SOC110 Flashcards
final
What is the target population?
A cluster of people you are the most interested in.
It needs to be chosen scientifically
If too broad it will be difficult to have a representative sampling, and if too specific it will be difficult to find enough participants
What is the accessible population?
The accessible population is the group that a researcher can actually measure
Accessibility depends on the context of the research and the researcher’s resources; there is always a group inside the target pop. that is more accessible
Even in the accessible pop. not everyone will accept to participate to the study.
What is a sample?
A subset of the accessible pop.
The individuals who actually take part in research
Group from whom we gather data
*Sometimes, the subgroup does not benefit from the research results
In the eligible sample, what are inclusion criteria?
Guidelines about who can participate
- It ensures coherence of the sample with the research topic
-scientifically supported
In the eligible sample, what are the exclusion criteria?
Guidelines about who cannot participate
-ethical considerations
-reduce the gathering of unrelevant data
Give an example of inclusion and exclusion criteria with research about the effects of music on homeless people who have a substance abuse problem
Inclusion:
-experience problematic substance use
-experience homelessness
-Having been exposed to music at least once a week in the last 12 months
Exclusion:
-Experience psychotic, non-stable symptoms
-State of intoxication at the moment of research interview
-Experience music anhedonia
What can you do to make sure that people respect your eligibility criteria?
You can develop an eligibility questionnaire or use a questionnaire that was already done previously for things that are often studied, like substance use.
What is sampling?
Sampling is the process of selecting observations that will be analyzed for research purposes
sampling is the strategy to select the right sample and unit of observation. In human sciences it is often individuals or groups of individuals
sampling is about how you will reach out the accessible pop and how will you form the sample
It can be probabilistic or non-probabilistic
What are the characteristics of probabilistic sampling techniques?
Representativity: resembles the target pop. eg: a phenomenon is experienced by 70% women and 30% men, then you will have those numbers in the study
Generalizability: Results apply to the target pop. eg: rare things that are not experience by the target pop. should not be overseen in the sample.
Radom selection: Equal chances of being invited to take part in a study. eg: more coherent with quantitative study
Name the 4 probabilistic sampling techniques and exlpain them briefly
simple radom : Researcher randomly selects elements from sampling frame
systematic : researcher selects every xth element from sampling frame
stratified: Researcher creates subgroups then randomly select elements from each group
cluster: Researcher randomly selects clusters then randomly selects elements from selected clusters
Explain this probabilistic sampling technique: Systematic sampling
- Develop a sampling frame: list all members of the accessible pop.
- Randomly select individuals: using a systematic strategy. Every Xth person on the list. So, it is the same as the simple random technique, but you have an interval between every participants. So, you start by randomly choosing the starting point and then every “x” number for every participant. eg: you need 15 participants from the 100, so every 6 participants you take that person
comment: good for small scale research
Explain this probabilistic sampling technique: Simple random sample
- Develop a sampling frame: list all members of the accessible pop.
- Number potential participants
- Randomly select individuals: random number table, so you randomly select the first number and you go on
Limitation: Difficult to apply because it requires you to know every possible participant, so it it not feasible for big sample but it can be achieve with a smaller sample eg: Bishops teacher
comment: this technique is simple to understand, but hard to do. So it is rare to see it in research
Explain this probabilistic sampling technique: Stratefied sampling
- Divide the accessible pop in subgroups
- Draw a sample for each subgroup: either by random sample strategy or by systematic sampling strategy
comment: It is a good strategy if you want people with some characteristics represented eg: separate men, women, non-binary, and agender. This technique makes sure that you have minorities in your research, which can not always be the case in random strategies
Explain this probabilistic sampling technique: Cluster sampling
- Divide the accessible pop in subgroups: subgroups have the same characteristics like living in Sherbrooke
- Select random subgroups
- Sample all members of the selected subgroups
Comment: This technique is useful when you do not have access to all the information about the potential pop.
The sample might be heterogenous since participants might be more different from one another.
This technique is use for larger pop.
What are the strengths and limitations of probabilistic sampling techniques?
Strengths:
-Representative samples
-Generalizability of the research results
-Less likely to be biased
Limitations:
-Resources needed may limit its feasibility (time, costs)
-Requires a high level of skill and experience, so it is best accomplished by an experienced research team
-May not represent minority groups
What are the characteristics of non-probabilistic sampling techniques?
Used to…
-Describe, explore or examine small groups
-Explore a phenomenon that is unknown/misunderstood
-Understand the perspective of particular individuals
-Develop an in-depth understanding of a phenomenon
Comment:
-The sampling is not based on statistical techniques
-mostly used in qualitative techniques
-you recruit people by identifying how you can reach people
-Made to recruit people with a very specific “condition” that would be difficult to find in the general pop.
