WANT: Stats Tutor or Stats Question Solutions

[heart_veemars]

Need help for Assignment for ECON 221 (UWaterloo)

If you took Floyd's Stats course at University of Toronto or if you are taking ECON 221 DE this term, please let me know! Or if you're a Stats tutor in York Region/Thornhill/North York area, PM please, or the other option is if you want to get paid to do my stats assignment for me....

2 questions.. maybe more.

1) A pollster wants to estimate the difference between the proportion of men and women who favour a particular national candidate using a 90% confidence interval of width 0.04. Suppose the pollster has no prior information about the proportions. If equal number of men and woman are to be polled, how large should the sample size be?

2) One of the most pressing problems in high-technology industries is computer security. Computer security is typically achieved by a password - a collection of symbols (usually letters and numbers) that must be supplied by the user before the computer system permits access to the account. The problem is that persistent hackers can create programs that enter millions of combinations of symbols into a target system until the correct password is found. The newest systems solve this problem by requiring authorized users to identify themselves by unique body characteristics. For example, system developed by Palmguard, Inc. tests the hypothesis:

H0: The proposed user is authorized
vs.
H1: The proposed user is unauthorized.

by checking characteristics of the proposed user's palm against those stores in the authorized users' data bank (Omni, 1984)

a) Define a Type I error and a Type II error for this test. Which is the more serious error? Why?

b) Palmguard reports that the Type I error rate for its system is less than 1% where as the Type II error rate is 0.00025%. Interpret these error rates.

c) Another successful security system, the EyeDentifyer, "spots authorized computer users by reading the one-of-a-kind patterns formed by the network of minute blood vessels across the retina at the back of the eye." The EyeDentifier reports Type I and Type II error rates at 0.01% (1 in 10,000) and 0.005% (5 in 100,000) respectively. Interpret these rates.

[Brandon]

2) One of the most pressing problems in high-technology industries is computer security. Computer security is typically achieved by a password - a collection of symbols (usually letters and numbers) that must be supplied by the user before the computer system permits access to the account. The problem is that persistent hackers can create programs that enter millions of combinations of symbols into a target system until the correct password is found. The newest systems solve this problem by requiring authorized users to identify themselves by unique body characteristics. For example, system developed by Palmguard, Inc. tests the hypothesis:

H0: The proposed user is authorized
vs.
H1: The proposed user is unauthorized.

by checking characteristics of the proposed user's palm against those stores in the authorized users' data bank (Omni, 1984)

a) Define a Type I error and a Type II error for this test. Which is the more serious error? Why?

b) Palmguard reports that the Type I error rate for its system is less than 1% where as the Type II error rate is 0.00025%. Interpret these error rates.

c) Another successful security system, the EyeDentifyer, "spots authorized computer users by reading the one-of-a-kind patterns formed by the network of minute blood vessels across the retina at the back of the eye." The EyeDentifier reports Type I and Type II error rates at 0.01% (1 in 10,000) and 0.005% (5 in 100,000) respectively. Interpret these rates.

Not sure if this is right, but this is from what I remember from my stats courses. I'm not at UofT or York, but it's all universal.

a) Type I Error = rejecting null hypothesis when it's true = person who is authorized isn't authorized
Type II Error = not rejecting null hypothesis when it's false = person who is not authorized is authorized

b) 1% of people who are authorized are not authorized by security
0.00025% of people who are not authorized are authorized by security

c) same as b), just change the numbers

For Q1, what is width = 0.04? Is there another term for what this may mean? I don't know if it has to do with the answer, but to use a normal distribution for proportional confidence intervals, you have to have a large sample size (>=30).

[SuperGT]

for # 1, its equal men and women, so let p=.5 and use the standardize formula backwards (solve for n)...
use 90% confidence, find the Z+/- 0.04... thats my hint!

[heart_veemars]

SOLVED, sorry for wasting everyone's time haha

Not sure how to approach it:

Some college professors make bound lecture notes available to students in an effort to improve teaching effectiveness. Two groups of students were surveyed - 86 students enrolled in an eco100 class that required purchase of the lecture notes and 60 students in an eco100 class that did not offer lecture notes. In both classes, the instructors used lectures as main method of delivery. At the end of the semester, sutdents asked to respond to the statement "having a copy of the lecture notes was [would be] helpful in understanding material" Respnses were measured on a 9 point scale where 1 = strongly disagree and 9 = strongly agree.

Results can be summarized as follows:

students buying notes
n = 86
mean = 8.48
standard deviation = 0.94

students not buying notes
n = 35
mean = 7.8
stdev = 2.99

c) construct and interpret a 99% confidence interval for the difference in the population means.

d) would a 95% confidence interval for the difference in the population means be wider or narrower than the one you found in part b? Why?

[Brandon]

Also need help on another question if anyone can help:

Not sure how to approach it:

Some college professors make bound lecture notes available to students in an effort to improve teaching effectiveness. Two groups of students were surveyed - 86 students enrolled in an eco100 class that required purchase of the lecture notes and 60 students in an eco100 class that did not offer lecture notes. In both classes, the instructors used lectures as main method of delivery. At the end of the semester, sutdents asked to respond to the statement "having a copy of the lecture notes was [would be] helpful in understanding material" Respnses were measured on a 9 point scale where 1 = strongly disagree and 9 = strongly agree.

Results can be summarized as follows:

students buying notes
n = 86
mean = 8.48
standard deviation = 0.94

students not buying notes
n = 35
mean = 7.8
stdev = 2.99

c) construct and interpret a 99% confidence interval for the difference in the population means.

d) would a 95% confidence interval for the difference in the population means be wider or narrower than the one you found in part b? Why?

Not too hard of a question, just a basic fill in a formula question. Sample sizes are both greater than 30, so you can assume normal distribution. And with the values you're given, you can easily sub it into any Z value formula for two different sample sizes (with st. dev. and mean being known) to get your test statistic. And as for your Z for your hypothesis test, just lookup in the Z-table for 99% (critical value).

The second formula:
http://en.wikipedia.org/wiki/Statistical_hypothesis_testing#Common_test_statist ics

For any hypothesis testing (non-ANOVA) I learned the 5-step method from my prof:
1. State Ho and Ha (research claim goes into Ha unless using = sign)
2. State significance level and whether test is one or two tail
3. Calculate critical value, draw distribution and highlight rejection area (helps visualize it), state decision rule
4. Calculate test statistic
5. Conclusion

[reqle55]

Its actually an interesting idea that people come online for homework help. I wouldn't be surprised one day if RFD got swamped with HWQ's, and to a point where it may even have its own forum...hehe:lol:

[SuperGT]

i remember back in highschool, going on IRC for tips on physics questions hhehehe