Question #2: In their 2014 paper in the Proceedings of the National Academy of Sciences, Jung et al. concluded that hurricanes with traditional female names were more deadly than hurricanes with traditional male names. The authors use data to claim that feminine-named hurricanes cause significantly more deaths compared to masculine-named hurricanes, attributing this to the lower perceived risk and consequently less preparedness for a femininenamed hurricane. U.S. hurricanes used to be given only female names (with a couple exceptions), but in 1979 an alternating male-female naming system was adopted.

a) Open the Excel workbook titled Hurricanes.xlsx. The data set contains data on fatalities caused by 94 hurricanes in the United States between 1950 and 2012. We will explore this dataset to determine if the data supports the claim that femalenamed hurricanes are more deadly than male-named hurricanes.

b) Although we will learn more later, create a scatterplot of the total number of deaths (y-axis) by year (x-axis). Comment on any obvious pattern(s).

c) Temporarily remove the observations for Hurricane Katrina and Hurricane Audrey. How does the scatterplot you made in a) change in appearance? Replace the observations for Katrina and Audrey.

d) Use a PivotTable to determine how many observations there are for female and male hurricanes, separately.

e) Using the data from your answer in d), create a bar graph of the total number of hurricanes (y-axis) by the gender of hurricane names (x-axis). Ensure the graphic has appropriate labels.

f) Make a PivotTable that calculates the mean (average) and standard deviation of the number of deaths for female and male hurricanes. Which type of hurricane, male or female, has the largest mean? Which has the smallest standard deviation?

g) Calculate the five-number summary (p. 34 of the book) for female and male hurricanes separately.

h) Based on your responses in e) and f), which measure, the mean or the median, do you think better represents a typical number of deaths from a hurricane? Why?

i) Make a boxplot of the number of deaths from female and male hurricanes. Describe the trends you see in the boxplot.

j) Based upon the numerical calculations, do you think that the female named hurricanes are more deadly? Why or why not? How could the fact that nearly all hurricanes had Female names until 1979 bias the results? Could extreme values be influencing you interpretation? If so, what should you do next?

Hurricane Year Gender Deaths Easy 1950 Female 2 King 1950 Male 4 Able 1952 Male 3 Barbara 1953 Female 1 Florence 1953 Female 0 Carol 1954 Female 60 Edna 1954 Female 20 Hazel 1954 Female 20 Connie 1955 Female 0 Diane 1955 Female 200 lone 1955 Male 7 Flossy 1956 Female 15 Audrey 1957 Female 416 Helene 1958 Female Debra 1959 Female 0 Gracie 1959 Female 22 Donna 1960 Female 50 Ethel 1960 Female 0 Carla 1961 Female 46 Cindy 1963 Female 3 Cleo 1964 Female 3 Dora 1964 Female 5 Hilda 1964 Female 37 Isbell 1964 Female 3 Betsy 1965 Female 75 Alma 1966 Female 6 Inez 1966 Female 3 Beulah 1967 Female 15 Gladys 1968 Female 3 Camille 1969 Female 256 Celia 1970 Female 22 Edith 1971 Female 0 Fern 1971 Female 2 Ginger 1971 Female 0 Agnes 1972 Female 117Carmen 1974 Female 1 Eloise 1975 Female 21 Belle 1976 Female 5 Babe 1977 Female 0 Bob 1979 Male 1 David 1979 Male 15 Frederic 1979 Male 5 Allen 1980 Male 2 Alicia 1983 Female 21 Diana 1984 Female Bob 1985 Male O Danny 1985 Male Elena 1985 Female 4 Gloria 1985 Female 8 Juan 1985 Male 12 Kate 1985 Female 5 Bonnie 1986 Female W Charley 1986 Male U Floyd 1987 Male 0 Florence 1988 Female 1 Chantal 1989 Female 13 Hugo 1989 Male 21 Jerry 1989 Male 3 Bob 1991 Male 15 Andrew 1992 Male 62 Emily 1993 Female 3 Erin 1995 Female 6 Opal 1995 Female 9 Bertha 1996 Female 8 Fran 1996 Female 26 Danny 1997 Male 10 Bonnie 1998 Female 3 Earl 1998 Male 3 Georges 1998 Male Bret 1999 Male 0