Quantitative Research Evidence Flashcards
What is the difference between quantitative and qualitative?
Quantitative
- Generalisation
- Concise and narrow focus
- Theory driven
- Begins with a hypothesis (hypothesis testing/confirmatory)
- Requires greater number participants
Qualitative
- Contextualisation
- Complex and broad focus
- Data driven
- Hypothesis generating (exploratory)
- Requires fewer participants
If researchers assign participants to receive different treatments (exposures) is it experimental or observational?
Experimental
What is a cohort study?
- Looks forward in time
- Investigator identifies groups that differ according to the exposure of interest e.g. group with healthy BMI vs increased BMI. Groups are then followed forward in time to determine if they differ in relation to outcome of interest e.g. heart disease.
What is a case control study?
Looks back in time to investigate exposures that caused outcome. e.g. people with heart disease and without HD (control) and see if they differ in terms of exposures
What is a cross-sectional study?
- Snapshot at one timepoint
- Looks at association between factors
What is sampling?
The way in which a group of people are taken from a larger population
What are inferential statistics?
Allows us to make inferences about the population from the results of a smaller sample
What is a simple random sample?
Every individual in the populations has an even chance of being selected e.g. tossing a coin
What is a systematic sample?
Performed by assembling the population into a numbered list, randomly selecting the starting point and then selecting individuals using a fixed interval
What is a stratified sample?
Splitting population into smaller groups that have similar characteristics (strata). Then we randomly sample a proportional amount of individuals from each group. e.g. splitting by gender then selecting a proportion from each group
What is a cluster sample?
When you have naturally occurring groups from which you wish to sample e.g. patients belonging to different GP surgeries
What are the 4 types of non-probability sampling?
Convenience
Voluntary response
Purposive sample
Snowball sample
What is a convenience sample?
Selecting individuals that are easy to reach e.g. survey to nursing students
What is a voluntary response sample?
Researcher using adverts, and volunteers contact researcher. This bias’ the study
What is a purposive sample?
Researchers specifically select participants based on characteristics they require in the study population
What is a snowball sample?
Participants are asked to refer researcher to people just like them. Useful when studying hard to reach individuals e.g. drug users
What is generalisability/external validity?
The extent to which the findings of a study can be applied to the population from which the study sampled. A sample should be representative of the population to ensure generalisation
What is an exposure?
Any factor that may be associated with an outcome of interest
e.g. smoking -ve exposure; exercise +ve exposure
What is an outcome measure used for?
Used to assess the effect, both positive and negative of an intervention. Quantitative outcome measures can be both objective and subjective.
- Objective - blood pressure, white cell count (not open to outside influence)
- Subjective - visual analogue scale for rating pain (patient reported outcome)
What is data collection?
The process of conducting the outcome measures and recording results.
What is a randomised controlled trial?
- A study in which a number of similar people are randomly assigned to 2 or more groups to test a specific drug, treatment or other intervention.
- One group has drug, other has placebo/control/other intervention
- Groups followed up to see how effective the experimental intervention was
- Outcomes measured at specific times
- This method reduces bias
How does randomisation work?
- Individuals are assigned to different groups in a random manner e.g. random numbers table or computer generated sequence
- Each person has the same chance of receiving each intervention.
- Attempts to create groups that are similar at baseline
- Differences in outcome measures at the end of the study can be attributed to the intervention
Why is stratified randomisation used?
Used to create treatment groups with similar characteristics
Particularly useful when certain characteristics are though to affect response to the intervention
What is allocation concealment
Concealment of which group each participant will be assigned to
Prevents selection bias
Neither investigators nor participants know groups
Single vs double blinding
Single - Patients prevented from discovering which treatment they are receiving
Double - Patients and investigators prevented from discovering which treatment they are receiving
When would you conduct a non-randomised control trial?
- Random allocation inappropriate or not possible
- Failure to ensure true randomisation
What are the advantages and disadvantages of non-randomised controlled trials?
- Advantages - allowing for participant preference can improve adherence and reduce drop out rates
- Disadvantages - Cannot control for confounding factors; increased risk of selection bias (If groups are so different at start of trial, results of study cant be reliably attributed to the intervention being investigated
What is an observational study?
When the exposure is not assigned, it is observed
