PYB350 Advanced Statistical Analysis/MIS770 Foundation Skills in Data Analysis/COMM1190: DATA, INSIGHTS, AND DECISION/AERO4560 Flight Mechanics 2

Tuesday 4th October 11:59 pm, 2022

Assignment Resources:
The Assignment folder in the PYB350 Blackboard site assessment tab contains the following resources for this written assignment:
· An Unfinished Manuscript: this comprises the introduction and methods sections of a research report. This unfinished manuscript provides a brief rationale and background to an experimental study (quantitative data) with an open ended question (qualitative data) on parents’ attitudes to risky play and the effect of playground type and supervision on children’s risky play behaviour. (You need to write up the qualitative and quantitative results, and a preliminary discussion.)
· A link from which you can download your raw data for the study.

· A document with the participants comments from the open ended question.

The assignment has been very broadly drawn from the reference below that may will give you some basic background to the study. You are NOT expected to read this reference it just there for some general background. You do not need to research the content to complete this assignment, everything you need is provided in the unfinished manuscript.
Little, H., Wyver, S., & Gibson, F. (2011). The influence of play context and adult attitudes on young children's physical risk‐taking during outdoor play. European Early Childhood Education Research Journal, 19(1), 113-131.
Assignment Task:

Your task is to complete the manuscript, by writing the up the RESULTS (Qualitative & Quantitative), and a preliminary DISCUSSION section for the report, in APA 7th format (Maxium 10 pages including title page not including references and appendix). To do this, you should:

· Conduct a simple thematic analysis on the comments provided. This analysis should result in the identification of 1-3 key themes (and possibly some sub themes underlying these) with example quotes to support the theme. When reporting your qualitative results please ensure “quotes” are clearly distinguished from your text, using quotations marks, italics or separated text.

· Enter the raw data provided on the link in the assignment folder on Blackboard into an appropriately formatted data file in SPSS, and carry out the analyses required to report on the results of the quantitative study. (NOTE: Do NOT assume that your data is the same as that of your fellow students, the link takes you to individual data sets.) You must enter your own data, and upload your data file to Blackboard when you submit your assignment. Your data file should be appropriately formatted, including value labels for any categorical variables.
· Complete a Results section for the study, with separate sections for the qualitative and quantitative analyses.
· In your qualitative results section you will need to describe your key themes and any subthemes and provide evidence for these in the form of quotes from the transcripts.
· In your quantitative results section, you should provide descriptive information about the variables, if assumptions for your analysis are met by the data (you can assume your analysis is robust to any breaches of normality), the main analyses, and any follow up analyses required to test the hypothesis. Your results section should include a figure illustrating your findings, correctly formatted in APA 7th format.
· Complete a preliminary discussion. You are NOT expected to do further background reading about the phenomenon under investigation. So you are NOT expected to write the final part of the discussion, where you locate this research in the broader empirical and theoretical context. You should provide the early part of the discussion, where you interpret your results with respect to the study’s research questions and hypotheses. You should consider and describe their meaning in light of the research questions/hypothesis presented in the introduction. You should also briefly note any strengths and limitations of the research design that can be ascertained from the information provided in the methodology. Please note that you are not expected to integrate the findings, and may discuss the qualitative and quantitative results separately.
· There is no need to provide references, but if you wish you may provide references to statistical texts or qualitative analysis texts to support the decisions you have made in analysing the data.

Please Note: there is a strict 10 page limit for this assignment. Your title page, results and discussion are included in this 10 page limit. References and your appendix containing your output are NOT counted in this 10 page limit. Pages in excess of this limit will not be read by your markers so will not contribute to your marks. Marks will be deducted if the prescribed APA formatting is not followed (i.e. not correct font or not double spaced or narrow margins). If you do not use the correct formatting and your assignment would exceed the page limit if you had, you may receive both penalties.

