Quiz Over 2nd
Half of Chapter 7 (pp. 144159) of the
6^{th} Edition
Hypothesis Testing
A different quiz covered the 1st half
of Ch. 7 (Steps 1, 2 , 4, & 6)
Step #5: The Criterion for Evaluating the Sample Evidence
 In testing Ho: m1=m2,
what decision would be made (in Step #6) if the 2 sample means are the
same.
 In a correlational study, if we somehow knew that r
= 0.00, should we expect a random sample to yield data such that
r = 0.00?
 "If the difference between the data and __ (Ho/Ha)
is judged to be ___ (small/large), then the sample data are looked upon
as being inconsistent with Ho and, as a consequence,
Ho will be ___ (rejected/retained)."
 In Step #5, the calculated value can be compared against a criterion
number that comes from a statistical table. That criterion number
is called the _____ .
 If the calculated value exceeds the criterion number obtained from
a statistical table, will Ho be rejected?
 (T/F) Normally, the critical value is not presented
in the journal article.
 If a researcher compares the databased pvalue with the level of
significance (instead of comparing the calculated value with a critical
value), Ho will be rejected if p is ___ (smaller/larger)
than the level of significance.
 A small pvalue indicates that the sample data deviate ___ (a little/a
lot) from what would be expected if the null hypothesis was true.
 Which procedure gives the researcher the better chance to reject
Ho?
 comparing the databased pvalue with the level of significance
 comparing the calculated value with a critical value
 neither procedure gives the researcher a better chance of rejecting
Ho
Step #3: Selecting the Level of Significance
 (T/F) The level of significance is like a scientific cutoff
point that allows the researcher to determine whether the discrepancy
between the sample evidence and Ho should be labeled "small"
or "large."
 A researcher should determine the level of significance ____ (before/after)
the sample data are collected.
 What level of significance is used most frequently by applied researchers?
 What lowercase Greek letter denotes the significance level?
(Answer with both a word and a symbol)
 (T/F) The hypothesis testing procedure does not allow one to prove
that the null hypothesis is true, but it does permit one to prove
that the null hypothesis is false.
 What type of mistake is made if a true null hypothesis is rejected?
If a false null hypothesis is not rejected?
 The chances of making a Type __ error are equal to the significance
level.
 If the level of significance is made more rigorous, this change will
decrease the chances of a Type __ error but at the same time increase
the chance of a Type __ error.
 Which kind of error (Type I or Type II) is generally thought to be
worse by the scientific community?
 Do researchers ever choose a level of significance that's more "lenient"
than the popular .05 level?
 In Excerpt 7.23, the statement "p < 0.05" means the same thing as
 a < .05
 a = .05
 a > .05
 (T/F) The chance of a Type II error (like the chance of a Type
I error) is determined solely by the alpha.
 After making a decision about Ho, will the researcher
be able to tell whether his/her decision was correct?
 If a researcher rejects his/her Ho, does the level of
significance indicate the probability that Ho is true?
Results That Are Highly Significant and Near Misses
 (T/F) The level of significance influences the size of the
tabled critical value..
 A researcher might use the phrase "highly significant" if
 p > .999
 p < .001
 (T/F) Whenever you see the notation "p < .01," you
should presume that the researcher set alpha equal to .01.
 A researcher might say that his/her results "approached significance" if
the databased plevel ends up being just slightly ______ (lower/higher)
than the level of significance.
 Suppose a researcher sets alpha equal to .05 and then discovers,
after analyzing the data, that p = .051. In summarizing his/her finding,
what might this researcher might say?
 "There was a trend toward significance."
 "The result was not significant."
 He/she might say either of the things contained in the first two
options.
A Few Cautions
 Alpha should be a number close to 0 if it's referring to _____ but
close to 1.0 if it's referring to _____ .
 Regarding Ho & Ha, which is/are normally
specified in a journal article?
 Ho
 Ha
 Both
 Neither
 If a researcher says that "the data support the hypothesis," to which
hypothesis is he/she probably referring?
 Ho
 Ha
 Hr
 It could be Ho or Ha
 It could be any of one of the 3: Ho or Ha
or Hr
 If Ho is rejected at a very rigorous alevel, is
it OK to conclude that something important has been revealed?
 (T/F) If a finding is "statistically significant," this means that
it's big, important, and useful.
Click here for answers.
