Midterm Flashcards
What is a research paradigm?
A distinct set of concepts, theories, and research methods that define how knowledge is understood in a field.
Positivism
know where knowledge comes from (facts, measurable numbers). Deductive approach
Ontology
The study of reality and what exists.
Epistemology
The study of knowledge and how we acquire it.
What were the consequences of the Wakefield study?
Small sample size (no statistical significance).
No randomization (high bias).
Fraudulent data selection.
Published in 1998; retracted in 2010.
* Led to decreased immunization rates.
* Measles outbreaks increased worldwide.
* Wakefield lost his medical license.
Linked MMR to autsim
What is “fast thinking” (Kahneman)?
Unconscious, emotional, instinctive decision-making. Leads to snap judgments & bias.
What is “slow thinking”?
Conscious, deliberative, and rational thought.
Clinical question
Arises from a clinical context/situation.
Research question
Arises from a gap in existing knowledge.
What is the PICO(T) framework for structuring a clinical question?
P = Population
I = Intervention
C = Control
O = Outcome
T = Time (optional)
What are the two types of research evidence?
Primary evidence: Original research data (quantitative or qualitative).
Secondary evidence: Summarized research (e.g., systematic reviews, clinical guidelines).
Primary evidence
Original research data (quantitative or qualitative).
Secondary evidence
Summarized research (e.g., systematic reviews, clinical guidelines).
What are four attributes of rigorous quantitative evidence?
- Reliability
- Validity
- Reproducibility
- Generalisability.
Reliability:
Produces stable and consistent results.
Validity
Measures what it claims to measure.
Reproducibility
Yields consistent results using the same methods.
Generalisability:
Applies to broader populations or contexts.
What are the key elements of a well-posed research question?
Describes variables.
Specifies the population.
Examines testable relationships.
What is a hypothesis?
Hypothesis is a suggested idea or an educated guess or a proposed explanation made based on limited evidence, serving as a starting point for further study.
What are the characteristics of a good hypothesis?
Specifies variables.
Defines the population.
Predicts outcomes.
Can be tested.
What is a variable in research?
Any characteristic, number, or quantity that can be measured or counted.
Independent variable (X):
The factor being manipulated or observed as the cause of what you changed. Yes or No question
Dependent variable (Y):
The outcome being measured. Cause or effect
What are the four main types of hypotheses?
- causal
- simple
- complex
- statistical
causal
Suggests cause-and-effect.
simple
Examines the relationship between two variables.
complex
Examines relationships among three or more variables.
statistical (null hypothosis)
States no relationship between variables.
Directional
Predicts the direction of the relationship.
Non-directional
States a relationship exists but does not specify the direction.
What are key elements of a research purpose statement?
- Identifies a problem linked to existing evidence.
- Defines a gap in knowledge.
- Explains why the research matters.
- Includes objectives and hypothesis.
What makes a hypothesis testable?
It can be measured using quantitative methods.
What are the three key components of an experimental design?
Randomization
Control group
Manipulation
Manipulation
The researcher actively changes an independent variable.
Control group
A comparison group to measure effects.
Randomization
Assigning participants randomly to groups.
What is a randomized controlled trial (RCT)?
to evaluate the effectiveness of an intervention, treatment, or program by randomly assigning participants into two or more groups. It is considered the gold standard for clinical and public health research because it minimizes bias.
Simple RCT
Single intervention (e.g., medication).
Complex RCT
Multiple interacting components (e.g., behavioral interventions).
Why is randomization important in RCTs?
Ensures equal probability of group assignment.
Reduces selection bias.
Creates comparable groups at the start.
What is blinding, and why is it important?
Blinding prevents bias by keeping participants or researchers unaware of group assignments.
Single-blind
Participants don’t know their group.
Double-blind
Both participants and researchers don’t know.
What is allocation concealment in RCTs?
Hiding group assignments before and during allocation to prevent selection bias.
Are RCTs always the best research design?
No,
Over-reliance on RCTs (discounting other evidence).
Ethical concerns (some treatments can’t be randomly assigned).
Generalizability issues (may not apply to all populations).
Hawthorne Effect (people change behavior when observed).
What is the Hawthorne Effect?
When participants change their behavior because they know they are being observed.
What is confounding in research?
When an external factors affect cause/outcome, may be hard to read results or could be missleading
What is clinical equipoise
It means there is no known “better” treatment at the time of the study, uncertainty of what works better.
