Midterm Flashcards
Issues connected to ethics
- no coercion
- no persuasions (figure of authority)
- voluntary participation
- research should not add risk
- participants can skip sections in the questionnaire/surveys, refuse to answer all questions during interviews
- benefits should be provided in any case
- treating people fairly & with respect
- discourage people from withdrawing if they decide not to do so
Main Principles in Ethics
protect participants from harm
- research with human participants requires Ethics approval
- balance between potential harm and benefits of research
Ethics Review Board (ERB)
- usually, universities have two Ethics Review Boards
1) Full Research Ethics Board - they receive greater than minimal risk research, committee of faculty members, REB, community & other members
2) Delegated REV Review - they have minimal risk review, individual will review (REB officers)
Tri-Council Policy Statement Core Principles
concern for human welfare - means seeking concrete ways to promote the physical and mental wellbeing of participants and protect them from harm justice - treat people fairly and equitable respect for persons - intrinsic values of human beings - incorporates dual moral obligation to respect autonomy and protect those who have limited autonomy (vulnerable groups, disabilities, PTSD) - people should choose freely
Hypothesis
- 2 variables
- expresses relationships between variables
- expressed as prediction or expected future outcome
- logically linked to research question and theory
- possible to test it to see if it is false or true
null hypothesis
no relationship or no effect between variables
alternative hypothesis
states that independent variable has an effect on dependent variables
sampling
process of selecting observations in research
probability sampling
involves random selecting
probability sampling techniques
simple random sampling
simple random sampling
- researcher assigns a single number to each element in the sampling frame without skipping any number in the process
- a table of random numbers is then used to select elements for the sample
limitations:
- entire list of population should be accessible but not usually the case
- could be quite extensive and expensive
small random sample may not be representative of the population of specific ethnic groups
probability sampling techniques
systematic random sampling
systematic random sampling
- design requires a list of certain types of population to be researched
systematically takes every 10th, 20th, element into sample
weakness
- periodicity of people in the list could be done in such a way that sampling will represent only specific gender, race, age, etc.
probability sampling techniques
stratified sampling
stratified sampling
- method for obtaining a greater degree of representativeness
- it is necessary to divide the population into different subgroups or strata, & then randomly selects final subjects proportionally from the different strata
- stratification is process of grouping members of a population into relatively homogenous strata before sampling (could be race, gender, age, socioeconomic status)
probability sampling techniques
cluster sampling
cluster sampling
- one of most efficient ways of sampling
- may be used when not possible to compile a list of all elements that compose the larger population
- useful when clusters occur naturally such as in schools, universities, hospitals. cluster becomes a unit of analysis and therefore considered a representative of population
non-probability sampling
Not random sampling
Selective
non-probability sampling techniques
quota sampling
quota sampling
- gather data from individuals in the same proportion as they are represented in population, (gender, sex, age, race, education)
- could be face-to-face, email, phone, mail
weakness
- allows research to decide whom to include or not include into study, therefore creates biases
non-probability sampling techniques
reliance on available subjects
reliance on available subjects
- accidental sampling, procedure that relies on available subjects
weakness
- findings would represent the opinions only of people with characteristics of those passing the sampling point of specified times