with interviewee perceptions, best left until later in the interview so as not to affect the procession of the interview.
What are the characteristics of semi-structured interviews?
- Interview guide (list of fairly specific questions) is used
- Interviewee given leeway to respond
- Questions may not follow exact order
- Questions may be changed/added contextually
What are the characteristics of structured interviews?
All respondents are asked the exact same questions with the help of a formal interview schedule
What are the characteristics of an in-depth interview?
- Open-ended questions
- Response flexibility
- One-on-one interaction
- Contextual understanding
What is an indirect question?
Answered with interviewee perceptions of others
What is an open-ended question and what are its strengths and weaknesses?
A question that allows respondents to answer however they want.
+ Encourages respondents to share their thoughts and feelings
+ Useful for in depth analysis
- Time consuming
- Limited standardization
What is true of natural settings?
- Not a lab
- In the field
- No measurement instruments used
- Talk to people and observe their behaviour in the context being studied
- Face to face contact
What is a closed-ended question and what are its strengths and weaknesses?
A question that gives respondents a set list of answers to choose from
+ Easy to collect structured data
- Limited insight
- Response bias
What is inferred by “mixed methods” research?
Use of both qualitative and quantitative methods in a study.
What are the arguments against multi-strategy research?
- Particular research methods are associated with particular epistemological and ontological positions (embedded methods argument)
- Qualitative and quantitative research methods are inherently incompatible on epistemological and ontological grounds (paradigm argument)
How can researchers gain access to closed settings?
- Use friends/colleagues
- Use gatekeepers
- Offer something in return
What is ethnography?
A study of people and their culture in naturally occurring settings.
What is true of ethnography?
- Unstructured observation such as in-depth interviews
- Describes the life of a community from participant POV without outside interference
- Researchers are immersed in a particular setting for a long time
- Can also refer to a written account about particular qualitative research
How can researchers gain access to open settings?
Similarly to closed settings, but may have to answer pointed questions.
What is structured observation?
Researchers follow rules of what to look for as well as when and where, and how to record their findings.
What is participant observation?
An observational aspect of ethnography that includes:
- Observing behaviour
- Asking questions
- Analyzing conversations
- Field immersion
What is the participant-as-observer approach and what are its associated risks?
- Researcher adopts a role in the group being studied
- Participants are aware who the researcher is
- Risk of reactivity
What is the complete participant approach and what are its associated risks?
- Covert operation, same as participant-as-observer but the group is unaware
- Researcher adopts a secret role in the group
- Risk adopting behaviours of the group (going native)
- Risk developing distaste for the research subjects
What is the observer-as-participant approach and what are its associated risks?
- Researcher observes from the edge of the group, participating minimally
- Risk of reactivity
- Risks misinterpretation of observed activity
What is the reactive effect (reactivity)?
The effect of subjects knowing they are being studied.