Stat 2000, Section 001, Homework Assignment 2 (30 Points)
(1/16/2015 - Due Friday 1/23/2015 by 9:35am)
- 0) Reading: Section 1.2
- 1) Please work on the following textbook exercises in Moore/McCabe/Craig:
Try to use a computer and software of your choice whenever possible.
Include printouts, screendumps, or photos of your results.
Note that neither CrunchIt nor StatCrunch allow you to produce back-to-back stemplots
as introduced in classe. Those have to be created by hand or by some alternative software.
- 2) Histogram vs Time Plot (5 points):
The file
http://www.math.usu.edu/~symanzik/teaching/2015_stat2000/Yellowstone_2001_2010.csv
was created from data accessible at the US National Park Service (NPS) web site at
http://www.nature.nps.gov/stats/park.cfm
(this URL no longer exists).
This file contains the total monthly number of visitors at Yellowstone National Park (NP) over a 10-year period
(i.e., 120 months) from January 2001 (MonthSeq = 1) through December 2010 (MonthSeq = 120).
Try to use a computer and software of your choice whenever possible.
Include printouts, screendumps, or photos of your results.
(a) Make a histogram of the number of "Visitors" and describe the pattern and
any striking deviations that you see. Create meaningful titles and labels!
(b) Make a time plot of the number of visitors and describe its pattern
in terms of long-term trend and seasonal variation (if any of these is
applicable).
(c) Which of these two plots is better suited to explain the
number of visitors at Yellowstone NP from 2001 through 2010?
Justify your answer.
- 3) Mean and Median (5 points):
Revisit the "Mean and Median" applet at
http://www.math.usu.edu/~schneit/Statlets/center/center.html.
Create the following plots and include printouts, screendumps, or photos of your results:
(a) A plot that consists of at least 10 observations where the mean is at least 3
bigger than the median.
(b) A plot that consists of at least 10 observations where the mean and
the median are exactly the same, but the plot is not
perfectly symmetric (i.e., the left and right sides may be similar, but they can't
be exactly the same).
(c) Calculate the exact values for the mean and the median
for your two previous plots. Show your calculations.
Note that all of your values are whole numbers in the range
from 1 to 10.
- 4) Facebook Friends (8 points):
The file
http://www.math.usu.edu/~symanzik/teaching/2015_stat2000/Survey_2015_Population.xlsx
contains our complete survey data. For this question, we are only interested
in the number of Facebook friends (FBFriends).
It is your task do determine whether men from our class tend to have more Facebook friends
than women (or at least, whether they may claim that they have). You should do this as follows:
- Create at least two relevant plots for each gender and discuss the
shape, center, and spread of the data, based on these plots.
Make sure that all plots have titles, that axes are labeled, and so on.
- Create relevant numerical summaries. Justify which measure for the
center (mean or median) might be better suited here.
- Discuss whether there are outliers. Do this manually, using the numerical
summaries and see whether your software automatically displays outliers or whether
you need to change some settings to display outliers (if any exist).
- Write a computer-based conclusive report of your results (a total of two to three pages,
including all figures).