Experimental Design (Labs 7 + 8) Flashcards
What are the components of good experimental design? (6)
1) Refutable Hypothesis (testable)
2) Well Controlled
3) Reproducible
4) Minimization of Bias
5) Representative
6) Applicable Statistics
How many groups (at minimum) must an experiment have (in the context of this course)?
TWO
1) Experimental Group
2) Control Group
Control vs Experimental Groups
Control group receives exactly the same treatment as the experimental group, EXCEPT for the variable that is being tested!
Control = NO treatment
Experimental = Treatment
What is a control experiment?
An experiment that has a “known” outcome
Positive vs Negative Control
(+ what are the purposes of each)
Positive control = Experiment that is expected to produce a KNOWN EFFECT
–> Used to make sure that your procedure is working properly
Negative Control = Experiment that is expected to produce NO EFFECT
–> Used to identify confounding variables
Null Hypothesis
The hypothesis in which NO relationship exists between two sets of data or variables being analyzed
–> Most statistical tests assume the null hypothesis to be true: operates under the assumption that there is no significant difference or association between groups unless the data strongly suggests otherwise
In what type of experiment are null hypotheses used?
Any experiment that is statistically testing for a significant difference or relationship between variables
Alternate Hypothesis
Implies that there IS a relationship between two sets of data/variables being analyzed
What is a p-value?
The probability (likelihood) of a result occurring by chance
The lower the p-value, the less likely the result occurred due to random chance (and was likely due to the treatment)
At What P Value Would a Scientist Reject the Null Hypothesis?
P < 0.05 (Probability of effect occurring due to chance < 5%)
Qualitative Data
Descriptive information that cannot be quantified by numbers; data sorted by category, not by number
–> Can consist of audio, images, symbols, or text
Quantitative Data
A type of data that can be expressed in numerical values
–> Countable and can therefore undergo statistical data analysis and can be plotted
What are the two types of qualitative data?
1) Nominal
2) Ordinal
What are the two types of quantitative data?
1) Discrete
2) Continuous
Nominal Data
Qualitative data that categorizes variables into distinct groups with no inherent order or ranking (essentially acting as labels or names without any numerical value)
Ex: Gender