===========================================================================================================================

MIS770 Foundation Skills in Data Analysis
15 Sept 2022

2.1 Scenario
Your company has previously commissioned you, Minjee Lee (Chief Analyst), to develop a survey to help them obtain
an overview of all facets of the company’s employees. Now, in order to obtain a better understanding of the make up
and type of their employees, your CEO (Margot Robbie), wants you to process and analyse the data set and then
answer several questions. The questions you need to answer are contained in the following memorandum.
Margot does not have an analytics background, so it’s important that you utilise “plain, easy to understand language”
in your answers. If you believe you need to include any technical terms, then you must explain these in a clear and
succinct manner using layman’s terms.
2.2 Memorandum
Date: 25th August, 2022
To: Minjee Lee, Chief Analyst
From: Margot Robbie, CEO
Subject: Analysis of Employee Survey Data
Dear Minjee,
Can you please carry out an analysis of the recent Employee Survey data (contained in the file EmployeeSurvey.xlsx)
and prepare a report containing answers to the following questions.
Q1. Summaries of key variables of interest
Can you please provide me with separate summaries of the following variables, just by themselves? In other words,
please investigate each variable individually without reference to any other variable in the dataset.
(a) “Income” – the amount we paid our employees last financial year.
(b) “JobSatisfaction” – how satisfied are our employees working for the organisation.
Q2. Exploring relationships between two variables
(a) I would like to know if there is a link between the level of annual income (“Income”) and the number of years
our employees have worked since the age of 16 (“WorkYears”). Historically I suspect that the more years you
worked, the higher your Income becomes, but in order to assist us with future budgeting, I’d like to know if
this is still the case. Therefore, I’d like you to establish from your sample data if there is any relationship
between these two variables.
(b) I’m also interested to establish if there is a relationship between gender (“Gender”) and an individual's status
as a union member (“MemberUnion”).
(c) Further, it would be helpful if we knew if the age (“Age”) of a person had any relationship to how an
employee describes their relations in the workplace with management (“ManEmployeeRel”).
Q3. Estimating employee measures
(a) To assist the organisation with its future budgets, I would like you to estimate the number of hours our
employees as a whole would work per week (“WorkHrs”) - the days of a 40 hour week have long gone.
(b) As you would realise, employee happiness with their organisation has a direct effect on a company’s culture
and productivity, therefore I’m interested to know if you can estimate the proportion of all employees who
are Very Proud or Somewhat Proud (“ProudOrg”) to work for the company.
Q4. Employee claims
(a) I read a report recently that indicated that the annual salary paid to staff in our industry (“Income”) was
$62,000. I’d like to think that we were paying our staff more than this figure. Is there any evidence to suggest
that this is the case?
(b) Another finding in that report concerned what job characteristics (“JobCharacteristics”) were most important
to employees. It claimed that less than 40% of employees enjoyed their work. Can you also check this claim
against your survey data of our employees?
Q5. Appropriate sample size
Finally, I am concerned that the sample of 400 employees is too small to provide accurate results as this seems hardly
enough data. If we ever decide to survey our employees again, I would like to be able to:
(a) calculate approximately the average annual income (“Income”) we are paying our staff to within $5,000, and
(b) estimate the proportion of employees that are satisfied (“JobSatisfaction”) with their job to within 3%.
Therefore, how many employees would we need to include in the next survey to satisfy both of these requirements?
Regards, Margot

=================================================================================================

COMM1190: DATA, INSIGHTS, AND DECISION

QUESTION 1
You have been brought in as a Data Science consultant on a court case. A chemical company
has been found negligent after a chemical spill at one of their plants. All that remains in the
court case is to decide on the extent of the damages for which the company is liable. One way
the court has been deciding on this amount is to look at the impact the spill has had on the
value of houses located near to the chemical plant where the spill occurred.
As the expert witness, you have been asked to evaluate some alternative strategies to
estimate the impact on housing prices (price). Strategy A involves taking a sample of sales
that occurred after the spill where the houses are classified as either being close to the plant
or not. This feature was designated by a variable near that was equal to 1 if the house was
deemed to be close to the chemical plant and zero otherwise. Then a regression analysis is
performed using the following model (MA):

MMMM: ppppppppppii = ββ0 + ββ1nnnnnnnnii + uuii.

