Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
EDU 8603
Educational Research
Richard M. Jacobs, OSA, Ph.D.
Statistics...
SEx = √ N - 1
…a smaller standard error indicates
less sampling error
…the major factor affecting the size
of the standard error of the mean is
sample size
…but, the size of the population
standard deviation also affects the
standard error of the mean
The null hypothesis (H0)...
True False
The researcher’s decision
True False
The researcher’s decision
Type II
True Correct Error
Type I
False Correct
Error
researchers use a test of significance
to determine whether to reject or fail
to reject the null hypothesis
…involves pre-selecting a level of
probability, “α” (e.g., α = .05) that
serves as the criterion to determine
whether to reject or fail to reject the
null hypothesis
Steps in using inferential statistics…
True
True and false…
…the difference between two
sample means being a true or
real difference means that the
difference was caused by the
dependent variable and not by
chance
False
True and false…
…the null hypothesis states that
any difference or relationship
found for the samples is the
result of sampling bias
False
True and false…
…the null hypothesis is the
research hypothesis
False
True and false…
…tests of significance deal with
probability not certainty
True
True and false…
…tests of significance enable the
researcher to know for sure
that the researcher’s analysis
correct
False
True and false…
…the researcher makes the decision
to reject or not reject the null
hypothesis with a given
probability of being correct
True
True and false…
…rejecting the null hypothesis
represents the researcher’s
conclusion that the means are
significantly different
True
True and false…
…a significant difference between
means indicates that they are too
different to be the result of random,
chance, sampling error
True
True and false…
…accepting the null hypothesis
indicates that the means are
determined not to be significantly
different, that is, the difference is
due to sampling error
True
True and false…
…researchers must always set the
probability level, α, prior to testing
for significance
False
True and false…
…testing for significance is actually a
matter of comparing the
consequences of making two
possible incorrect decisions
True
True and false…
…with α = .05, the researcher
believes the null hypothesis will be
true 95% of the time
False
True and false…
…as a researcher decreases the
chances of committing a Type I
error, the researcher increases the
probability of committing a Type II
error
True
True and false…
…rejecting a null hypothesis at α = .001
proves the research hypothesis, that
is, the independent variable causes
the dependent variable
False
True and false…
…a “more powerful” statistical test of
significance means that the
researcher is less likely to commit a
Type II error
True
True and false…
…a parametric test of significance
should be used when the data
represent an ordinal or nominal scale
False
True and false…
…generally speaking, a parametric test
of significance should be used when
the data represent interval or ratio
scale
True
True and false…
…a significant F ratio indicates that
there is at least one significant
difference somewhere among the
means but not which one it is
True
True and false…
…when many tests of statistical
significance are performed, the
probability level, α, tends to decrease
because performing a large number
of tests makes it more likely to obtain
significant differences
False
True and false…
…when the chance of finding a
significant difference between means
is increased, so is the chance of
committing a Type I error
True
Fill in the blank…
…an inferential statistic that tells the
researcher how much the researcher
would expect the sample means to
differ if the researcher used other
samples from the same population
standard error of the mean
Fill in the blank…
…a means by which researchers
determine whether there is a
significant of real difference between
the sample means, one due not to
random sampling error
tests of significance
Fill in the blank…
…the statement explaining that the
difference between two sample
means is the result of chance,
random sampling error
null hypothesis
Fill in the blank…
…the type of error when the null
hypothesis is true but the researcher
concludes that it is false
Type I error
Fill in the blank…
…the type of error when the null
hypothesis is false but the
researcher concludes that it is true
Type II error
Fill in the blank…
…the term indicating the probability
that the researcher is correct
level of significance
probability level
Fill in the blank…
…when α = .05, the probability that a
difference is significant will be
accurate within ___ standard
deviations of the sample means
(SEX)
+/- two SEX
Fill in the blank…
…when α = .01, the probability that a
difference is significant will be
accurate within ___ standard
deviations of the sample means
(SEX)
+/- three SEX
Fill in the blank…
…a null hypothesis which states that
one difference can only occur in one
direction requires a ____ test of
significance
one-tailed
Fill in the blank…
…the type of error committed when a
researcher does not reject a null
hypothesis that should be rejected
Type II error
Fill in the blank…
…a statistical test of significance
which determines whether the
observed difference is sufficiently
larger than a difference that would
be expected solely by chance
t-test
Fill in the blank…
…multiple comparisons of the means
that is decided upon before not after
the study is conducted and is based
upon research hypothesis
a priori comparisons
planned comparisons
Fill in the blank…
…the situation where multiple
comparisons of the means cannot be
decided upon before the study is
conducted and is based upon
research hypothesis
a posteriori comparison
inferential statistics
post-analysis considerations
and research reports