Name the 4 non-probabilistic sampling techniques
Purposive sampling
Quota sampling
Convenience sampling
Snowball
Explain this non-probabilistic sampling technique: Purposive sampling
- Identify the perspectives sought regarding the topic of interest: and what is the angle that you want to talk about the topic
- Identify diversification criteria: e.g., mental health problems, find people who were hospitalized in mental health facilities or hospitals, age range, gender, etc.; once you develop different characteristics, you need to find people.
- Seek individuals who respect those criteria and who can discuss these perspectives
Comment: this technique helps have a diverse perspective regarding the topic of interest
Explain this non-probabilistic sampling technique: Snowball sampling
Rely on initial participants to identify potential participants
Comments :
Useful to study stigmatized, hard to reach and socially excluded groups
Useful to reach groups unfamiliar to the researcher or new researcher
Explain this non-probabilistic sampling technique: Quota sampling
- Identify categories important to the study: eg: sex or gender
- Subgroups are created based on each category: eg: gender = subgroups of men, women and non-binary
- Select individuals from each subgroup
Comment:
- the number of participants that we want from each subgroup is called quota
-Selection of participants is not random and it is not representative of the general pop.
What can influence the selection of research participants? Sampling biaises
Access to invitation to take part in research: the sample will be influence by where and who can see the study eg: where is the poster, who will be able to reach the poster, if it is by internet older people might not be able to reach it
Recruitment strategies: Participants might have similar interests and similar ways of thinking
Personal interest from the participants in a research topic: often people who participate are truly interested and might taint the results
What are the strengths and limitations of non-probabilistic techniques?
Strengths:
-Fast and simple: it can be used by newer researcher
-Inexpensive
-Targets people with specific experience: which ensures that the participants have something to say about the research
Limitations:
-More at risk of sampling biaises
-Lack of representativity with the target population: people do not have the same chances to be part of the study which makes it possible that it is not generalizable to the general population
-Results cannot be generalized to the target population
Comment: these techniques are more used in qualitative research
Explain this non-probabilistic sampling technique: Convenience sampling (quantitative and qualitative)
Select individuals who are easily accessible: physical proximity, availability of the researcher, available resources
Comment: journalists use this strategy.
Useful in exploratory research: it takes less time to prepare and it is always a feasible strategy
What is a Survey?
The most use data collection tool
Quantitative method and sometimes qualitative
It has predetermined questions: -that are developed in advance with predetermined answers
-close-ended questions
-printed or online, so the survey can be done in person or online
Self-reported information: the participants answer the questions without the researcher’s interference.
What are the objectives of surveys?
Large amounts of data
Gather data quickly
Get details about a large population
To describe the characteristics of a population or a phenomenon: eg: the proportion and prevalence of people concerned with the cost of living
To investigate the opinions and behaviours of a population: eg: journalists have a new subject so they do surveys to have fast answers. It also helps characterize the population by having questions about sex, gender, age, etc.
Useful when you have limited resources
What are the differences and a similarity between a survey and a questionnaire?
Survey:
-collect data about a population
-Research method process
-Statistical analyses
-Draw conclusions
-Takes time to answer
Questionnaire:
-Collest data about one person
-Tool to gather information
-No formal analyses
-Will not answer generalizations about the population
-Only an instrument used to gather some information eg: can use it so know if participants have the prerequired necessities to be part of the research
-Can be quickly administered
Both have pre-determined questions
How can a survey be administered?
Self-administered:
-Hard copy:
-Online: Sometimes it is hard to see every participant; therefore, having the possibility to do it online is a great help. Also, the data can be gathered online through an online platform. However, having people take the time during the day to do it can be hard.
Administered by the researcher:
-by the form of an interview, so the researcher read questions and voice options.
What are the strengths of Surveys?
Method
-Gather lots of information in short amounts of time
-Reaches large population: you can do it online, so it is easier to reach diverse profiles
-Cost-effective: also less time consuming
-Reliable and consistent: ensures that the method is standard and that the questions are asked in the same manners
-Versatile: can adapt to many social contexts
Sample
-Representative
-Generalizable results that help draw conclusions about a larger population
What are the limitations of Surveys?
-Data collection rests on a single instrument: if there is an issue with a question you cannot go back, it also forces participants to choose between predetermined answers which might not represent the POV
-Validity of the results: fails to capture the true essence of the participants or they don’t understand the question which is going to affect the results.
-Misinterpretation of the questions: if the questions were interpreted differently it might not be generalizable data
-Risks of missing data
Participants
-Social desirability: tendency to answer questions to make themselves look good, some might lie or exaggerate certain things
-Biais in the persons who choose to participate
Name the different types of Surveys
-Cross-sectional surveys
-longitudinal surveys: trend, panel, cohort
-Retrospective surveys
What are retrospective surveys?
Administered only once, but ask participants about a series of past events
What are cross-sectional surveys?
