HAR6045 Assignment 2021 Introduction This assignment requires you to analyse the data set provided, and write a report of up to 2,000 words about this analysis. It will test your ability to select and conduct suitable methods of analysis for different types of data, and interpret and present the results of these in a style appropriate for a general audience. Required format The report should include the following sections: • A structured abstract of no more than 200 words • A brief introduction • A section on descriptive analysis of the data • A justification for the types of analysis conducted to answer the research questions • A full report on the analysis conducted, including statistical interpretation of the findings • A conclusions section The writing style should be appropriate for a medical journal such as the British Medical Journal (BMJ), however the content will probably be slightly different: it is not important to discuss the clinical issues in any real depth, and the statistical reporting will probably be slightly more comprehensive than that typically seen in medical journals. The following restrictions apply to reports: • They should start with an abstract of no more than 200 words • They should have no more than 2,000 words (excluding abstract, references, tables and appendices) • Up to six tables may be included in the main report • Up to four charts/figures may be included in the main report Please note that raw SPSS output tables should not be included in the report; you will need to extract he information that you think is relevant from your SPSS output and present it in a suitable format. If more than six tables or four charts are needed, some may be placed in an appendix. Technical appendices may also be used if required for other information, e.g. SPSS syntax. Research objectives The data come from a multi-centre randomised control trial of 240 patients with chronic obstructive pulmonary disease (COPD) designed to compare hospital-based rehabilitation (129 patients) with community-based rehabilitation (111 patients). The objective of the study was to determine whether community-based rehabilitation was better than hospital-based rehabilitation for the treatment of COPD. The primary outcome was whether patients responded to treatment or not after 18 months. This was a subjective assessment made by an assessor who was blind to the treatment. Secondary outcomes included Forced Expiratory Volume per second, Forced Expiratory Vital Capacity, and Peak Expiratory Flow. Also of interest to the investigators was whether patients withdrew from the trial, and if so after how long. The data The study data are provided in the data set “COPD data.sav”. This includes data for all 240 patients in the trial. The 38 variables are fully labelled; however, the following are the variables that it is expected you will include in the analysis: Grouping variable GROUP Hospital or community-based rehabilitation (1 = hospital, 2 = community) Outcome variables RESPONSE Responded to treatment? (1 = yes, 0 = no) FEV3 Forced Expiratory Volume in one second (litres) at 18 months FVC3 Forced Expiratory Vital Capacity (litres) at 18 months PEF3 Peak Expiratory Flow (litres/minute) at 18 months Other study variables WITHDRAWN Withdrawn from study? (1 = yes, 0 = no) TIME Time to withdrawal (weeks) Baseline variables FEV Forced Expiratory Volume in one second (litres) at baseline FVC Forced Expiratory Vital Capacity (litres) at baseline PEF Peak Expiratory Flow (litres/minute) at baseline Demographic variables AGE Age (years) SEX Sex (1 = female, 2 = male) Analysis required The report should include the following analyses: • Appropriate descriptive/summary analyses of all variables described above • An appropriate hypothesis test to examine whether there is any association between treatment group and the primary outcome, response after 18 months, with age and sex as covariates • Appropriate hypothesis tests to determine whether there is any difference in each of the secondary outcomes (FEV3, FVC3, PEF3) by treatment group. These should include age and sex and baseline measurements as covariates (e.g. the baseline measurement for FEV should be included when FEV3 is the outcome, and so on) • An appropriate hypothesis test to determine whether there is any difference in time to withdrawal, by treatment group, amongst those who did withdraw – again including age and sex as covariates Your analysis of all variables should involve careful consideration of what type of variable each is. Summary statistics should be presented in a table where appropriate, but any important information should also be described in the text. All hypotheses should be stated clearly, with the choice of hypothesis tests justified and any assumptions stated. All analyses should be reported with appropriate statistics (e.g. parameter estimates, confidence intervals, p-values) either in the main text or in tables. The report should include an interpretation of each of the hypothesis tests, and comments on other analyses as appropriate. Marking Marks will be awarded in the following proportions: Introduction and conclusion 10% Choice and justification of hypothesis tests 10% Descriptive analysis 20% Hypothesis tests: accuracy and reporting 30% Interpretation of results 20% Presentation and structure 10% These may be used as approximate guidelines for the emphasis given to each section. For example, as descriptive analysis is worth 20% of the marks, an appropriate length for this section would be 20% of 2,000 = 400 words. However, this is a guideline only, and is flexible; the important thing is to ensure the report covers the relevant sections in sufficient detail; you may feel this requires giving slightly more weight to one section than this method suggests. The report should state the number of words used (excluding abstract, references, tables and appendices). Usual penalties will apply for reports that exceed the word length, but no penalty will be applied to reports that are under this length (to encourage concise writing). Please see also the ScHARR standard marking criteria available on Blackboard. Submission Reports should be submitted by 1.00pm on Monday 7 June. Reports should be submitted via Blackboard, using turnitin.












































































































































学霸联盟