Final Flashcards
the way we measure something determines the …
type of data we get
levels of measurement include
nominal, ordinal, interval/ ratio
type of data include
qualitative, ranked, quantitative
qualititative categories
nominal
ranked
ordinal
puts observations on a scale w zero
interval/ ratio
words or codes that represent category
qualitative
that indicate order/ standing
ranked
indicate amount or count
quantitative
what summarizes, organzie data
descriptive stats
examples of descriptive stats
count, central tend, variability, correlatioin
use small sample to est. something and depends on quality and uses hypothesis testing
inferential
what tests are usually used for inferential
anova and t-test
isolated number seperated by groups
discerte
no restrictions and constant change
continuous variables
continous variables rounded
approximate numbers
manipuated by experimenter
independent var
something believed to be influenced by independent
dependent
variable that experimenter have failed to account for that compromise and inter of a study
confounding variable
a comprehensive group
pop
a subgroup that we are using to infer/ est things about a pop
sample
want to know middle (mean, median, mode)
central
variable and standard deviation
varability
sum of all divided by n, very susceptive to skew
mean
data into 2 halves, somewhat susceptive to skew
median
most comon no susceptible to skew and is binned
mode
simplest measure of var.
range
look at difference between values and the mean
variance
square root of variance
standard variance
deviation from mean , squared them, add together is
sum of squares
not affected by outliers
interquartile range
symbol or observatioins
n
one datum
X
mean
x bar
variance
s squared
SD
s
standard normal curve numbers percentages in order
0.1, 2.1, 13.6, 34.1, 34.1, 13.6, 2.1, 0.1
standard normal curve has mean of
0
standard normal curve has a SD of
1
can distributions be skewed and if so how
yes neg or pos
distributions can be peaky or taily
kurtosis
very flat, with long tails
platykurtic
point/peaky
leptokurtic
just right
mesokurtic
kolmogorov- smirnov and sapiro-wilk are tests of
normality
q-q plot is good visual method for doouble checking data especially for
large n
is K-S or S-W better
S-W
describes a relationship between two variables
correlation
Positive relationships are ones
where an increase in one variable
predicts an increase in the other
Negative relationships are ones
where the an increase in one
variable predicts a decrease in the other
most effective way of presenting relationship data
scaterplots
relationships best described by lines
linear relationship
best described w curves
curvilinear
pearson correlation varies from
-1 to 1
pearson correlation
Uses two variables
Variables are both quantitative*
Variable relationships are linear
Minimal skew/no large outliers
Must observe the whole range for each variable
parametric analysis
peason
nonparametic is
spearmeans rank, kendalls tau-b, eta
Random samples are not casual or haphazard,
getting truly random samples requires care
sampling
is the property of a dataset having variability
that is similar across it’s whole range
homoskedasticity
opposite of homoscedastic
heteroskedastic
is used when surveying, to obtain a
“snapshot” of the population
random sampling
is a process used in an experiment to
minimize bias in your experimental groups
random assignment
“Regardless of the shape of the population, the shape of the
sampling distribution of the mean approximates a normal
curve if the sample size is large enough”
the central limit theorem
type 1 error is
false alarm/ false positive
type 2 error
miss/ false negative
assumptions for binomial test
- All cases are mutually independent
- All samples have the same distribution
- You know the probability of the population
level of confidence is often what percentages
95 and 99
what percetnage for level of confidence is weaker
99
are t test parametic or nonparamentric
parametric
When you want to compare a sample mean to
some known or hypothesized value
one-sample t test
If you want to compare two groups to
one another
independent samples t-test
If you want to see how a group changes over time
repeated measures t-test
degrees of freedom for one sample t-test
n-1
degrees of freedom for independent samples t-test
n-2
degrees of freedome for paired samples t-test
n/2-1
he exact significance is calculated from all potential
distributions. It is very computationally intensive but works well with a small N
exact sig
calculated using an estimated curve; inaccurate
for small N but approaches the exact as N grows
aymptotic sig
uses a random process to estimate the
significance using areas under the curve. Less computationally intensive than
exact at high N, but not perfectly consistent
monte-carlo sig
the f-ration is the variability between groups divided by
variability within groups
degrees of freedom for one factor ANOVA
total df is
numbers of scores-1
degrees of freedom for one factor ANOVA between groups df
number of scores -1
degrees of freedom for one factor ANOVA within groups df
number of scores- number of groups
rejection of the null in an anova only means that all the population means are not
equal
when are post hocs done
after main analysis
