Hypothesis Testing Pg. 491 - 522 Flashcards
_____________________ are statistical tests that follow certain assumptions about the sample and
population they are testing
Parametric tests
____________ and __________ are a numerical way of describing the degree of
spread within a distribution
Standard deviation & Variance
_________________ is the degree to which a
distribution varies around the mean
- The smaller the value, the smaller the _______________-
Variance
____________________ is the typical amount
that each score varies from the mean
- The square root of variance
Standard deviation
_______________________ is an approximately normal distribution constructed
of means calculated from all possible samples of a given size from a given
population
Distribution of means
_______________________ is smaller than the standard deviation of a distribution of
scores
Standard Error
_____________________ is the idea that the world is chaotic and many phenomenon occur at random or by chance
Premise
____________________ is the specific value within a distribution that denotes how extreme the
data, and the sample test statistic, must be
to reject the null hypothesis
Critical Values
_____________________ are the areas of a distribution
beyond the critical value on either one or both
tails
Critical Regions
A ________________ error is rejecting the null hypothesis when it is
true or declaring a difference despite there
being none
Type 1
A __________________ error is failing to reject the null hypothesis
when it is false or declaring no difference when there is
one
Type 2
__________________ is the portion of a given distribution at either
end of the extremes
Alpha (⍺)
As _______ increases, the probability of a ___________ error
increases
⍺, Type 1
__________________ is the probability of correctly rejecting a false null hypothesis when a
particular alternative hypothesis is true
Power (1-β)
_________________ is the probability of committing a Type II error
β (Beta)
As ___ increases, power increases
n
As variance or standard error decrease, ______________ increases
Power
A ________________ is hypothesis testing using a z statistic to
compare two means (a sample and a
population) when population characteristics
are known
Z Test
A ______________ compares the magnitude of a sample mean
to known population mean with unknown
variance
One-Sample t-test
___________________ are an interval estimate, based on a sample
statistic, that theoretically includes the
population mean (μ) a certain percentage of
the time, if we were to sample from the
population repeatedly
Confidence intervals
A ___________________ compares the magnitude of means between two samples when
population parameters are unknown
Two-Sample t-test
A ___________________ Compares the magnitude of difference
between means from two separate matched
samples OR the change in performance on a measure
taken before and after a treatment or
manipulation when population parameters
are unknown
Dependent-Samples t-test
____________________ is an objective and standardized measure of the magnitude of an
observed effect
Effect size
_______________ alerts us to the percent of variance that can be explained by a given
variable
Effect size
_________________ assesses the size of an effect based on the differences between
groups or levels
d-family
____________________ represents a correlation between independent variables (measures of association)
r-family
_____________________ measures the difference (or shift above or below the
population mean as stated by H0) between two
means, and expresses this difference in standard
deviation units
Cohen’s d
___________________ compares the mean difference between
two independent groups
Independent-Samples t-test
The ____________________ tests whether the data from each group is
significantly different
Levene’s Test
___________________ allow us to reduce the influence of individual
differences between our participants (we are only looking at the
difference within the individual between conditions 1 and 2)
E.g., Do the same participants like chocolate more when they eat it with
someone else versus alone?
Paired-Samples t-tests
__________________ determine our direction before
collecting our data, and then we should only look at that specific
direction
One tailed tests
If ____________________________ have a significant t-statistic, then we can look
at the group means to determine the direction
Two tailed tests
_________________ is the theory that using an ideal sample size can bolster your ability to detect differences, should any exist
Power theory
_________________ is the probability of obtaining the observed difference, or one more
extreme, assuming the null hypothesis is true
p Value