Week 4 Quantitative Research Flashcards
What is a research design?
To structure the research, show all of the major parts of research project, samples, measures, treatments/interventions to address research question
Key Research Design features
- Intervention
- Comparison
- Control over confounding variables
- Time frames
- Relative timing
- Location
What are the key dimensions of design?
- Experimental vs non experimental (Control)
- Cross sectional vs longitudinal (How often)
- Retrospective vs Prospective (When)
What are the types of experimental design?
RCT & Quasi to identify causal relationship
What are the types of non experimental design?
Descriptive, correlational & comparative
What are the characteristics of true experiments?
- Intervention
- Control
- Randomisation
What is a true experimental design?
RCT: the gold standard for testing causal relationship: pre-test post-test with randomisation
What is a quasi experimental design?
Non-equivalent pre-test post test design
What is randomisation in RCT?
Allocation of subjects to control/exp groups is done by chance, without the ability to predict who is in the group
Every subject has equal chance
Minimise selection bias
What are the different steps of randomisation?
- Participant recruitment
- Participant allocation
- Intervention administration
- Outcome assessment
What are the key characteristics of Quasi?
- Intervention
- Absence of randomisation
- Control (yes/no)
What are the strengths & weaknesses of Quasi?
Practical - not all participants would want to be randomised in clinical trials
Cause & effect rs is weakened
Absence of randomisation & control group
When is non experimental research used?
When RQ cannot be addressed by experimental/quasi with no control over IV
What are the key characteristics of non experimental?
- IV inherently cannot be manipulated
- Unethical to manipulate IV
- Research aims to investigate a natural phenomenon
What are the types of non experimental designs?
- Descriptive
- Descriptive correlational
- Comparative descriptive
Descriptive study in non experimental
To observe, describe & document a situation
Descriptive correlational in non experimental
Examine relationship between variables
To describe & investigate without attempting to infer causal rs
Comparative descriptive
Compare variables between 2 or more samples on 1 or more variables often at a single point of time
What is a sample?
Representative of a subset of population
What are the key considerations for study sample?
- Representativeness - extent that sample is similar to accessible population
- Sampling bias - systematic over or under representation of population
Why probability sampling?
To reduce risk of bias & enhance internal & external validity
What are the types of sampling bias?
- Non response bias
- Systematic bias
- Attrition bias
What are the types of probability sampling designs?
- Simple random
- Stratified
- Systematic
What is stratified sampling?
Process of selecting sample to identify subgroups in a population that are represented in the sample
Adv + Disadv of stratified
Prevents possibility that some subgroups are under represented
Require more labour
Impossible if information is unavailable
What is systematic sampling?
Process of selecting every nth case from pre existing list
What are the types of non probability sampling?
- Convenience
- Consecutive
- Snowball
What is consecutive sampling?
Recruit entire accessible population within defined period of time (set time frame, e.g 3 months then stop)