Unformatted text preview: Paul Tucker STAT 280 16Jan08 HW 1 1.4 a. HEAVEN YES, DEFINITELY YES, PROBABLY NO, PROBABLY NOT NO, DEFINITELY NOT HELL YES, DEFINITELY YES, PROBABLY NO, PROBABLY NOT NO, DEFINITELY NOT 770 223 90 75 1158 66% 19% 8% 6% 100% b. 85% of the sample surveyed responded that they believe in heaven with some certainty. 74% of the sample surveyed responded that they believe in hell with some certainty. 11% More (146 respondents) people responded that they "definitely" believe in heaven than in 1.10 a. b. c. d. 1.14 a. b. The average of brides for that sample is descriptive. It gives an account "of the middle." The historian is making an inference when she estimates the average of the entire population that period. The population is of all brides in the period 18001820 in New England. 24.1 is a statistic. I would find the poll with the larger sample to be more surprising. Larger samples are more capable of pulling a random sample from the population and therefo more reflective of actual voting behavior. A sample size of 10 may not reflect the entire popu accurately and so the entire sample behaving in a certain way would not be that interesting. 1.20 a. b. c. n=10 n=1000 1 0.90 0.62 2 0.50 0.59 3 0.40 0.58 4 0.50 0.60 5 0.40 0.58 With a larger sample size, the results mirror much more closely the set population proportion. Results in the smaller sample vary greatly compared with the larger. Practical implication: Sample size effects margin of error when inferencing to populations. It makes sense to utilize sufficiently large sample sizes in the real world where possible. prop. of 1s: 0.6 try 1.24 a. This analysis is inferential because it is making a claim about the general U.S. population from relating to only 1001 Americans. The survey did not actually survey every U.S. citizen, theref Gallup had to estimate from the data what the numbers would look like if they actually did sur everyone. 1.26 a. b. Sample: 100 students surveyed Population: All students at Rice University (Undergrads). Descriptive statistics (like averages, standards of deviation, etc) will help the university to und the data, and then to interpret the data. In using the data to make estimates about the entire population, inferencing from the data, the university may then make a policy decision. Descriptive statistics (like averages, standards of deviation, etc) will help the university to und the data, and then to interpret the data. In using the data to make estimates about the entire population, inferencing from the data, the university may then make a policy decision. 1.32 a. I'm interested in economic development. I can use statistics to compare historical data about growth in south east Asia and foreign investment in manufacturing to see if there is a relation that might be useful in making policy elsewhere in the world. categorical quantitative categorical quantitative continuous discrete continuous discrete categorical Fish is the mode at over 40% of food selection. approximately 42% Pareto bar chart 2.4 a. b. c. d. 2.6 a. b. c. d. 2.10 a. b. c. d. 624 214 150 138 1126 55% 19% 13% 12% 100% with some certainty. some certainty. believe in heaven than in hell. ount "of the middle." ge of the entire population from he population and therefore are not reflect the entire population d not be that interesting. set population proportion. encing to populations. It then d where possible. neral U.S. population from data every U.S. citizen, therefore ke if they actually did survey help the university to understand stimates about the entire a policy decision. pare historical data about GDP o see if there is a relationship ...
View
Full
Document
 Spring '08
 Thomas
 a. b. c., Rice University, entire population

Click to edit the document details