Sampling Data& Descriptive Stats

Probability

making inferences from a ( the) population to an observation sample

Statistics

numerical characteristic of the sample ( what the direct data is telling us)

Population

the larger collection of objects being studied

parameter

the numerical characteristic that describes the whole population

sample

a smaller group that accurately represents a population

inferential statistics

observation sample

Representative Sample

a small, accurately selected subset of a larger population that mirrors the key characteristics of the whole group

Variables

is a characteristic or measurement that can be determined for each member of a population.

Categorical (qualitative)

responses in catagories and are not numerical

Numerical ( quantitative)

Responses that has values with equal units (numbers)

Simple Random Sample

a subset of individuals chosen entirely by chance from a larger population. In this method, every single member of the population has an exactly equal and independent probability of being selected, ensuring the sample represents the group without bias

Stratify

A sampling method where you divide the population into preexisting groups ( age,gender, race) = Stratra and proportionally sample each group

Cluster

A sampling method where you divide a population into groups and randomly select a specific cluster population to participate

Systematic

A sampling method where at a given interval you will interview ( every nth) after randomly choosing a starting point

Connivence

A sampling method where you do whatever is easiest typically confined location and sample size ( THE WORST) sampling method

Bias

a systematic error that causes a statistical model or estimation to inaccurately reflect the true reality of a population

Sampling error

A discrepency that occurs when the sample is not representative of the population

Sampling bias

When there is a systemic issue that prevents ALL or ANY member to be included within a sample

non-sampling error

When discrepancies occur from true population value that is caused by human mistakes, faulty design, or data collection flaws

Quantitative Discrete

Typically when counting items ( things that remain whole)

Quantitative Continuous

Any value within the number line and typically measurements ( has units!)

Nominal Data

Data that comes in the forms of names or other nouns = Pie Chart

Ordinal Data

Data that is in the form of ranking= bar chart

• Averaging is NOT good messes it up

Interval Data

Data measured on a scale with equal intervals between values, but with no true zero point.

Key traits

• Equal spacing between values

• Zero has a meaning, not the absence of the variable

• Can calculate mean, median, and standard deviation

Ratio Data

Data that is equal, with meaningful intervals between values, but also has a true zero point, meaning zero represents the complete absence of the measured quantity.

• Typically provides the most info

Frequency

The appearance of a specific value within a data set

Relative Frequency

shows how often a specific event or data point occurs relative to the total number of outcomes.

• typically a fraction, decimal, or percent

Cumulative Relative Frequency ( CRF)

A measure that tells you the proportion or percentage of a dataset that falls at or below a specific value

Lurking variables

Additional variables that are not researched/ analyzed but have an impact on the results of the data

random assignment

When subjects are assigned treatments randomly

• the best way to avoid lurking variables

Explanatory Variable

The variable that is maniupated or changed

Response Variable

The variable that changes as a result of the manipulated variable

Single- blind

when just the participant is unaware of which is a control and which is the treatment

Double-Blind

When both the participant and the administer are unaware of which treatment is which

Placebo

A fake treatment that will have no effect eliminating the power of suggestion

Bar Graphs

Used to visually represent frequency

• mostly nominal data!

histogram

a graphical chart that displays the distribution and frequency of continuous numerical data

Boxplot

ay of displaying the distribution of data

5 number summary

• min

• max

• Q1

• Q3

• median

Interquartile Range

measuring dispersion that describes the spread of the middle 50% of a dataset.

IQR=Q3-Q1

Outlier

a data point that is outside of 50% of the data set

• Q1-IQR or Q3+IQR