Strategy B involves taking a sample of sales for houses near to the plant but where some
sales occurred before the spill and some after. The variable after is equal to 1 if the house
was sold after the spill and zero if the sale was before. Then a regression analysis is
performed using the following model (MB):

MMMM: ppppppppppii = ββ0 + ββ1aaaaaaaaaaii + uuii.

Part A.
Explain A and B as strategies to estimate the impact of the chemical spill and critically
evaluate each of them. Is either preferable to the other?

Part B.
Suggest an alternative regression model that is preferable to MMMM given that you only have data
from after the spill. Does this address all your criticisms of Strategy A that you outlined in part
(a)?

Part C.
Using housing data models MA and MB are estimated, and the results given below. How do
you interpret these results? (Note that pppppppppp is expressed in $1000)

pppppppppp � = 131.9
(4.0)
− 40.0nnnnnnnn
(7.6)

nn = 142, RR2 = .165, ssssssssssssssss eeeeeeeeeeee iiii ( . )
pppppppppp � = 63.7
(5.9)
+ 28.3aaaaaaaaaa
(9.1)

nn = 96, RR2 = .094, ssssssssssssssss eeeeeeeeeeee iiii ( . )

Part D.
Suppose you have sales both near and not near to the plant as well as sales before and after
the spill. Suggest an alternative strategy to estimate the effect of the oil spill on housing prices
that is preferable to both MA and MB?

Word Limit: 800 words for entire question (i.e., all subparts).

QUESTION 2
Imagine you work for a large department store, which highly values customer service. The
following chart shows how customers contact the customer service centres.

You begin to discuss the chart with your manager. Immediately, she has the following queries:
“I want to see the overall trends, but it is difficult to see with all the seasonal spikes in the time

series. I’d like a simpler view into the trend.” You decide to create some charts to address your
manager’s queries.
Part A.
Using the four frameworks typology, identify the type of chart you would use to address the
query and explain why.

Part B.
Sketch two alternative charts for the query. For each chart, provide a brief explanation of your
design choices. To sketch the chart, you can use any tool you want (e.g., you can use a
software tool like infogram, excel, or R). Alternatively, you can sketch the chart using pencils,
pens or markers on paper, then take a picture of the charts and paste them into your solutions
document. You can access the underlying data “customer_service.xlsx” on Ed.

Part C.
Evaluate your two charts and explain which you would select to further develop to present to
your manager.

Word Limit: 500 words for entire question (i.e., all subparts).

QUESTION 3
As part of a preventive health program, we are interested in building a model for predicting
diabetes among women. We have access to the following information for each woman:
- diabetes: Yes or No for diabetic.
- npreg: number of pregnancies.
- glu: plasma glucose concentration
- bmi: body mass index
- ped: diabetes pedigree function
- age: age in years.
We have split the available data into a training dataset with 332 cases and a test dataset with
200 cases.
With reference to the above case, please answer the following questions:
Part A.
We have run a logistic regression to predict diabetes using npreg, glu, bmi, ped, and age.
The R output from this logistic regression is as follows.
Based on the output, write down the mathematical equation of the logistic regression
associated with this R output.

(2 marks)

Part B.
Based on the output in Q 3 Part A, provide an interpretation of the coefficients associated with
bmi and npreg?
Part C.
Based on the output in Q 3 Part A, explain and justify the predictors that have a statistically
significant relationship to the response?

QUESTION 4
As an alternative to the logistic regression in Q3, we have also fitted the classification tree
below:

Part A.
Based on this decision tree, would you predict that a woman with the following characteristics
has diabetes: npreg=0, glu=140, bmi=47.9, ped=0.259, age=26? Justify your answer.

Part B.
Provide TWO facts to demonstrate the consistency of results from the classification tree above
with the findings reported in Q3 Part A.

