DOC PREVIEW
ISU STAT 401 - Lab 12

This preview shows page 1-2 out of 5 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

1Stat401E Fall 2010Lab 121. In 1928 Karl Mannheim argued that people are most impressionistic during their late teens and early twenties. Subsequent research has shown that the age at which people are sensitive to (or impressed by) events differs according to the nature of the events. For example, one body of research has shown that it is around age 19 when people are most sensitive to (and thus more likely to remember) historical events (such as the bombing of Hiroshima).You are using data collected by the National Opinion Research Council in 1974. Your sample consists of 123 Americans, who were asked (among other things),"How vividly do you remember the bombing of Hiroshima?"You name this variable "recall". (Possible values on this variable are 1=do not remember the Hiroshima bombing, 2=barely at all, 3=somewhat vividly, 4=very vividly, and 5=extremely vividly.) Note: You should assume that this variable is a ratio-level measure.Since you also have data on each respondent's year of birth (yob) and you know that Hiroshima was bombed in 1945, you can calculate each respondents' age at the time of Hiroshima (hiroage) as follows:compute hiroage = 1945 - yob.a. What complete and reduced regression models (i.e., regression equations) would you use to evaluate whether Americans are more likely to have recalled the bombing of Hiroshima if they were 19 years old in 1945, than if they were older or younger than 19 at the time. (Do NOT attempt to find parameter estimates [i.e., no numbers please].) What does Mannheim's theory suggest will be the signs of all slopes in the complete model. (Explain your answer. Keep in mind that Mannheim's theory implies nothing about the marginal effect of "hiroage" on "recall".)b. The computer output from the two regressions described in part a, contains the following information:For the reduced model: Model Summary.16669Model1R Square2For the complete model: Which regression model provides the most parsimonious depiction of Americans' recall of the Hiroshima bombing? (Use the .05 level of significance.)c. After performing the above analysis, you learn that older Americans are more likely to recall historical events than younger ones. To evaluate these "maturity effects" you must create a new variable that measures age. Give SPSS, R, or SAS code that would create a variable (age) that measures each respondent's age at the time of the survey. What is the correlation between "age" (the variable you have just created) and "hiroage"? (Explain your answer. Do NOT provide a mathematical proof.)d. How might you adjust "recall" such that Americans' recall of the Hiroshima bombing no longer varies according to the linear effects of the variable, "age"? (Hint: No need for numbers here. Just explain how you would go about doing this.)e. A colleague suggests that you regress "recall" on both "hiroage" (to detect the linear effect of subjects' ages at the time of the bombing of Hiroshima) and "age" (to detect the maturity effects described above). Explain why this would NOT be a good idea.2. Consider the following three variables from a sample of 126 ISU students:ANTICIPATE (A) = "joy in knowing STAT 401 is almost at an end" on a scale from 0 = no joy to 100 = extreme joyNEAREND (N) = "number of STAT 401 classes left to be attended" on a scale from 0 = no more classes to attend to 45 = forty-five more classes left to be attendedREQUIRED (R) = "if STAT 401 was taken as a required course" on a scale from 0 = not required to 1 = requiredSome data: rAN = -.22 rAR = -.58 rNR = .60Some regression models:Model a: Model b: Model Summary.19836Model1R SquareAˆaaˆabˆa1N+=Aˆbaˆbbˆb1R+=3Model c: Model d: Model e: Model f: Model g: Model h: Model i: Model j: Model k: Model l: Model m: Model n: Model o: Model p: Model q: Model r: a. Indicate which of the models (i.e., Model a through Model r [or none of these]) would be used as the complete and reduced (if applicable) models in testing each of the following research questions: Model(s) usedComplete Reduced Question_______ _______ Does joy increase as each STAT 401 class is finished (and thus as there is one fewer STAT 401 class left to be attended)?_______ _______ Does joy increase at a faster pace as there are fewer and fewer STAT 401 classes to be attended?Aˆcaˆcbˆc1RR–2+=Aˆdaˆdbˆd1NN–2+=Aˆeaˆebˆe1RR–NN–+=Aˆfaˆfbˆf1Nbˆf2R++=Aˆgaˆgbˆg1Nbˆg2NN–2++=Aˆhaˆhbˆh1Nbˆh2RR–2++=Aˆiaˆibˆi1Rbˆi2NN–2++=Aˆjaˆjbˆj1Rbˆj2RR–2++=Aˆkaˆkbˆk1Rbˆk2RR–NN–++=Aˆlaˆlbˆl1Nbˆl2RR–NN–++=Aˆmaˆmbˆm1Rbˆm2Nbˆm3RR–2+++=Aˆnaˆnbˆn1Rbˆn2Nbˆn3NN–2+++=Aˆoaˆobˆo1Rbˆo2Nbˆo3RR–NN–+++=Aˆpaˆpbˆp1Rbˆp2Nbˆp3NN–2bˆp4RR–NN–+++ +=Aˆqaˆqbˆq1Rbˆq2Nbˆq3RR–2bˆq4RR–NN–+++ +=Aˆraˆrbˆr1Rbˆr2Nbˆr3NN–2bˆr4RR–2bˆr5RR–NN–+++++=4_______ _______ In comparison with students who take STAT 401 as an elective (i.e., not as a required course), do students who take STAT 401 as a required course exhibit a greater increase in joy for each remaining STAT 401 class to be attended?(Hint: One of these three research questions does not call for a reduced model.)b. Selecting from among Model a through Model r , a t-test would be preferable to which one of these regression models?c. Seven of the regression models are misspecified (i.e., ill-formulated). Which are these?d. Each of the seven models (referred to in part c) is misspecified for the same reason. What reason is this?e. Referring to Model f, assume that and that . Give an interpretation of what each of these estimates means in words.3. You are doing research on how well people manage their money. Some research has shown that drug abusers manage their money worse than people who do not abuse drugs. The central theoretical argument in this research is that drugs impair people's judgment and make them less capable of making good decisions regarding their families' finances. As a social worker, you find yourself working with 21 drug-free families (i.e., families in which no member abuses drugs), 11 crack families (i.e., families in which at least one member is a regular user of crack-cocaine), 5 heroin families (i.e., families in which at least one member is a regular user of heroin), and 3 amphetamine families


View Full Document

ISU STAT 401 - Lab 12

Download Lab 12
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Lab 12 and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Lab 12 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?