Stats terminology

Statistics

Statistics

methods used to measure certain aspects of nature and quantify their uncertainty and test hypotheses about them

Population

The group of interest

Sample

Subset/section of the population

Sample Unit

subjects/replicates; individuals or groups that make up the population

Parameter

a quantity associated with a population

Estimation

inferring an unknown quantity of a population from a sample

Sampling error

the difference between the estimated value and the true parameter value

Precision

Having a low sampling error

Bias

a systematic difference between the estimates and the true parameter value

Accuracy

Having low bias

Random sample

• has an equal chance of being sampled

• independent

• reduce bias

Sample of Convenience

a sample of individuals that are easily accessible to the researcher

• biased

Volunteer Bias

systemic difference between the pool of volunteers and the population to which they belong

Variable

a characteristic that varies among individuals

Data

Raw measurements of values of one or more variables from the individual in a sample

Categorical Variables

Define membership in a group

• male/female

• size class

• Nominal or ordinal

Nominal

no order to the categories

Ordinal

order but no magnitude

• egg, larvae, juvenile, subadult, adult

Numerical Variables

quantitative measurements of magnitude

• Continuous or Discrete

Continuous

can take any real number value

• temperature

Discrete

only take integer values

• number of offspring

Hypothesis testing

determining how well a hypothesis about the population parameters fits the data from a sample

Null Hypothesis (Ho)

hypothesis stating no change

Experimental Studies

researcher assigns a random treatment to each individual

Observational Studies

researcher has no control over which units fall into which groups

• can only determine associations

Explanatory

a change that predicts or affects the other variable

Response variable

affected by the explanatory variable

Probability distribution

distribution of a variable in the whole population

Normal Distribution

theoretical probability distribution

• symmetrical bell

Skew

measures whether data are symmetrical or spread out to one side

• Right/Left skew → side with longer tail