Statistics in a Modern World 800
Solutions to Exam 2

1.  
   1. 1st, 4th, average of 7th and 8th, 11th, 14th = 1, 2, 3, 15, 30
   2. (iii) skewed right/high outliers
   3. (iii) greater than the median
2.  
   1. (i) histogram (for values of one measurement variable)
   2. (iii) scatterplot (for relationship between 2 measurement variables
   3. (ii) piechart (for one categorical variable)
   4. (ii) barchart (for two measurement variables)
3. (iii) +1 (as students get older, their year at Pitt gets higher, with few exceptions)
4.  
   1. z=(95-115)/13=-1.54; the proportion below is .06 or 6%
   2. highest 1% have 99% below, so z=+2.33, observed value = 115 + (2.33)(13) = 145.29
5.  
   1. predicted height = 44.2 + .37(65) = 68.25
   2. (i) 60 is outside the range of x-values used to produce the regression line
   3. r=.5 remains unchanged because it is independent of unit of measurement
6. $5(152/167)=$4.55 [Note that the 1995 amount should be less than the 1999 amount of $5.]
7. I think (c) is the case, and the graph is cluttered with chart junk. However, I did not take points off if you believe the graph displayed (a) or (b).
8.  
   1. (7)
   2. 1
   3. 2
   4. 4 (the chemical emulsifiers are a confounding variable)
   5. 5 (but I accepted 6 too)
   6. 3 (chocolate is a contributing, but not sole, cause)
   7. 6 (but I accepted 5 too)
9.  
   1. (i) being male or female may affect your decision to live on or off campus, but living on or off campus will not affect your gender
   2. calculate (column total * row total)/table total: 8, 13, 18, 30
   3. 2, 1.2, .9, .5
   4. 2 + 1.2 + .9 + .5 = 4.6
   5. 4.6 > 3.84, so we have statistical evidence of a relationship
   6. The proportion of males living on campus is 12/26=.46, whereas the proportion of females living on campus is 9/43=.21. This difference, along with the answer to (e), lend support to Nancy's claim

[ Home | Calendar | Assignments | Handouts ]