Exam 3 Lecture 5 Flashcards
Left-Skewed distributions are _______ skewed, and Right-Skewed distributions are _________ skewed.
Left-Skewed distributions are negatively skewed, and right-skewed distributions are positively skewed.
What defines a skew?
The direction of the long tail defines the skew because it indicates where you’ll find the majority of exceptional values.
Left-skewed distributions
Left skewed distributions occur when the long tail is on the left side of the distribution. Statisticians also refer to them as negatively skewed. This condition occurs because probabilities taper off more slowly for lower values. Therefore, you’ll find extreme values far from the peak on the low side more frequently than the high side. The crucial point to keep in mind is that the direction of the long tail defines the skew because it indicates where you’ll find the majority of exceptional values.
Data MUST be quality checked before any type of ________ can be made.
Inference
What is an inference in statistics?
Using information from a sample to infer properties about a population. This is done using statistical tests.
Protocol vs. Measures vs. Participants
The Protocol
- 1 mile run
- 100 pullups
- 200 pushups
- 300 squats
- 1 mile run
The Measures
- Body composition
- Upper & lower body muscular strength
- Muscular endurance
- Anaerobic
- Aerobic power
Poor ______ _________ is the #1 way that people lie with or botch statistics. You should not believe every scientific result you hear or read about. There are a lot of bad study designs, bad samples, bad data & bad graphs. To be able to detect the junk, you must understand the basic principles of statistics!
DATA QUALITY
Why is it important to differentiate the good from the bad in statistics and graphs?
Statistics and graphs can be used incorrectly (by accident and on purpose). In our fact-oriented world, this has power.
- Misleading people with partial information
- Misleading people with manipulated graphs
Even simple “fact checking” doesn’t work.
- There are too many facts/ too many things passed off as facts
- Every piece of evidence must be vetted
- We all need to be 1- part scientist
You must be able to read things, think them through, and make an informed decision about the accuracy, quality & value.
Beware of deception
Graphs can be manipulated
- Look at x-axis LABELS and VALUES
- Look at y-axis LABELS and VALUES
Don’t forget to ask yourself whether the relationship or comparison makes sense!
Is it just the central tendency that is shown, or is variability also shown?
Is a mean used when the data are non-normal?
A bad graph can be dangerous because
It preys on people’s biases, their lack of attention to detail/education/interest.
It feeds them “facts” (data points) to bolster their preconceived notions.
A good graph…
- Uses a reasonable range
- Uses a logical range
- Is logical and easy to follow
- Is simple
