Statistics
Statistics - a branch of mathematics
concern with the collection, summary of , description of and inference
of data or observations or results of an experiment.
also statistics - a number representing a numerical property
of a sample of a population
Descriptive statistics -
that area of statistics that involves collecting, summarizing and describing
data or observations or results of an experiment.
Inferential statistics -
that area of statistics concern with making conclusions or judgment (inference)
based on analysis of data.
Parameter - a number representing a numerical
property of a population, e.g. Mean, median, variance; also called a statistics
when
the numerial property is made from a sample of the population.
Sampling
Population - the collection of all members
or elements of a group of interest.
Sample - a collection of a portion of the
population, a sample involving all members of a population is called a Census.
Random sample - a sample obtained
from the population in which every member of the population is equally
likely of being selected.
Stratified sample - is sampling
taken from subgroups (strata) of a population, each subgroup is called
a stratum.
Cluster sample - sampled obtained
from the grouping of many strata (subgroups).
Systematic sample - samples taken
at prescribed intervals usually start at some random point along an ordered
list.
Sample of convenience - a
sample taken because of its easy availability (convenience) often not taken
at random.
Bias - an inclination or preference, especially
one that interferes with impartial judgment: prejudice. Webster's
II
Response bias - occurs when results
are obtained falsely either by the respondents (sample) or the process
or methods used to obtain the results
Nonresponse bias - occurs when
segment or portions of the population are inadequately represented.
Experiments
Treatment - a property or characteristic
being studied
Treatment group - the group being
studied having the property being studied
Control group - the group being studied
not having the property being studied
Experiment - any activity design to study outcomes , results,
possibility of chance.
Controlled experiment - an
experiment in which subjects given treatment is controlled.
Observational study - an experiment
where administration of treatment to subject is not controlled.
Randomly controlled
experiment - an experiment where subject assigned treatment randomly.
Confounding factor - is a property
other than the treatment that can influence the results of an experiment
or study.
A double-blind study -
an experiment in which neither the subjects nor those conducting the experiment
know which subjects are in the control group or treatment group.
A cross-sectional study
- different subjects are compared at he same time
A longitudinal study -
the same subjects are compared with themselves over different periods of
time.
|