Reference no: EM131711609

Statistical Programming for Data Science

Does Public Transit Encourage Physical Activity?

In recent years there have been a number of health interventions conducted world-wide, with the goal of encouraging adults to become more physically active on a daily basis. A motivation for such interventions is that physically active adults have been reported to have lower rates of chronic diseases as well as a reduced risk of obesity, diabetes and heart disease. However, with demanding careers and busy personal lives, many adults do not make time for regular exercise.

Active public transport has long been the focus of health initiatives, as it involves walking to bus or tram stops or train stations, as well as cycling using public bike lanes and roadways. In some countries active public transport has become public policy - for example, in Denmark, walking and cycling are popular modes of public transport and now form part of the government outdoor recreation strategy.

In 2010-2011 the NYC Department of Health and Mental Hygiene conducted the Physical Activity and Transit Survey (PAT). The PAT Survey consisted of three parts:

1. A telephone survey of physical activity and health
2. A weeklong accelerometer device component of a sub-sample of participants; and
3. A weeklong GPS device component.

The data can be downloaded from the COMP 5070 website.

Your analysis and conclusions in your report should focus on whether the use of Public Transport encourages physical activity. You have been provided a number of data sets you are free to use as you see fit, providing the requested components (below) are included. You can focus on one file or merge data files - wherever your analyst's brain takes you

For the report you are free to analyse whichever aspects of the dataset(s) interest you, with the following caveats:

1. Movement achieved via public transport must be part of the analysis.

2. MVPA (Moderate to Vigorous Physical Activity) forms part of your analysis.

3. The report focuses on answering 2 questions of interest - posed by you - and answered by your data analysis.

Programming language: R is probably easier to use however you are free to use just R, just Python or R and Python - whatever you need to get the job done

The report should contain the following:

1. Report Aim: Write a short explanation of the analysis to be performed and an explanation of the question(s) you are investigating within the data (1-2 paragraphs).

2. Data Summary: Write a short explanation of the data you will analyse - e.g. demographics of interest and/or other variables you will investigate (1-2 paragraphs).

3. Data Cleaning an explanation of any data cleaning performed (including merging) to enable the analysis to be performed.

4. Descriptive statistics including visualisations to support your analysis including the 2 questions you have posed as part of your analysis.

5. Statistical comparisons: besides focusing on MVPA there should be a comparative aspect of interest in your analysis. E.g. MVPA by activity type or MVPA by demographics, MVPA by day of the week ... these are just suggestions - it's really up to you

As a minimum you need to produce one comparison for each question of interest you wish to answer.

6. Exploratory Factor Analysis: conduct an exploratory factor analysis for the data in PublicTransportSurvey.csv - a codebook has been provided in the file PublicTransportSurvey.txt. Partial code has been provided with this take-home exam in the file TakeHomeExam_EFA.R. Please use this code as your starting point and follow the prompts given to you inside the file. In particular look for lines starting:
### !!! or ### R
as there is a question in the code you need to answer and code you need to write, respectively.

For the Exploratory Factor Analysis you need to include the following in your written report here:

- The Cronbach Alpha output and a short discussion (2-3 sentences) as to whether the data is trustworthy and why you conducted the Cronbach Alpha test.

- Correlation output of your choosing (graphical and/or numerical) with an accompanying discussion (3-4 sentences). If numerical, round the correlations to 2 digits. There is a hint below regarding rounding.

- A single paragraph explaining the outcome of the determinant test, Bartlett's test of sphericity and the KMO statistic for the data. Do not include R output.

- Your decision regarding the number of factors to estimate (scree plot may be shown, do not show the R console output).

- The FINAL factor solution you have chosen - you may wish to present this as a table. You do not need to discuss results of any other solutions tried, however you should justify your final factor solution.

- The final factor solution should include names of the factors in each analysis and an explanation as to how you come up with these names.

- Are there any survey questions you think could be dropped? Explain.

- The plot of your final factor solution. In the workshop example we plotted four in one overall graph; in this case you only need to produce one plot and it should match your chosen solution.

7. Conclusion: Draw conclusions about:
i. The use of public transport by visitors to Adelaide (1 paragraph).
ii. Your analysis overall (1 paragraph).

Data Files

https://www.dropbox.com/s/ehjdso1unzrkpjg/Exam%20and%20data%20files%20-%2011-8-2017.zip?dl=0