QUESTION 5
You are a data analyst for AppCo. AppCo produces a smartphone app that allows users to
virtually try on clothes. It is funded by having sponsored links to online clothing retailers.
AppCo tells its users, “top brands, all sizes, best prices”. AppCo has a dashboard that is used
by Board members that shows sales, revenues and a comparison of sales by retailer. You
have been asked to consider whether having a selection of the most popular sizes at a slightly
lower price would increase revenue. The CEO has said, “well just use some of your magic A/B
testing ...”.

Part A.
Identify potential legal issues that may arise from A/B testing if AppCo users are unaware the
experiment is taking place.

Part B.
Evaluate whether A/B testing would lead to any legal consumer issues?

Part C.
Recommend steps to ensure that your organisation maintains good governance when
developing analytics at AppCo.

Word Limit: 500 words for entire question (i.e., all subparts).

===================================================================================================

AERO4560 Flight Mechanics 2
4 Sept 2022

All analysis should be done on the aircraft (Pilatus PC9) supplied to you in the assignment
package. A non-linear flight simulation framework will also be supplied to you to assist your
analysis.
Non-linear simulation framework state vector:

X = [u, v, w, p, q, r, φ, θ, ψ, xe, ye, ze]
T

Non-linear simulation input vector:
U = [δt
, δe, δa, δr, δf ]
T

All variables are in SI units with angles defined in radians. δt has a range of 0 (idle) to 1
(full throttle), you can assume all other control surfaces have a limit of 25 degrees on each
side. Equilibrium flight condition and trim settings are available on Canvas. Note that δf
represents flap setting and it is included to generalise the framework. It should be removed
from your linear analysis.
The numbers in square brackets at the beginning of each question indicate how much each
question is worth.

Week 2 MATLAB Tasks

Q 1 Linearisation
MATLAB Tasks
(a) MATLAB Grader [30]
Write a function to determine the A and B matrices describing small perturbation
motions about the equilibrium condition.
(b) MATLAB Grader [20]
Isolate the lateral components of the A matrix, the B matrix and the state
vector X. Use suitable C and D matrices to represent v as side-slip angle β.

Xlat = [β, p, r, φ, ψ]
T

(c) MATLAB Grader [5]
Extract the Eigenvalues of Alat matrix.
(d) Set up the linear lateral state equations of motion in state-space form. Use a
suitable integration procedure to generate time domain responses of β, p, r, φ and
ψ to control surface step inputs equivalent to 1 degree rudder input. Using the
non-linear simulation framework, generate equivalent time domain responses.
Analysis Questions
(i) Canvas Quiz [5]
Identify the dynamic modes of the aircraft in the given flight condition using the
Eigenvalues in Q1(c) and describe the aircraft stability based on the eigenvalues.
(ii) Canvas Quiz [10]
Using the results from Q1(d), plot the responses of each state from linear and
non-linear equations of motion on the same plot. Compare and explain any
differences.

Q 2 Heading Angle Response
MATLAB Tasks
(a) Using a suitable C matrix, generate the heading angle response (ψa = ψ + β) of
the aircraft to rudder input.
Analysis Questions
(i) Canvas Quiz [5]
Using the results from Q2(a), plot ψa response together with β and ψ responses
to the same input. Comment on the results.

Week 3 Questions

Q 3 5th Order Transfer Functions
MATLAB Tasks
(a) MATLAB Grader [15]
Use the MATLAB functions ss2tf,tf,ss2zp,zpk etc. to generate the 5th order
transfer functions describing the β, p, r, φ and ψ responses to a rudder input, in
both polynomial and pole-zero forms.

Q 4 Dutch Roll Mode Modal Approximation
MATLAB Tasks
(a) MATLAB Grader [10]

From the five dimensional space model in Q1, generate the two dimensional state-
space approximation corresponding to the Dutch Roll mode.

