This graph is referred to as
stem-leaf plot.
histogram.
box plot.
bar chart.
none of the above.
Which one of the following statements is correct?
Gender and score are nominal scales.
Score is an interval scale.
Gender and brands of DVD player are categorical variables.
Price and score are continuous variables.
None of the above.
The 80th percentile of ease of use score for the sampled female clients is
0.8
4
9
0.2
None of the above.
Which one of the following statistics can not measure the variability or
dispersion of a data set?
interquartile range.
variance.
covariance.
standard deviation.
coefficient of variation.
An investigator carries out a large sample two-sided test of the null hypothesis
that μ = 100 . He reports the value of test statistic of 2.12 with a p-value of
0.034.
Which of the following statement is correct?
The 99% confidence interval for μ covers 100.
The null hypothesis is rejected at α = 0.01 .
The 95% confidence interval for μ covers 100.
The null hypothesis can not be rejected at α = 0.10 .
None of the above.
An investigator carries out a large sample two-sided test of the null hypothesis
that μ = 100 . He reports the value of test statistic of 2.12 with a p-value of
0.034.
Let z = 2.12 . Which one of the following formulae is used to compute the
p-value?
P(Z > z)
P(Z > 2z)
P(Z > z) / 2
2P(Z > z)
P(Z > z) − P(Z < −z)
The data set collected here is based on
simple random sampling.
clustered sampling.
stratified sampling.
completely randomized design.
randomized complete block design.
Based on the descriptive statistics given above, the correlation of the sales
between display X and display Z is
0.4513
0.8065
0
4.8392
0.9678
How can we interpret this correlation?
No meaning for this study, since this is not a paired data set.
A mistake, since the correlation must be less than one.
A mistake, since the correlation should be negative.
If sales of display X increase then sales of display Z increase too.
There is a strong interaction between display X and display Z .
What is the best reason for randomly assigning treatment levels to the
experimental units?
Randomization makes the experiment easier to conduct.
Randomization will tend to average out all other uncontrolled factors so
that they are not confounded with the treatment effects.
Randomization is required by statistical consultants before they will
help you analyze the experiment.
Randomization makes the analysis easier because the data can be
collected and entered into the computer in any order.
Randomization is used to remove the effects of another factor from the
comparison.
If sales are normally distributed, which testing procedure would you use to
compare the average sales from these displays?
F test in a one-way ANOVA table for completely randomized design.
F test for “displays” in a two-way ANOVA table for block design.
Chi-square test for a 3 by 6 contingency table.
Paired t-test for a block design.
Two-sample t test for independent random samples.
the average customer used the equipment for 30 minutes last week.
who has owned the equipment an extra month used the equipment 30
minutes less last week than the average customer who has owned it one
month less.
who just bought the equipment used it 30 minutes last week.
who bought the equipment one-half month ago.
None of the above.
How can we interpret the number 10 in the equation?
The average hours that the customer used the equipment last week is
10.
The maximum hours that the customer used the equipment last week is
10.
The median hours that the customer used the equipment last week is
10.
It is meaningless for this study.
None of the above.
If one customer has owned the equipment for 2 years, how many hours he
or she used this equipment last week?
−2 hours.
We should not use this regression to make such prediction since the
value of explanatory variable (X) is outside the range of X when
this regression line is constructed.
The customer will not use the equipment any more.
2 hours.
None of the above.
Which one of the following is not the assumption of the error terms in a
simple linear regression?
The error terms are independent.
The error terms have constant variances.
The error terms are normally distributed.
The error terms have zero expected value.
None of the above.
Any reason(s) why we should question the use of a simple linear regression
for this data set for prediction?
No, the assumptions required by the analysis of a simple linear
regression are all satisfied.
No, the R-square is very high.
Yes, the sales may be serially correlated.
Yes, the relationship between years and sales is not linear.
None of the above.
A survey shows that 30% of the fashions that were found to be unprofitable
were marketed by the major fashion clothes stores; 60% of the fashions
found to be profitable were marketed by the major fashion clothes stores. If
70% of all fashions are profitable to market,
(a) What is the probability that the major fashion clothes stores market a
particular fashion? (3%)
(b) Find the probability that a fashion will be profitable if the major fashion
clothes stores market it. (5%)
可觀看題目詳解,並提供模擬測驗!(免費會員無法觀看研究所試題解答)