Normal dist + X2 dist = t dist!
t = (xbar - mu) / (s / sqrt(n) ~ t (n-1)
o degrees of freedom: (all of the things we CAN use to estimate - the number of
things were estimating)
COMMANDS:
o To access Rcmdr: in R, library(Rcmdr)
o To name variables: name = input
ex. data = 2
ex. data <- 4
You can use periods and numbers in names, like data.new3e5 (as long
as the first character is a letter) which I guess you cant do in SPSS. H
Intro to Statistics
o Definition:
Collecting
Organizing
Displaying
Interpreting
Analyzing
Guessing
about/for data.
plural of datum
Latin for given
objective, not subjective (thats where interpretation comes in)
generic term for numerical infor
Where does Cocaine come from?
Erythoxylon Coca. Grown in Andes, high altitudes. Alkaloid when chewed freed from organic
compound. Leaves contain 1% cocaine
Fame Cocaine
Vin Mariani endorsed by the pope. Freud's Uber Coca (cure to gastric disorder, asthma,
4MethodGoalsofScience
Describeaneventorphenomenon
Discoverrelationships
Prediction
Control
InductiveGoals
Describeaneventorphenomenon
Discoverrelationships
DeductiveGoals
Prediction
Control
Prediction
Guess;seeifitcomestrue
Control
Gettingbacktoanswers;if
correlational research
measuring two variables and determining whether they are associated with each other (studies relating
smoking to lung cancer are correlational, for example)
prospective research
a research strategy in which people are followed forwa
health psychology
the subarea within psychology devoted to understanding psychological influences on health, illness,
and responses to those states, as well as the psychological origins and impacts of health policy and
health interventions
health
the abse
cognitive restructuring
a method of modifying internal monologues in stress-producing situation; clients are trained to monitor
what they say to themselves in stress-provoking situations and then to modify their cognitions in
adaptive ways
self-talk
inter
clinical breast exam (CBE)
a thorough physical examination of the breast by a health care professional to detect changes or
abnormalities that could indicate the early signs of breast cancer
aerobic exercise
high-intensity, long-duration, and high-enduran
Most common dist. shapes:
o symmetric - if right/left sides of histogram are approx. mirror images of each other
o skewed to the right - right is longer and shorter tail! (shorter = skewed)
o ^same for left, but obviously the opposite scenario
Complex vs
Section 1.3
Density curves - mathematical models of dists.
o Total area under curve, by definition, is equal to 1, or 100%
o Area under the curve for a range of values is the proportion of all observations for
that range
a histogram with a curve juxtapo
When talking about probability of an event happening under a Normal symmetric density
curve:
o 1 st. dev.:
P (mu - sig [< or =] X [< or =] mu + sig) = .68
o 2 st. devs.:
P (everything above) = .95
o 3 st. devs.:
P (everything above) .997
Standard Norm
Nonparametric method: the sign test:
o A distribution-free test usually makes a statement of hypotheses about the median
rather than the mean
are the medians different?
makes sense when the distribution may be skewed
less reliable than parametric meth
Choose one- or two-sided tail test BEFORE performing the test, NOT based on results or
anything
The p-value is the chance of obtaining a particular random sample result IF the null
hypothesis is true, as quantified by a test of statistical significance.
Statistical Inference!
Methods:
o Confidence intervals - estimating a value of a population parameter
o Tests of significance - assessing evidence for a claim about a population
Each sample has its own individual sample mean (xbar); if you keep drawing s
If the sample size is much smaller than the size of the population w/ proportion p of
successes, then the mean and SD of p^ are:
o MUp^ = p
o SDp^ = sqrt(p(1-p)/n) = SDx/n
o Because mean is p, we say that the sample prop in an SRS is an unbiased
estimato
Conditional Probabilities
o probability of event B given event A (provided that P(A) is NOT 0) =
P(A and B) / P(A)
consequently, P(A and B) = P(B|A) P(A) when events are not independent
when independent, P(A and B) is just P(A) P(B)
ex. probability o
Xbar = arithmetic average of a set of observations
Mu = weighted average of possible values of random variable X
o ^this reflects the fact that all outcomes might not be equally likely
o mean of discrete random variable
Find this by multiplying Xi by Pi
P(A or B) = P(A U B) = P(A) + P(B)
o ^addition rule for disjoint events
IF NOT DISJOINT:
o P (A U B) = P(A) + P(B) - P(A and B)
Finite sample spaces deal with discrete data - data that can only take on a limited number
of values.
We can assign probabilit
regression line - straight line that describes how a resp. var. y changes as an exp. var. x
changes
o use it to predict value of y for given x
o ^distinction between exp. and resp. vars. is important here
if you exchange them, youll get the wrong line
o
Marginal dist. of two-way table:
o ex. marg. dist. of gender
o ex. marg. dist. of class rank
Joint dist. of two-way table:
o ex. joint dist. of gender and class rank
Conditional dist. of two-way table:
o ex. Given freshman, .4 are male, and .6 are female
Correlation
o coefficient = r
o r doesnt distinguish between x and y
treats them symmetrically (see the formula; theyre treated equally)
o r has no units of measurement
we get rid of our units when we standardize
o r ranges from -1.0 to 1.0
o its assum
health belief model
a theory of health behaviours; the model predicts that whether a person practices a particular health
habit can be understood by knowing the degree to which the person perceives a personal health threat
and the perception that a partic