1 Flashcards
what is postive control group should have
should show a particular change during the experiment
postive control
a group in an experiment that receives a treatment with a known result, there fore should show a particular change during the experiment
positive control group is used for
1used to asses test validity so to asses a new test’s ability to setect a disease = sensitivity
2Used to control for unknown variables during experiment
3to give the scientist something to compare with the test group
negative control are
groups where no phenomenon is expected, they ensure that there is no effect when there should be no effect
postive and negative control group are used for
Used to eliminate most potential confounding variables
Vehicle control
is a solution that contains everything but the test substance you are evaluating
characterise of vehicle control
Often times you may need to use solubilizing factor or reduced pH or ethanol to solubilize your drug
Sham or procedural control
a control that has the exact same procedure performed without the factor be evaluated
how to overcome Sham ethical issue
In many clinical trials you can run a comparative effectiveness study where you compare your new drug or procedure to the current standard of care.
Complete Randomized Design
All subjects are assigned randomly to a group or groups and then the IV is randomly assigned to each group.
Factorial Designs
One of the more common designs used in experiments. The simplest being a comparison between a treatment and a control. This would be a 1 x 2 factorial design
. Traditional dose-response curve is
a 1 x 6 factorial design
Factorial designs can also examine
the effects of more than one IV (called factors) simultaneously.
what is the thing that can be a factor?
Essentially anything that can be discretely identified can be a factor (dose, time, type of treatment
Repeated Measures Design (within subjects design)
The same test is performed multiple times on the same subject
Repeated Measures Design required
Special statistical analyses
the data is no longer completely random.
Randomized blocks designs
Similar to completely randomized designs except that blocks of subjects are generated or self-select, and then the blocks are assigned randomly across the factors
block
group of experimental units are similar ,
bock are important in
decrease the variability within each group , which allows us to detect differences caused by the treatments more clearly
types of experimental designs
completely randomized design, randomized block design , matched pairs design
a stratified, randomized block design
If there is an inherent level involved in the randomized block design
Matched Samples design.
A special case of repeated measures where two different groups are pre-matched to be as identical as possible.
This pre-matching introduces
a bias and is therefore treated statistically as a special case of repeated measures
advantage of Repeated measures
reduces within subjects variability
thereby increasing precision and power
leading to a reduction in the number of subjects.
Three major disadvantages of repeated measures
1- practice effects (only the first test is immune).
2- Differential carry over effects. Prior exposure affects subject in second exposure one way and a different way under another condition.
3- Statistical issues often assume total independence to meet the criteria of randomness (the basic probability o which al stats are based). Since they are repeated measures, randomness is impossible
practice effects overcome
This problem is overcome by counterbalancing. Practice effects affect all treatment conditions equally.
Repeated measures cross over design/two period two treatment design
Repeated measures designs are threatened by
carry-over effects, practice effects, and priming.
There is often a washout period (especially true with drug studies).
multiple blocks of experimental units and worried about sequence or carry over effects that could threaten the design
. You can randomize blocks,
or alternatively using a Latin Square.