Administered once
At a particular point in time: which gives a picture of a certain phenomenon at a certain point in time
Does not consider evolution in phenomena and population:
-gives a stagnant picture
-difficult to generalize
-difficult to know if the results apply long after the assessment
What are longitudinal surveys
Repeated observations: eg. to know what are the substance use over time and the frequency
Assessment over long period of time: from multiple months to years
Follow the same individuals over time:
-Risk of attrition: might lose participants over time. Generally, 70 to 80% must remain in the study for it to be accurate. Depending on the participants it might be hard to keep them. We accept certain attrition rates like for homeless participants they are easily “lost”
In longitudinal surveys, what are trend surveys?
Evolution and changes over time
-opinion
-characteristics
-behaviours
Administer the same questions at different moments:
-the interest are trends, not specific groups
-Respondants are not necessarily the same on each occasion: as long as the participants are representative of the general population, it does not matter
In longitudinal surveys, what are cohort surveys?
Track change over time: need to use the same question over time
Regularly survey people of the same category of interest
-But not necessarily the sample individuals: not necessarily the same participants but they need to have a particular characteristic eg. physical and mental health of people born in 2003
In longitudinal surveys, what are panel surveys?
An audience is surveyed multiple times: each set of surveys is called a wave
-track changes in behaviours, attitudes, thoughts, etc.: Therefore, they need to have very similar questions in the different waves
-relationships between variables: can help identify relationships between natural and individual variables
The same persons participate in each assessment:
-difficult and costly: You need to be aware of each participant’s information. High risk of attrition
-requires lots of resources
How can you design effective survey questions?
Define the goal and purpose of the survey
Get a clear idea of what you want to learn
Identify relevant questions:
-begin with a brainstorm of potential questions
-work with a team
-organize the potential questions: eg. you can do it by themes so you can notice if you are lacking some questions or if you have similar ones
Comment:
-Should never ask the most sensible question in the beginning
-there are no clear-cut manner to answer questions, but it should be done intelligently so it flows
-the length of the survey should be linked with the topic
-all questions should be important and not redundant
What are the characteristics of good survey questions?
Clear and unambiguous: only one idea per question and it should not include multiple point
Appropriate wording: depending on the population, the question should be adapted on their capacity to understand them
Minimize biais: we want to avoid questions that would give the impression that their is a right and wrong answer
-As neutral as possible
-Want to avoid social desirability
Obtain feedback/pre-test the questions: this is a normal process to develop the best question
What are the characteristics of good response options?
Generally one possible answer
Generally close-ended: the participants should not creat their own answer
-mutually exclusive: an idea can not be present in two choice eg. What is your age 13-15, 16-18 not 13-15, 14-20
Example:
-yes /no questions
-likert scales
-level of agreement
What are experiments?
Quantitative data collection methods
Test hypotheses in controlled conditions:
-laboratory design for a certain type of research eg. music, soundproofing, good headphones
-free from disruption and biais from outside
-provide the same conditions for each participant
-More used in psychology
What are the goals of an experiment?
Determine causal relationships between variables
Explain relationships between variables
What are the characteristics of experimental research?
Control over variables: strict control over the variable that we study. It is important to identify the variables. Manipulations are the same for each participant
Control over environment: it must be the same for each participant. eg. conducting a study in a private and calm space, if those are not possible you need to make sure that the participants are exposed to the same conditions
Control over participants: have clear eligibility criteria to make sure that we have a homogenous group.
-random assignment
Explain the independent variable in an experiment
Independant variable: The one we want to study
-factor: also called independent variable, can have more than one but not too many
-level value of an independent variable: eg. if you look at the effect of relaxing and exciting background music. independent variable is the background music, but it has two levels relaxing and exciting
Explain the dependant variables in an experiment
Outcome: also called dependant variable, this one gives us the results using data collection tools
Explain the confounding variables in an experiment
The variable that is not of interest, it is not measured but might influence the causal relationships that we are looking at. We try to eliminate them as much as possible. eg. you are doing an experience on music, but one of your participants suffers from a lack of enjoyment from music. This can make you think that the independent variable is not good, but actually, it is just the conditions
What are three important variables in experiments and explain them?
Independant (Factor)
Dependant (Outcome)
Confounding
What is the difference between the experimental and control groups in experiments?
Experimental group:
-Exposed to the stimulus
-Test group
Control group:
-Is not exposed to the stimulus
-Meant to be a point of comparison: it allows the researcher to have a conclusion.
Both groups are tested the same way
What are the aims of random assignment in experiments?
*Each participant as equal chance of being assigned to each group
Pre-test equivalence of the groups
Control over individual variables
*Each participant as equal chance of being assigned to each group. Groups should be similar and the only difference should be the presence or absence of stimulus. One way to do it is by assigning numbers to participants and even numbers are the control and odd numbers are doing the experiment
it is never possible to control every variable but by doing random assignment it reduces it
What are the two kinds of experiments?
Within-subjects/within group and between-subjects/between group
