You are on page 1of 6

Week 1

1. Study 2
2. μ1 −μ2=0; μ_1 > μ_2
3. If the null hypothesis is true, the chance of finding that people’s voting habits are
influenced by pictures of baby animals is less than 2.5%
4. We can reject the null hypothesis because there is a 5% chance that this would happen if
the null hypothesis was true.
5. Type I error
6. Increase the level of alpha Reduce the variance in the population Use more reliable
measurement instruments
7. 4.679102
8. 2.125
9. Yes
10. 4.8235368
11. 31.2388
12. 48.7612
13. For option 1 we would use an independent groups analysis. For option 2 we would use a
“dependent” or “matched” groups analysis.
14. 1.517929376
15. xd = 0.4, t = 0.3888, fail to reject null hypothesis in two-sided test.
16. Mediator, moderator, confounder
17. You would reject the null hypothesis.
18. You would fail to reject the null hypothesis

Week 2
1. We want to see if gender and drink preference are independent
2. Neighborhood2
3.

Wrong
Our null hypothesis is that neighborhood and type of crime are independent.
If neighborhood and crime are not independent, we can calculate the observed
marginal frequencies by multiplying the joint frequencies, divided by the sample
size.
If the expected frequencies under the null hypothesis are significantly different
from the observed frequencies, then the categories are dependent.

Wrong
Our null hypothesis is that neighborhood and type of crime are independent.
If neighborhood and crime are independent, we can calculate the observed joint
frequencies by multiplying the marginal frequencies, divided by the sample size.
If the expected frequencies under the null hypothesis are significantly different
from the observed frequencies, then the categories are independent.

Wrong
Our null hypothesis is that neighborhood and type of crime are dependent.
If neighborhood and crime are not independent, we can calculate the observed
marginal frequencies by multiplying the joint frequencies, divided by the sample
size.
If the expected frequencies under the null hypothesis are significantly different
from the observed frequencies, then the categories are dependent.

Wrong
Our null hypothesis is that neighborhood and type of crime are dependent.
If neighborhood and crime are not independent, we can calculate the observed
marginal frequencies by multiplying the joint frequencies, divided by the sample
size.
If the expected frequencies under the null hypothesis are significantly different
from the observed frequencies, then the categories are independent.

Wrong
Our null hypothesis is that neighborhood and type of crime are dependent.
If neighborhood and crime are independent, we can calculate the observed joint
frequencies by multiplying the marginal frequencies, divided by the sample size
If the expected frequencies under the null hypothesis are significantly different
from the observed frequencies, then the categories are dependent.

Wrong
Our null hypothesis is that neighborhood and type of crime are dependent.
If neighborhood and crime are independent, we can calculate the observed joint
frequencies by multiplying the marginal frequencies, divided by the sample size
If the expected frequencies under the null hypothesis are significantly different
from the observed frequencies, then the categories are independent.

4. Wrong - 3.447173
Wrong - 0.443
Wrong - 2
5. None of the above (not sure)
6. 3.3106
7. 0.361
8. Republican and democratic voters seem to have cats and dogs, while independent voters
have other pets
9. 3, 12, 15
10. No answer
11. Visitors come to the cafeteria at random, not all visitors buy anything
12. He had too few cases to reliably reject the null hypothesis, the sampling was not at random
13. You would reject the null-hypothesis at a significance level of 0.05
14. Table 2
15. The frequency of correctly identifying a liar is equal to the expected frequency of 6.
Correct and incorrect responses are independent from lying and not lying.

Week 3
1. predictor response
2. 2635
3. --
4. wrong
5. https://www.algebra.com/algebra/homework/Probability-and-statistics/Probability-and-
statistics.faq.question.1073405.html D
6. 6Because we assume the dependent variable is normally distributed at each level of the
independent variable in the population and the line goes through these means.
7. R-squared
8. --
9. --
10. a c
11. Yes, 0 is inside the confidence interval.
12.
13. Outliers
14. In 95% of repeated samples, the population mean shoe size will lie in the interval, so we
find it plausible that mean shoe size lies between 10.65 and 18.32.
15. Prediction Interval
16. μy=αβx
17. wrong
Week 4

1. Scariness = a + 1.6*(blood) + 2.8*(teeth) - 3.2*(cuteness)

2. The model’s R-squared.


3. 5.072
4. --
5. 0.227918
6. Blood, teeth and cuteness can account for less than half of the variation in monster
scariness.
7. 5
8. 5.518868
9. –
10. 1.8
11. 1.7011 wrong
12. 4.47 wrong
13. No
14. Wrong- bde
wrong – bcd
wrong – bcd
wrong – bce
wrong – bcf
wrong - abd

15. -3 or less
16. The intercept is 2.6 units higher when cash is offered rather than vouchers.
17. 4
18. The odds of voting will change by a multiplicative factor of 3.210 when Conscientiousness
increases by 1.
19. --
20. 0.89
21. 0.82

Week 5

1. We retain statistical power


2. Each plant group may come from a different population distribution
3. 1.583333
4. –
5. The researcher cannot reject the null hypothesis.
6. Collect more data and rerun the study.
7. 6 tests, 0.008
8. –
9. --
10. 2 x 3 factorial design.
11. Interaction between chemical and insect.
12. The difference between mean growth with each pesticide will be diff with each pest
group
13. yes, provided that all other assumptions are met.
14. The interaction between gender and student status is significant, but the main effect of
gender is not significant.
15. –
16. --

Week 6
1. Parametric
2. The order of runners up
3. 1 3
4. –
5. The median rating is not significantly different from 2.
6. ?
7. ----
8. Wrongx
9. At least one breed of cat has a significantly different size.
10. If he had more cats
11. Plot 3 is monotonic, plot 2 is linear.
12. -
13. There is a weak positive relationship between study hours and grades.
14. 8
15. No

Week 7
1. Ttest for 2 ind groups
2. Correlation/Bivariate regression
3. Chi-square test
4. Kruskal-Wallis
5. –
6. Ordinal
7. Logistic regression
8. Atleast one group mean is significantly diff from the other
9. he prediction interval at a particular value of x.
10. in the left critical zone, so ho is rej
11. Exponential regression
12. 8.93
13. Main effect of faculty
14. (0.039, 0.223)
15. Procrastination, gender
16. Made a Type II error
17. Taking a larger sample, increasing the level of significance
18. 2.88
19. The correlation coefficient significantly differed from the value of 0 at a significance
level of 0.05 with two-tailed test.
20. The regression model is significantly better than the model that only uses the average
of y.
21. 8.5923
22. Wrong
23. 4.6183??
24. 15.7333
25. 2.9022
26. 0.6124
27. 6.5709
28. 0.925387993
29. 0.3142606

You might also like