What is the intention-to-treat (ITT) principle?
Data is analyzed based on initial treatment assignment, not whether participants completed the treatment.
What is loss to follow-up, and why is it a problem?
Participants drop out before the study ends, which can bias results.
What is a non-experimental research design?
A study that observes relationships between naturally occurring variables without manipulating them.
What are the two main types of non-experimental designs?
Descriptive (focus on frequency) and analytical (quantifies relationships between variables).
What is a survey (cross-sectional descriptive) study?
is a research method that gathers information from people at one specific time to understand trends, opinions, or health conditions.
What is the key limitation of survey studies?
They focus on the larger scale of things rather than depth of the information, can only make statements about asked questions
What is a cohort study?
A study that follows a group of people over time to examine associations between exposures and outcomes.
What is the difference between retrospective and prospective cohort studies?
Retrospective studies use existing data, while prospective studies follow participants over time.
What is a key advantage of cohort studies and some cons?
They are useful for assessing risk factors and disease prognosis.
Expensive and time consuming
What is a case-control study?
A study that compares individuals with a disease (cases) to those without (controls) to identify potential risk factors for that disease. (bacon makes you fat)
What is the main outcome measure in case-control studies?
They look back to assess whether there is a statistically significant difference in the rates of exposure to a defined risk factor between the groups.
What is a limitation of case-control studies?
They are susceptible to recall bias and cannot establish risk.
What is a cross-sectional (observational analytic) study?
A study that measures exposure and outcome at the same time to assess prevalence.
Why are cross-sectional studies considered “ethically safe”?
They do not involve interventions or prolonged follow-up, reducing risk to participants.
What is a systematic review?
a method to find answers to a certain research question, by collecting all available studies related to that question and reviewing and analyzing their results.
What is the Cochrane Collaboration?
Systematic reviews that help guide health care decisions. Like BPG
What are the types of Cochrane reviews?
Intervention, diagnostic test accuracy, methodology, qualitative, and prognosis reviews.
Cochrane review Intervention
asses benefits and harm
Cochrane Review diagnostic test accuracy
review how they are performed and how accurate they are
Cochrane review methodology
reviews how clinical trials are ran
Cochrane Review qualitative reviews
addresses questions of intervention rather than effcetiveness
Cochrane review Prognosis
adress the probable cause of a disease with future outcomes
What is a population in research?
The entire group of individuals that a study aims to learn about
What is testability
measured and tested using number that can be statistically verified to ensure its real data
What is a sampling frame?
A specific list, database, or source from which a sample is recruited (e.g., a clinic list, OHIP database).
Homogeneous sample
all the items in the sample are chosen because they have similar or identical traits
Heterogeneous sample
occur when the items have differences.
What are inclusion criteria in research?
Key features of the target population used to define eligibility for a study, such as socio-demographic, clinical, or geographic characteristics.
What are the two main types of sampling techniques?
Probability sampling (random selection) and Non-probability sampling (non-random selection).
Probability sampling
random selection, everyone has an equal chance at being chosen
Non-probability sampling
non random selection
What are the four types of probability sampling?
Simple random, stratified random, cluster, and systematic sampling.
What is simple random sampling?
A method where each member of the population has an equal chance of being selected.
What is stratified random sampling?
A method that divides a population into smaller groups (strata) based on shared characteristics before random sampling.
What is cluster sampling?
A method where the population is divided into clusters, and entire clusters are randomly selected for inclusion in the sample.
What is systematic/strategic sampling?
A method that selects every nth participant from a list after a random starting point. (every third person on a list)
What is non-probability sampling?
A sampling method where not all individuals in a population have an equal chance of being selected.
What are common types of non-probability sampling?
Convenience, quota, purposive, snowball, and self-selection sampling.
what is convenience sampling
easy to reach, first come first serve
what is purpose sampling?
chosen based on certain characteristics
what is snowball sampling
using current participants to recruit other participants
what is quota sampling
choosing a certain person of a population
What is sampling error?
Random variation in a sample that occurs by chance.
What is sampling bias?
A systematic error that results in a non-representative sample, leading to skewed results.
What are common sampling errors?
Population specification error, non-response error, sample frame error, and selection error.
What is statistical power?
A measure of how likely a study is to detect a true difference between groups rather than one due to chance. Needs to be done at the star of the study
association vs causation
association indicates a relationship between variables, but it doesn’t confirm one causes the other.
cross sectional always uses what design
retrospective
