### Types of Variables
How does each of these factors in experimental design help allow them to indicate cause and effect?
1. Randomize: ==this ensures there is no systematic difference between groups.==
2. Hold Constant: ==every other variable should be controlled for in your experimental design so you know the independent variable was the cause of the depend variables variation.==
3. Matched groups/Counterbalance(only used in some experimental design): ==if you have one variable that you are particularly worried will dominate the outcome matching/counterbalancing helps, you can match/counterbalance the groups by splitting groups in series. (For example if you are measuring bench press max based on creatine use and your sample is very small with a massive amount of variation in normal bench press maxing. You would split groups by putting the lowest max person in group 1 then second lowest in group 2 then third lowest in group 1 and so on).==
What are the four scales of measurement?
1. ==Nominal scale:== ==an unordered set of values identified only by name. ![[Pasted image 20220829151535.png]]==
1. ==Ordinal scale==: ==an ordered set of categories. A score can be identified as 'larger' than another, but not by how much. (for example rock climbing difficulty or class rankings, freshman, sophomore etc. ![[Pasted image 20220829151633.png]]==
3. ==Ratio scale:== ==a numerical scale where the value zero indicates "non of" the variable. This means measurements compared by ratios are meaningful. ![[Pasted image 20220829151946.png]]==
4. ==Interval scale:== ==another numerical scale but the zero point is located arbitrarily.![[Pasted image 20220829152057.png]]== ^a50fc7
What are the two types of variables?
1. ==Discrete variables:== ==like integers, or nonnumeric categories (such as ranks like AP test scores, class sizes, and military rank). Separate indivisible categories. Discontinuous.==
2. ==Continuous variables:== ==like real numbers. Sample values are not limited to a fixed set of categories but can vary continuously.== ^fc2e25
# Summarizing Data Sets
## Frequency Tables, Histograms and Bar Graphs
==Binning data== is common when creating Histograms so that we don't get a histogram with thousands of variables. Importantly, bin boundaries constitute the ==real limits== of that continuous variable. Example of Binned data: ![[Pasted image 20220829153442.png]]
Real limits do not apply to ==discrete== variables.
==Real limits== are boundaries between adjacent score bins (categories), and govern the grouping of scores for binning purposes.
What are the real limits of the Bin called 3? ![[Pasted image 20220829153704.png]] :: ![[Pasted image 20220831150000.png]]
==Nominal or or ordinal== independent variables are displayed on ==bar graphs== not histograms.
In this binned frequency distribution what is the maximum value of a number in the bin 34-35? ![[Pasted image 20220829154452.png]] :: 35.5
A bracket in a range means that end of the range is ==inclusive -- it includes the element listed==. A parenthesis means that end is ==exclusive and doesn't contain the listed element.==
## Percentile Ranks
^13693d
Percentiles and percentile ranks are used to describe the position of individual scores within a distribution. Percentile rank gives the cumulative percentage associated with a particular score. A score that is identified by its rank is called a percentile.
Percentiles work off the same idea as quartiles. Quartiles divide the dataset into four where as percentiles divide them into 100.
Assuming continuous data what is the 83rd percentile? ![[Pasted image 20220831145556.png]] :: ![[Pasted image 20220831145627.png]]