Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
46 Cards in this Set
- Front
- Back
Statistics
|
A quantitative process of gathering and analyzing data to make inferences. Form the Italian word Statista meaning of the state.
|
|
Data (plural)
|
Pieces of information
|
|
Datum (singular)
|
a single piece of information
|
|
Quantitative
|
inherently numeric
|
|
Qualitative
|
non-numeric
|
|
Variable
|
name of the data
|
|
discrete
|
counting; expressed only as whole numbers
|
|
Continuous
|
measuring; able to be expressed as any real number(s)
|
|
Experimental Research
|
requires an intrusive "treatment" and a control group
|
|
Observational Research
|
non-experimental: not intrusive: passive
|
|
Population
|
all possible data elements under consideration, as defined.
N= population size= number of elements in population. |
|
Sample
|
any subset of the population.
n= sample size= number of elements in sample. |
|
Census
|
Population = census
|
|
What is the criteria for a statistically valid random sample?
|
1. every element of the sample is also an element of population.
2. Every element of the population has an equal chance of being selected. |
|
Simple Random Sample
|
best for small populations and samples
randInt>(1,N,n) |
|
Systematic Random Sample
|
variant of sample method; best for large populations
use sample interval i= N/n (round up) Select random start (btw 1-i, or 1-N) beginning with random start, select every i-th element. |
|
Stratified Random Sample
|
used if representation of subpopulations is desired or required
|
|
Stratum
|
homogeneous subpopulation
Population N= N1+N2+.... |
|
Cluster Random Sample
|
best for very large population with many strata
a. take a random sample of all strata= cluster. b. take random subsample from each member of the cluster only. |
|
Array
|
a preliminary informal listing of data collected: unordered or ordered.
|
|
Table
|
what, where, when, how many.
|
|
Components of every table
|
Title: what, where, when, how many.
2+ columns, with column headings 2-10 lines no more then 20 |
|
Grouped data or frequency table
|
1st column: numeric data, usually in groups/ classes
2nd column: frequency: number of occurrences per class |
|
Rules for constructing a grouped data/ frequency table:
|
1. the first datum is in the first class; last datum is in the last class.
2. No data value is in more than one class; (no overlapping) 3. uniform class widths for all classes is recommended (not required) 4. closed interval format if data is discrete, open interval format if data is continuous. |
|
W= class width
|
(max-min)/k (round up!)
k= # of recommended classes |
|
lower class limits (l.c.l.)
|
1st value in each class
|
|
upper class limits (u.c.l.)
|
last value in each class
|
|
lower class boundaries (u.c.b.)
|
lowest possible value in each class
|
|
upper class boundaries (u.c.b.)
|
highest possible value in each class
|
|
midpoints
|
(l.c.l. + u.c.l.)/2
|
|
Bar graph
|
vertical or horizontal separated bars, categorical axis
|
|
Pareto Chart
|
vertical bar graph with descending separated or joined bars.
|
|
Circle graph
|
proportional wedge segments for each datum
|
|
line graph
|
chronological, points connected by line segments.
|
|
histogram
|
contiguous vertical bars, numeric axes
|
|
Polygon
|
same axes as histogram, consecutive, connected straight line segments.
|
|
Central Measures
|
averages, measure the typical or central value of datum
|
|
Mode
|
the observed average, the value that occurs most often
|
|
Median
|
the position average, requires data to be in order.
|
|
Mean
|
the calculated average
|
|
Measure of dispersion/ variation
|
measure the amount of variation among the data.
|
|
Range
|
R= the quick and dirty" measure of variation: R= max-min or R=H-?
|
|
Standard Deviation
|
Svx the required standard dispersion measure; based on mean.
|
|
Standard deviation
|
Sx= the required standard dispersion measure, based on mean.
|
|
Variance:
|
the estimated accumulated amount of variation from the mean. variance= sx
2 |
|
Skewness
|
the existence of a minority of atypical, extreme values in distribution of data, which may significantly affect the mean and all measures of dispersion, but not the meadian or mode.
|