Chapter 1 Flashcards
Statistics
The study of methods to describe and measure aspects of nature from samples.
Statistical Hypothesis Testing
Specific claim about a population parameter
Population
All of the individuals in the world
Samples
A subset of the population of interest
- Can be individual or group of individuals
Parameter
A quantity describing a population (measurement or observation)
Estimate
A quantity calculated from a sample
Sampling Bias
A systematic difference between parameter and its estimate
- Occurs when the samples aren’t representative of the population
Sampling Error
Undirected deviation of estimates away from parameters
Volunteer Bias
Volunteers are likely to be different on average from the population
Properties of Good Sample
Random and Independent (Equal chance of selection) selection
Variable
Any characteristics or measurements that differs from individual to individual
Data
Measurement of the variables
What ate the two types of variable?
Numeric (Quantitative) and categorical (Qualitative)
Numeric Variable
Magnitude on a numeric scale
Categorical Variable
Describe members in a category or groups
1. Ordinal: can be ranked (sizes)
2. Nominal: cannot be ranked (sex chromosomes)
Numerical Data
Continuous - continued with decimals or fractions
Discrete - whole number, can be counted
Explanatory Variable
Manipulated by the researcher: “Independent”
Response Variable
Measured effect: “Dependent”
Frequency Distribution
Shows how often each value of the variable occurs in the sample: common values, rare values, average, and variation.
What are the two types of Probability distribution?
Actual and Theoretical (normal distribution)
Actual/Real distribution is almost never known…Why not?
For actual, needs to measure every member of the population!
Experimental Studies
Researchers assign treatments to individuals
1. Controlling a variable (placebo/non placebo)
2. Cause and effect
Observation Studies
Researches do not assign treatments
1. Observe association
Observation studies CANNOT
Prove causation or disentangle cause and effect
Why bother with observation studies?
Gather information to plan an experiments and ethics cannot do experiments such as on pregnant women and smoking
Confounding Variables
Variable that are not considered in an experimental study but they may affect the response variable
How could experimental studies remove confounding?
Through random assignment of treatment
Examples: “The number of violent crimes tend to increase when ice-cream sales increase”
What is the confounding variable?
Heat
Well designed experiment can reveal causation. What could be considered as a well designed experiment?
Random assignments and artifacts (placebo effects)
