L2: Study design Flashcards
things to think about when designing a research study?
Research question: What is the research question?
Study design: How will you undertake the study?
Subjects : Who are the subjects? How will they be selected?
Data: What data needs to be collected? What kind of measures?
Analysis: What analyses do you need to undertake?
Interpretation: What do the results mean? How valid are they?
types of study design?
your research question determines what type of study you do. can be quantitative which branches into experimental (controlled trial, quasi-experimental) or observational–> ecological and observational branches into: cross-sectional, cohort or case-control. or research can be qualitative.
quantiative vs qualitative?
Quantitative: Uses numbers; exposures and outcomes are measurable
How many?
Who is at risk?
What causes this disease?
Is there an improvement?
Qualitative: Uses words; stories, experiences, observations
Why do people do ……?
How do they feel about…?
What is their experience of…?
observational vs experimental? how do you tell which is which?
Does the researcher control the exposure?
Yes: experimental
No: observational
Do you want to determine if something is “causal”?
Yes: experimental
No: observational
Difficult to establish causation in observational studies (because of bias, confounding, temporality?)
types of observational studies?
Descriptive: used to formulate a certain hypothesis
Examples: case studies, cross-sectional studies, ecological studies
A descriptive observational study aims to observe and describe characteristics of a population or phenomenon without manipulating any variables. In this case, the study focuses on “the prevalence and trends in obesity.”
Descriptive: The study is trying to describe the extent (prevalence) and patterns (trends) of obesity within a population. It’s not attempting to find cause-and-effect relationships or influence any factors, just documenting and summarizing data.
Observational: The study observes and gathers data on obesity rates over time (trends) and across different groups (prevalence) without intervening or assigning any treatments or conditions to the participants.
This makes it descriptive because it’s not testing hypotheses but rather collecting and presenting data, and it’s observational because the researchers are simply watching and recording the data as it naturally occurs.
Analytical: used to test hypothesis
Examples: case-control, cohort
An analytical study is designed to explore relationships between variables and test hypotheses. In this case, the study question, “How much exercise is necessary to reduce the risk of specific diseases?” is considered analytical because:
Analytical: The study is not just observing the amount of exercise and disease risk but is trying to analyze and quantify how much exercise directly impacts the risk of specific diseases. It involves examining the relationship between two variables: exercise and disease risk. The goal is to determine whether a certain amount of exercise reduces risk, which involves testing hypotheses, measuring associations, and possibly drawing conclusions about causality.
Hypothesis-driven: The study may involve comparing groups of people with different levels of exercise and examining how these levels affect disease risk, testing whether more or less exercise leads to a measurable change in health outcomes.
So, it’s analytical because it goes beyond simple description to explore the impact of exercise on disease risk.
target, source and study population?
Target population: group of individuals about whom you want to make inferences
Source population: group of individuals from whom the study population is drawn
Study population: group of individuals that serve as study participants
Creating a valid link between the target population and the source population refers to ensuring that the sample or group you’re studying (source population) is representative of the group you ultimately want to draw conclusions about (target population). This is crucial for the generalizability or external validity of your study findings. you then have to choose a study population who represents the source population.
dependent and independent variables?
Independent variables:
Risk factor
Exposure
Intervention
Dependent variable
Disease
Outcome
cause—-> effect
confounding factors?
Confounding factors:
A distortion of the measure of association between the exposure and the
outcome due to the mixing of the effect of the exposure with another risk factor.
E.g: when seeing if physical activity leads to increased risk of getting heart disease: smoking is a confounder.
Need to collect measures of confounding factors aswell as exposure and outcome.
reliability, validity and precise measures?
Systematic Error (Bias)
Leads to consistently wrong results in the same direction.
Example: A scale that always adds 2 kg to your weight.
It is precise (consistent) but not valid (incorrect). *chatgpt
More systematic error → More chance of a wrong result.
Random Error (Variability)
Due to chance and unpredictable fluctuations.
Example: A scale that gives slightly different readings each time you step on it.
It reduces precision (repeatability).
Less random error → More precise results.
Reliability vs. Validity
Reliable = Precise (low random error, results are consistent).
Valid = Accurate (low systematic error, results are correct).
A study should be both reliable and valid for the best chance of getting the true result.
Systematic error- more chance of finding wrong resukt.
Reliable, precise, lack of random error
Reliable and valid- best chance of finding true result
Random error- less chance of finding the true result
Valid, lack of systematic error
Check what this means no clue what this is saying
choosing between different observational methods?
Choosing between different observational methods
Depends on:
how rare the outcome is
what data exists for the population of interest
whether the temporal relationship is important
i.e. Exposure → Outcome (cause → effect)
how quickly you want the answer
money / resources
