[SOLVED] R Scheme statistic RSM2: Coursework Report (25% final grade)

RSM2: Coursework Report (25% final grade)
Research Methods and Statistics 2 PSYL10126
Coursework set: 17:00, Thursday 31st October 2019 Coursework due: 12:00, Monday 25th November 2019
Queries concerning the task
This document contains the details of the quantitative research report for RMS2. There is an on-line discussion board specifically for the coursework. If you have a question, it is likely your class-mates may have the same question. Before posting a question, please check the on-line board in case it has already been answered. You can post anonymously if you wish.
You can also make use of office hours for coursework questions.
Please do not work on coursework in labs. Neither tutors nor staff will assist with this assignment during lab hours.
Format & Grading
Your report will be graded based on the university common mark scheme. Note that this is different to the RMS report you completed in year 2. This task requires you to make a set of decisions in order to provide analyses to answer a set of questions. You will likely all approach this in slightly different ways, and explain your choices in different ways.
The report should be structured in a similar way to a published paper. Below (see Task) you will find a description of a study and some data. You are expected to conduct the appropriate analyses to answer the questions given, and to provide a write up split into the sections described below. The page suggestions are approximate, but indicate the relative weighting of the sections for grading.
Analysis Strategy (Approx. 1.5 pages): This section should outline the steps you will take in your analysis. It should include:
a. checking data
b. specification of the model
c. identifying which coefficients in the model test specific research questions
d. description of how model assumptions will be checked
What we are looking for: A good analysis strategy will provide details on all of the above, and rationales for your choices. These rationales should be clearly written. We should be able to tell from your presentation exactly what models you will be running. The proposed analyses should be able to answer the questions posed in the task.

Results (Approx. 2.5 pages): Here you will report the results of the analyses as set out in the analysis strategy. Results should be clearly presented (choice of text, table and plot dependent on what is most appropriate). The section should contain brief interpretations of the results to demonstrate knowledge of the different elements of your results. You should report appropriate assumption checks for your models, the conclusions from which should be described here, and the associated plots presented in the appendix and referenced in text.
What we are looking for: Clear presentation and accurate interpretation. Results presented should be consistent with the plan outlined in the first section.
Discussion (Approx. 1 page): You should provide a discussion that briefly summarises your findings with respect to the questions (See `Task`) and then evaluate your analyses.
What we are looking for: You should be succinct, and connect the statistical findings to the applied questions. These statements would likely only be brief. You should include a short discussion of any observations you have about the data, questions and analyses, what you think was good, bad, what limits you see in the data etc. Please keep the page limit in mind here. We do not expect in depth discussion of the topic.
Note your report should:
Be no more than 5 pages long including tables and figures (excluding appendix see next point). Anything which appears beyond these limits will not be marked.
You may also include ONE additional appendix that contains the plots from your assumptions checks. There is no page limit on this appendix, but anything included in the appendix should be referenced and discussed in the main text. You may not include additional analyses of the main research questions, or additional plots/tables for the main analyses in this appendix.
Use a regular font (e.g. Times New Roman), size 12, double-spaced. Include standard 2.54cm (1 inch) margins and follow APA formatting and writing style.
Failure to follow these formatting rules will lead to a loss of marks for overall report presentation.

Hand-in
Online Hand-in: Your report and the associated R-code should be submitted electronically via the Turnitin links on the LEARN page for RMS2. There will be two links located under Assessment Details & Submission tab. One will be clearly labelled for the report, the other clearly labelled for the R-script.
IMPORTANT: Please name your files using the following format:
ExamNumber_RMS2Report ExamNumber_RMS2Code
Replace ExamNumber above with your actual exam number. Your exam number begins with B0, e.g. B088888. Do not include any identifying information in the file name (so no name and no matriculation number). This is very important to allow us to link the code to the report.
Please also include at the top of your code a list of exam numbers for people you worked with in completing the coursework. Line 1 of your R-code should be something like:
# Produced in collaboration with B01111111 and B022222222
The course administrator will email you to remind you of these details as the deadline approaches.
R-Code
You are required to submit your R-code that reproduces the results provided in the report.
If the code fails to run, or does not exactly reproduce the results reported, you will lose 10% points (e.g. 72% becomes 62%).
Please check that your code runs completely error free before submitting. To do this:
1. Open RStudio
2. Empty your working environment type in the console: rm(list=ls())
3. Open the R script you plan to hand in
4. Run the script line by line in order
5. If an error appears, correct it, then loop back to point (2) and start again
6. Continue until you get all the way through your script without errors
If you have any doubts as to whether something appearing in your script is an error please come to office hours before you submit your coursework.

Task: Investigation into Teacher Training and Bullying
Background: Scientist and Researcher (2015) conducted a study to evaluate the efficacy of specialised teacher training on bullying reduction. Two hundred teachers from multiple schools were randomly allocated to the training condition or the control condition, and the percentage of all reported bullying incidents in which teachers intervened for the following year was recorded. Orthogonally to this, mandatory bullying feedback sessions were added to half of all regular parent-teacher meetings (i.e. meetings between a given teacher and a specific students parents) to assess if the training produced results independently of communication about bullying; the other half of teachers did not have these sessions. In order to control for possible confounds of some teachers being more sensitive to bullying than others, or some being more experienced teachers than others, baseline scores on an empathy measure and years of teaching experience were also collected.
Research Questions: In your report, provide appropriate descriptive statistics for the data and then answer the five questions below. For each question, provide the appropriate statistical test and interpretation to answer the question. For every test you perform, conduct appropriate model diagnostics and check assumptions.
1. What is the relationship between empathy and experience?
2. Controlling for empathy and experience, does training increase bullying interventions by teachers?
3. Similarly controlling for confounds, does feedback produce increases in intervention?
4. Which produces a stronger effect?
5. Is there any interaction between training and mandatory feedback? If so, what does this effect mean/signify?

Data Dictionary
The data set is available on LEARN in the Assessment Details tab. The file is titled RMS2_report_1920.csv.
The table below provides a full description of all the variables in the data set:
Label
Description
subject
Unique identifier for each participant
incidents
[0%-100%] Percentage of total reported incidents that teachers intervened in.
training
[1 = no training, 2 = training]: whether the teacher underwent specialised training
feedback
[1 = no feedback, 2 = feedback]: mandatory bullying feedback discussions with parents during regular meetings
empathy
[1 to 20]: baseline empathy measure, 1 = low empathy, 20 = high empathy
years
[0 to 30]: years of teaching experience (rounded to whole number)