Hypothesis
Represents the proposed answer to the experimental question
Independent Variable
The variable ring manipulated to test the hypothesis
Dependent Variable
the variable being affected and changed
Control Group
The group that the experimental group is being compared to and is not exposed to the independent variable
Positive Control
A control that tests that the experiment is working and has a known result
Negative Control
A control that should not change and provides the baseline
Controlled Variables
Variables that ensure that any observed effects are due to the independent variable and not other variables
Reproducibility and Repeatability
More replicates, large sample size, confirmation of results via other researchers
Tips about writing an experiment
Be brief and only state the main parts
Graphing
Bar Graph
Shows visual summary of how a data set is spread across various categories
Not continuous numerical data
Scatter Plot
Allows you to establish relationships between two variables
Can draw trend line (line of best fit) to form a line graph
Line Graph
Used to represent changes over time
Want to see changes in slope
No line of best fit
Normal Curve
Describes most natural phenomena
Does science usually work with populations or samples?
Samples
Precision
Degree of closeness of the measurements with each other
AKA repeatability and reproducibility error
Accuracy
Degree of closeness of the measurements to the target value
AKA bias error
p=
probability value
Likelihood something occurs by chance
Variability
More consistent data, the more confidence we have in it and a tool for predictions
Standard Deviation measures…
Precision
Standard Deviation
Absolute measure of dispersion of a series
Higher standard deviation = less confidence
square root of the sum of the squared values of the difference between the value and the mean divided by degrees of freedom
Standard Eror
Tests reliability of data
Standard deviation in repeated samples from a population bc standard deviation only measures one dataset
Standard deviation divided by sqaure root of sample size
Small number = closer to true mean (more values)
2SEM
Gives us 95% confidence
If error bars overlap then…
There is not a statistically significant difference between the two
SD differs between each bar