Hi guys, can you please help with this questions and explain your working! Thank you!
Q3 [10 marks] It is just over 100 years since the Titanic hit an iceberg on its maiden voyage and sank. There are questions about the exact number of deaths, but this disaster rates as one of the most famous of all time due to the Titanic's supposed \"unsinkability\". The structure of the data is a follows: 'data.frame': 16 obs. of 4 variables: $ Survived: Factor w/ 2 levels "No","Yes": 1 2 1 2 1 2 1 2 1 2 $ Class : Factor w/ 4 levels "1st","2nd","3rd",..: 1 1 2 2 3 3 4 4 1 1 $ Sex : Factor w/ 2 levels "Male","Female": 1 1 1 1 1 1 1 1 2 2 $ Freq : num 118 62 154 25 422 88 670 192 4 141 The class variable is often used as a substitute for economic class, but is actually linked to the location of the cabin in which the passenger was sleeping. We wish to investigate the relationships between Survived [1], Class [2] and Sex [3]. In this analysis we wish to know if these three factors are associated or not using Loglinear models. The outcome from fitting all possible log-linear models (using the three factors in the order given above) was: Model df Deviance 1 [1][2][3] 10 1018.32 2 [1][23] 7 605.72 3 [13][2] 9 583.86 4 [12][3] 7 837.42 5 [12][23] 4 424.82 6 [13][23] 6 171.25 7 [12][13] 6 402.95 8 [12][13][23] 3 65.18 9 [123] o 0.00 a) Which model is the best one to explain the interrelationships among the three factors? Justify your selection. [2 marks] b) What table (or tables) of the data would you recommend be constructed that will show the interactions from your chosen model. [2 marks] c) What would the difference between models 2 and 5 be testing and what conclusion would you draw from this comparison? [2 marks] d) Which pair(s) of models test the association between Survived and Sex? [2 marks] e) Given your answers above, what table (or tables) would you recommend presenting to the researcher to justify your preferred model. [2 marks] Page 3 of 11 C05