Wk 2- Biomed Skills: Scientific Method and Experimental Design Flashcards
Scientific method
The acceptable, logical path used to gain evidence that answers Qs about observable phenomena
Mnemonic for steps in scientific method
Only hippos eat dirty coke cans
Steps of scientific method
Observe
Hypothesis
Design and perform exp
Collect, analyse, interpret data
Conclusion
Communicate findings
Observe
Observe phenomena and research
ID gaps in knowledge
Hypothesis
What causes the phenomenon?
Aims: What Q do you want to answer?
Design and perform exp
Protocol and procedure
ID type of data
Collect, analyse, interpret data
What data
How is it collected
How is it analysed
Conclusion
Does the data support your hypothesis?
Yes- Repeat 3* for reliability
No- Repeat and make sure, adjust hypothesis if necessary
How are ideas generated
Reading literature and observing analogies b/n systems or patterns / breaks in patterns
New tech thus new Qs
Intuition, imagination, observation
Talking to clinicians and patients
How was Penicillin discovered?
1928
Alexander Fleming found that staph on culture plate were killed by Penicillium notatum
Grew culture and produced penicillium
What is outside scientific investigation?
Phenomena that cant be repeated or measured
Phenomena that cant be tested
Homeopathy
Alt medicine
Diliution of disease causing agent in the theory that water holds the ‘memory’ of it
Proven to be false
6757 papers on Pubmed as of March 2024
e.g. Paper showing activation of basophils using dilute IgE antiserum shown to be false
What should be considered when planning and designing an exp.?
Type of study
Control vars
Min bias
Analysis of results
Presentation of results
Logistics
Types of studies
Observational
- Descriptive (single case/group)
- Analytical (comparison b/n groups)
Experimental ( control vars)
Longitudinal (group over time)
Prospective (present/future)
Retrospective (past)
Cross-sectional (Single point in time)
Types of variables
Independent (X)
Dependent (Y)
Nuisance
Confounding
Relationship b/n X and Y
X causes Y
Nuisance vars
Change Y outside of X
Must be controlled
Examples of nuisance vars
Bio var
Envir.
Time
Sampling bias
Researcher bias
Participant bias
Social and cultural context
Confounding vars
Change at the same time as X
e.g.
X= Alzheimer’s
y= Memory loss
Confounding var= Age
How are confounding vars controlled?
Case control design- Study group compared to like controls
+ controls
Ensure result obtained is actually +
- controls
Ensure no false +
Example of a control
DNA of mice labelled w/ brdU
Mice performed dif tasks
Control group allowed stat analysis to be performed on exercise groups
Increased brdU indicated neurogenesis
How can vars be controlled?
Randomisation
Double-blind placebo-control trials (remove patient and clinician bias). Code unlocked at end
Controlled exp conditions recorded in lab journal
Sample size
Why is it important to record exp conditions?
So that it can be repeated by others to ensure validity
One exp only valid for one time, place, operator, subject, technique