17.1 Matched pairs
17.2 Two independent samples
17.3 Single sample
17.4 Two independent samples
17.8
a) If the loggers knew that the effects of logging would be assessed, they may log in such a way as to try to minimize the effect. If they did not normally log in this manner, it would not be possible to generalize to logging practices at other times and locations.
b) H0: μLUL=0, there is no difference in the mean number of species between logged and unlogged plots 8 years after logging.
Ha: μL - μUL< 0, the mean number of species in logged plots is less than the mean number of species in unlogged plots 8 years after logging.
The t-statistic is 2.1141. Using the conservative value of 8 degrees of freedom, the p-value is between .05 and .025 for the one-sided alternative. The test is significant at the .05 level, but not the .025 level.
17.10 A 90% confidence interval (using a conservative 8 degrees of freedom) for the difference in the mean number of species between logged and unlogged plots is 3.8 ± 1.86*1.81 or (.433, 7.27).
17.12 The two-sample t confidence interval for the difference in mean yields may not be accurate because the sample size is smaller than the recommended minimum of 5 in each group.
17.24
a)H0: μFM, the mean score for females is the same as the mean score for males.
Ha: μFM, the mean score for females is lower than the mean score for males.
The test statistic t=6.13. The p-value is less than .0005. There is strong statistical evidence that the mean score on the interview questions is lower for women than men.
b) A sample of drivers selected by random digit dialing may not be representative of the population of drivers.
17.26 H0: μ911, the students in grades 9 and 11 have the same mean score on the CDI.
Ha: μ911, the students in grade 9 have a mean score lower than those in grade 11 on the CDI.
The test statistic t=1.9. Taking 69 as the conservative degrees of freedom, the p-value for the one-sided test is between .05 and .025. Therefore, there is statistically significant evidence that 11th graders score higher on the CDI than 9th graders.
17.27 H0: μMF, male and female students have the same mean score on the CDI.
Ha: μM≠μF, male and female students have different mean scores on the CDI.
The test statistic t=.25. Taking 103 as the conservative degrees of freedom, the p-value for the one-sided test is greater than .25. Therefore, there is no significant evidence that male and female students have different mean scores on the CDI
17.28 H0: μTC, The mean improvement in the treatment group is the same as the mean change in the control group.
Ha: μTC, the mean improvement for the treatment group is greater than the mean improvement for the control group.
The test statistic t=1.91. The approximate degrees of freedom is 13.919. The p-value is 0.038, so the improvement of those in the receiving the treatment was significantly greater than the improvement of those receiving the neutral message. This is true at both the 5% and 10% level.
17.30 A 90% confidence interval for the mean difference in gains between treatment and control groups is (0.249, 6.051).
17.38
a) The mean score for boys is 110.96 and the mean score for girls is 105.84.
b) For Boys:
779
8
90377
100234556667779
1100001123334556899
120334467788
136
For Girls
724
869
91368
10023334578
111122244489
1208
1302
c)H0: μBG, the mean score of boys on the IQ test is the same as the mean score of girls.
Ha: μBG, the mean score of boys on the IQ test is the greater than the mean score of girls.
The test statistic t=1.64, with an approximate degrees of freedom of 56.9. The p-value=0.052.
d) It would be important to know whether the students who took the test are in fact representative of all 7th grade students in the district.
17.40 The critical value for a 90% confidence interval given a t-distribution with 56.9 degrees of freedom is 1.67. A 90% confidence interval for the difference between the mean IQ scores of all boys and girls in the district is (-0.088, 10.325).
Math 280 | Teaching | Home