Statistics: CH 1

Comprehensive vocabulary flashcards covering the basic definitions, sampling techniques, bias types, study designs, and data classifications from Chapter 1 of Statistics.

Statistics

The science of collecting, analyzing, and drawing conclusions from data.

Inferential Statistics

The process of taking samples to draw a conclusion on the population.

Population

The entire collection of objects about which information is desired.

Sample

A subset of the population selected to study.

Parameter

Any number derived from a population; usually denoted with Greek letters such as $\mu$ (Pop Mean), $\sigma$ (Pop Standard Deviation), $\sigma^2$ (Pop Variance), or $p$ (Pop proportion).

Statistic

Any number derived from a sample; usually denoted with English letters.

Law of Large Numbers

As sample increase in size, the resulting statistics better reflect the value of the population parameter.

Variable

A characteristic of interest that will vary with each member of a population or sample.

Data

Observations or values of a variable.

Simple Random Sample

A sample in which every object has the same chance of being selected.

With replacement (WR)

Once selected, the object from the population has the potential for future selection.

Without replacement (WOR)

Once selected, the object from the population is excluded from future selection.

Systematic Sampling

A sample in which we pre-assign numbers to the objects in the population and then select every kth observation.

Stratified Sampling

A sample in which the population is partitioned into groups, and then a simple random sample (SRS) is taken from each group; described as taking SOME from ALL.

Cluster Sampling

A sample in which the population is partitioned into groups, a simple random sample of groups (clusters) is taken, and then ALL members of the selected group(s) are taken; described as taking ALL from SOME.

Convenience Sampling

A bad sampling technique that uses a readily available or convenient group to form a sample, which results in bias.

Bias

Occurs when a study or experiment fails to represent the population value (can be intentional or unintentional).

Unbiased

Occurs when a study or experiment tends to have the correct estimate of a population value, on average.

Sampling Bias

Occurs when some members of the population are more likely to be asked than others.

Voluntary Response Bias

Occurs when you leave it up to the people in your sample to respond.

Self-Interest Bias

Occurs when claims are made but no data is supplied and it is evident that the researcher has an interest in the outcome of the experiment.

Social Acceptance Bias

Occurs when people are hesitant to report the truth because it may reflect negatively on them.

Leading Question Bias

Occurs if a question in a survey leads people to a response rather than letting them answer objectively.

Nonresponsive Bias

Occurs when part of your sample fails to respond, causing results to be inaccurate.

Observational Study

A study where the investigator observes specific characteristics of the sample without changing or controlling anything.

Randomized Experiment

A study where an investigator observes how some specific characteristics change when they manipulate one or more factors.

Double-Blind

An attempt to prevent bias where neither the subjects nor the medical professionals know what is being administered.

Placebo

A treatment given to a control group, such as a 'sugar' pill, to provide a baseline for comparison.

Confounding Variable

Occurs when an effect is noticed, but it is unclear which factor was involved.

Categorical Data

Qualitative observations that describe qualities, opinions, or characteristics rather than numbers.

Ordinal Data

Qualitative data where order or rank is deemed meaningful, such as letter grades.

Nominal Data

Qualitative data where order or rank is NOT deemed meaningful, such as favorite color.

Numerical Data

Quantitative observations consisting of quantities or numbers.

Discrete Data

Quantitative values that can only be represented as isolated parts on a number line (counted), such as the number of pets.

Continuous Data

Quantitative values that can take on any value within an interval on a number line (measured), such as height, weight, or time.

Frequency Distribution

A table that summarizes the distribution of data values for a variable.

Relative Frequency (RF)

The proportion of observations in a class of a distribution, calculated using the formula $RF = \frac{x_i}{n}$.

Upper Class Limit

The largest data value occurring in a class or category.

Lower Class Limit

The smallest data value occurring in a class or category.

Class Width

The difference between consecutive lower-class limits.