Homework answers / question archive /
QUESTION 1 [15 MARKS]
The following graph appears online in the New York Times article “How Much Warmer Was Your City in 2019?” published on 17 January 2020
QUESTION 1 [15 MARKS]
The following graph appears online in the New York Times article “How Much Warmer Was Your City in 2019?” published on 17 January 2020
Statistics
Share With
QUESTION 1 [15 MARKS]
- The following graph appears online in the New York Times article “How Much Warmer Was Your City in 2019?” published on 17 January 2020. According to the article “Temperature and precipitation data are from AccuWeather. The normal temperature range is determined using the averages of the highs and lows for each day from 1981 to 2010.”
-
- Comment on why this is an example of a good graphic.
- Also, comment on how this graphic can be improved.
(Source: https://www.nytimes.com/interactive/2020/world/year-in-weather.html#assy)
- The following infographic about Covid-19 incidence in the United States appeared on the New York Times website on 30 September 2020. Please note that Porto Rico (P.R.) is an unincorporated territory of the United States.
-
- Explain why this is an infographic.
- Looking at middle panel titled “Where cases are highest per capita” summarise the message in the information presented.
- Comment on whether this infographic achieves the objective of presenting the information quickly and clearly. Justify your answer.
QUESTION 1 CONTINUED
(c) The publication Universities 2018 audits, 31 May 2019, New South Wales Auditor-
General’s Report, Financial audit produced by the Audit Office of New South Wales contained the following graphic on page 26. The following information provides a definition of a repeat finding: “There were 38 repeat findings (24 in 2017) identified in 2018. Repeat findings arise when the university has not implemented recommendations from previous audits. Twenty-eight repeat findings related to IT control deficiencies.”
- Critique this graphic.
- Describe an alternative graphic for presenting repeat internal control deficiencies by area and risk.
[22 MARKS]
(a) The table below comes from a David Spiegelhalter talk that was streamed live on 16 May 2020. Sir David Spiegelhalter OBE FRS is a British Statistician and Winston Professor of the Public Understanding of Risk in the Statistical Laboratory at the University of Cambridge. The data is for England.
- Use it to produce a publication standard graphic.
- What is the message conveyed in the graphic you have produced.
CONTINUED
(b) The following graphic appeared in the article “Top 4 countries exporting COVID-19 cases to Sydney” by Kate Aubusson in the online edition of The Sydney Morning Herald published on 1 October, 2020 – 3:45 pm.
- Critique this graphic.
- Describe a more appropriate graphic for presenting this data.
(Source: https://www.smh.com.au/national/nsw/top-4-countries-exporting-covid-19-cases-to-sydney20201001-p560y1.html)
CONTINUED
(c) Consider the following graphic from the article “No more damned lies…” by David Spiegelhalter which appeared in the New Scientist 15 August 2020 edition. It is presenting a graphical statistical analysis of the age and sex of mass murderer Harold Shipman’s victims. Harold Shipman worked in England as a medical doctor, he killed at least 214 of his patients over 24 years, in the Hyde suburb of Manchester, before his arrest in 1998. David Spiegelhalter worked as one of the statisticians on the Shipman enquiry.
- Critique this graphic.
- What conclusions can you draw from this graphic?
QUESTION 3 [8
Suppose, for the data set mydata.csv containing information on a stratified random sample of people by location, you want to produce a publication standard trellis plot for the variables height (cm) and weight (kg) with gender (Male, Female) recorded at five locations (Albury, Casino, Cooma, Eden, Narrabri) in New South Wales.
- Provide sample code for reading the data into R. Make sure you state any packages you use to do this.
- Provide a rough sketch of the trellis graphic you wish to produce.
- Discuss how you would handle missing values. Provide sample code for doing this.
- Explain how you would use the R package ggplot2 to produce this trellis plot. Provide sample code for doing this.
QUESTION 4 [16 MARKS]
The table below information about the 780 convicts on the First Fleet who departed England on 13 May 1787 arriving to the first settlement in Australia between 19 January 1788 and 20 January 1788. Some convicts had their original death sentence commuted to transportation and so were on the First Fleet.
(Source: http://firstfleet.uow.edu.au/search.html)
Original Sentence
Ship Death Transportation Death Transportation Total
Alexander 63 149 0 0 212
Charlotte 36 51 6 14 107
Friendship 31 45 6 15 97
Lady Penrhyn 0 0 23 79 102
Prince of Wales 0 1 5 45 51
Scarborough 74 137 0 0 211
- Produce a Mosaic plot of this data. Show all intermediate steps.
- What can you conclude from your mosaic plot? Please answer in less than 100 words.
QUESTION 5 [6
The following figure comes from the article “Ecosystem decay exacerbates biodiversity with habitat loss” by Jonathan M. Chase, Shane A. Blowes, Tiffany M. Knight, Katharina Gerstner and Felix May published in Nature, Volume 584, pages 238-243 (2020).
(Source: https://www-nature-com.simsrad.net.ocs.mq.edu.au/articles/s41586-020-2531-2)
- Comment on plot b (has “Standardized nestedness ~ fragment size slope” on horizontal axis).
- Assess whether the straight line model fitted to the data in plot b. can be used to make inference. Justify your answer.
- Do you think a better model can be fitted to the data in plot b? If you think a better model can be fitted explain what it is. Justify your answers.
QUESTION 6 [8
For the following scenarios describe the most appropriate statistical graph/s to use. Give reasons for your choice.
- Skulls are the most commonly found material of marine mammals and can be found in natural history museums. A researcher has recorded various skull measurements on South African fur seals, which are of interest in determining relationships useful for predicting the age of future South African fur seals skulls from easily made physical measurements. The age of South African fur seals is determined from the counts of incremental lines observed in the dentine of tooth sections. For each skull the researcher takes ten linear measurements as well as assessing age from a tooth section. The researcher wants to assess which of these ten measurements which are easier to obtain, are useful to predict the age. These ten measurements are all recorded in millimetres.
- Three similar products have ratings collated from a popular product review website along with gender. One star is the lowest rating that can be given and five stars is the highest rating. This website does not allow half stars.
- Seven hundred dogs are being sampled and the following characteristics are being recorded: height (cm), length (cm), weight (kg), body structure (ectomorphic, endomorphic, mesomorphic), personality trait (Neurotic, Extroverted, Dominant, Impulsive, Agreeable).
- Daily rainfall data recorded for several years at several locations.