(b) MATLAB Grader [10]
Using Laplace transforms, generate G
β
δr
(s) and Gr
δr
(s) using the two dimensional

Dutch Roll mode state-space system.
(c) MATLAB Grader [10]
Using Cramer’s rule, generate G
β
δr
(s) and Gr
δr
(s) using the two dimensional Dutch
Roll mode state-space system. Compare to the results of 4(b) and explain any
differences.
(d) MATLAB Grader [0]

Using the results from Q3, apply frequency domain modal approximation pro-
cedures to the 5th order transfer functions G

β
δr
(s) and Gr
δr
(s), generate the 2nd
order transfer functions which approximate the Dutch Roll mode response.
(e) MATLAB Grader [0]
Generate G
ψ
δr
(s) from both the 2nd order system in Q4(b) and the 5th order

system by applying modal approximations.
(f) MATLAB Grader [0]
Generate G
ψa
δr
(s) from both the 2nd order system and the 5th order system by

applying modal approximations.
Analysis Questions
(i) Canvas Quiz [7]
Compare the 2nd order transfer functions in Q4(d) to the results of Q4(b) and
Q4(c) and explain any differences.
(ii) Canvas Quiz [8]
Compare the two results in Q4(e) and explain any differences.
(iii) Canvas Quiz [8]
Compare the two results in Q4(f) and explain any differences.

Week 4 Questions

Q 5 Step Responses Comparison
MATLAB Tasks
(a) Using G
β
δr
(s) from Q4(b), apply partial fractions and inverse Laplace transforms
to determine an expression for the step response in β to an rudder step input
equivalent to 1 degree.
(b) Use the MATLAB function step to generate the equivalent Q5(a) step response
using the 5th order transfer function obtained in Q3 (exact solution).
(c) Use the MATLAB function step to generate the equivalent Q5(a) step response

using the 2nd order transfer function obtained in Q4(b) (Dutch Roll modal ap-
proximation).

(d) Use the MATLAB function step to generate the equivalent Q5(a) step response
using the 2nd order transfer function obtained in Q4(d) (frequency domain modal
approximation).
Analysis Questions
i Canvas Quiz [15]
Plot the resulting time domain responses from Q5(a), Q5(b), Q5(c) and Q5(d)
on one figure. Compare and explain any differences.
ii Canvas Quiz [15]
Compare the results of Q5(a), Q5(b), Q5(c) and Q5(d) to the simulated responses
obtained in Q1 from the non-linear simulation and explain any differences.

Week 5 Questions

Q 6 Frequency Responses Comparison
MATLAB Tasks
(a) Use the MATLAB function bode to plot the exact frequency response from the
5th order transfer function G
β
δr
(s) obtained in Q3.

(b) Use the MATLAB function bode to plot the frequency responses of the 2nd order
approximations to the transfer function obtained in Q4(b) and Q4(d).
(c) Use the MATLAB function bode to plot the 5th order frequency responses for
G
β
δr
(s), G
ψ
δr
(s) and G
ψa
δr
(s).

Analysis Questions
(i) Canvas Quiz [7]
Plot the results from Q6(a) and Q6(b) on the same figure. Comment on and
explain the similarities and differences between the results.
(ii) Canvas Quiz [10]
Explain the differences in the three responses in Q6(c) with respect to their
frequency content and phase relationships. Relate these behaviours to the time
domain results obtained in Q1.

Q 7 Aileron and Rudder Control Effect
MATLAB Tasks
(a) Use MATLAB function bode to generate the exact frequency responses for β and
φ resulting from δa and δr excitation.
Analysis Questions
(i) Canvas Quiz [5]
Plot the frequency response of β and φ to δa input on the same figure, then plot
the frequency response of both states to δr on the same figure. Comment on the
relative activity of β and φ in response to δa and δr.
(ii) Canvas Quiz [5]
Plot the frequency response of β to both δa and δr on the same figure, then plot
the frequency response of φ to both control inputs on the same figure. Comment
on the relative effect of δa and δr on β and φ.