Chapter:
7 . 97103 (done)
, up until but not including "geographic examples of confidence intervals"
“When sample stats represent a larger pop accurately, they are considered unbiased estimators”
point estimation
: stat is calculated from a sample and then used to estimate the corresponding
pop parameter…(ex w probability estimation best estimate for a population mean is sample
mean…etc)
interval estimation
: range that a sample stat is likely in (ex confidence intervalsrepresent level
of precision associated w pop estimate)width determined by –sample sizeamount of variability
in pop probability level or level of confidence selected.
W confidence intervals: upper/lower bound. Confidence level: prob that the interval surrounding
sample mean encompasses the true pop Mean. Significance level (alphaequals total sampling
error): Prob that the interval surrounding sample mean fails to encompass true pop Mean.
8.
all

9.
all
10
. all
11.
skip KolmogorovSmirnov tests (156158 and 163166), cover everything else
12
. focus on
Dr. VasquezProkopec's presentation
; book pages 171176 (do not need to know
how to do the calculations)
Book: Nearest Neighbor Analysis to determine whether a random (Poisson) process has
generated a point pattern. Requires a random sample of points from a population and that sample
points are independently selected.
13.
193201
(up until but not including section 13.3)
14.
all
ENVS 260: Quantitative Methods in Environmental Studies
List of key topics and concepts
Data:
Terminology:
o
Singular (one “data point”) =
datum
o
Plural =
data
Different types of data:
o
Categorical/nominal
(not numerical, not ordered)
o
Ordinal
(may be categorical or integer scale, orderedstrongly or weakly)
o
Continuous
(numerical, not text; with decimal points, not integers)
Measures of central tendency and of dispersion:
What is a measure of central tendency? Mean, median, mode: describe characteristics of the
data…
When does it make sense to focus the mean of a sample? If normally distributed, few
outliers…
When does it not make sense to focus on the mean? If not normally distributed, many
outliers…
**If have greater than about
35%
coefficient of variation (sd/mean or s/xbar) then mean is prob
not best to use…
What are measures of dispersion? Standard deviation, variance
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
How are standard deviation and variance related to one another? Variance= SD squared
Probability distributions:
What are the key features of the normal distribution? Symmetrical, mesokurtic?,
How does a t‐distribution differ from a normal distribution? T distribution if have sample size
less than 30: z dist if sample size greater than or equal to 30. How is the normal
distribution a special case of the t‐distribution? Normal distribution is t with infinite degrees of
freedom…
Central limit theorem
: (law of large numbers)
What does the CLT predict? “Suppose all possible random samples of size n are drawn from an
infinitely large, normally distributed pop having mean Mu and standard dev sigma. The
frequency distribution of these sample means will have:
This is the end of the preview.
Sign up
to
access the rest of the document.
 Fall '10
 BrosiandSize
 Statistics, Normal Distribution, Statistical hypothesis testing, Type I and type II errors

Click to edit the document details