3.2 Experimentation Flashcards
Validity
variables controlled so that any measured effect is likely to be due to the independent variable
variables controlled so that any measured effect is likely to be due to the independent variable
Validity
Reliability
consistent data obtained through repeats and replicates carried out independently
consistent data obtained through repeats and replicates carried out independently
Reliability
Accuracy
accurate data that are close to the true values for the measurements made
accurate data that are close to the true values for the measurements made
Accuracy
Precision
precise data values in replicated measurements that lie close to each other
precise data values in replicated measurements that lie close to each other
Precision
Multifactorial studies
have more than one independent variable
have more than one independent variable
Multifactorial studies
Analysing multifactorial studies
More difficult to analyse and confounding variables will be present
What studies difficult to analyse and confounding variables will be present
Multifactorial studies
Observational studies
Where is there more than one independent variable
Where is there more than one independent variable
Observational studies
Analysing multifactorial studies
More difficult as confounding variables are present
What is a pilot study used for
Help plan procedures
Assess validity
Check techniques
What is used to..
Help plan procedures
Assess validity
Check techniques
A pilot study
What is the result of a pilot study
Allows for evaluation and modification of experimental design
What allows for evaluation and modification of experimental design
Pilot study
What can a pilot study ensure
An appropriate range of values for the independent variable
What ensures you have an appropriate range of values for the independent variable
Pilot study
What does a pilot study give an idea of to do with averaging results
The number of repeat measurements required to give a representative value for each independent data point
Independent variable
The variable that is changed in a scientific experiment
The variable that is changed in a scientific experiment
Independent variable
Dependant variable
The variable being measured in a scientific experiment
The variable being measured in a scientific experiment
Dependant variable
Independent and dependent variables can be
_____________
continuous or discrete
Experiments involve doing what to the independent variable
The investigator must manipulate the independent variable
What do you compare the experimental treatment group to
A control group
Simple experiment
One independent variable
One independent variable
Simple experiment
Multi factorial experiment
More than one independent variable
Multi factorial experiment
More than one independent variable
More than one independent variable
Multi factorial experiment
Experiments that don’t have a truly independent variable
When investigators use groups that already exist
Result of investigators using groups that already exist
There is no truly independent variable
Observational studies advantage and disadvantage
They are good at detecting correlation
Since they do not directly test a hypothesis, they are less useful for determining causation
They are good at detecting correlation
Since they do not directly test a hypothesis, they are less useful for determining causation
Observational studies
Independent variable in observational studies
It is not directly controlled by the investigator, for ethical or logistical reasons
In observational studies, what is not directly controlled by the investigator, for ethical or logistical reasons
Independent variable
what means that other variables besides the independent variable may affect the dependant variable
The complexities of biological systems
What must be done to confounding variables
Must be held constant if possible, or at least monitored so that their effect on the results can be accounted for in the analysis
what must be held constant if possible, or at least monitored so that their effect on the results can be accounted for in the analysis
Confounding variables
What to do if confounding variables cannot be easily controlled
A randomised block design could be used
When might you use a randomised block design
When confounding variables cannot be easily controlled
How do randomised blocks of treatment and control groups work
They can be distributed in such a way that the influence of any confounding variable is likely to be the same across the treatment and control groups