Guidelines for Completing and Submitting POLS0008 Report

▪   Read the below guidelines to avoid losing unnecessary marks.

▪   The assessment is due on 3rd May 2023 at 14.00 hours. Please follow all designated Department of Political submission guidelines. THESE MAY BE DIFFERENCE TO THOSE OF YOUR HOME DEPARTMENT. The submission guidelines are available on the Moodle page for this module. The word limit is 3,000 words, excluding tables and graphs, references, and your R script appendix (see below).

▪   Name the file you submit using your 2022-23 Candidate number.

▪   This is an assessed piece of coursework for the POLS0008 module; collaboration and/or discussion with anyone is strictly prohibited. The rules for plagiarism apply and any cases of suspected plagiarism of published work or the work of classmates will be taken seriously.

▪   Along with the report outline, the datasets  for the report can be found in the ‘Datasets’ folder on Moodle.

▪ The data come from the British Social Attitudes – a face to face survey of 3,224 adults 18 and over.

▪   You may open the dataset and work on the report anytime up until the submission date. There is no limit on the number of times you may open the data files. Be sure to save your data file and R script file.

▪   Where appropriate, answers should be written in complete sentences; no bulleting or outlining. Be sure to answer all parts of the questions posed and interpret output statistically and substantively. An attempt will almost always be worth more marks than no attempt.

▪   You should include tabular and graphical output alongside your written answers.

▪   You are required to submit a copy of your R script appended to your report. You can copy and paste this from your R script window to the end of your word- processed document.

▪ FAILURE TO SUBMIT YOUR R SCRIPT WILL INCUR A 5-POINT

PENALTY.  Note that your R script file should include comments indicating the question   being   addressed.   Your   R   script   file   should   contain   only   the exercises/questions asked here.

▪   You do not need to explain in detail the background to the study or methods of data collection.

▪   You should not cite or discuss relevant literature.

▪   You should discuss the interpretation of your results and how they relate to the questions you were asked.

▪   You may  assume the methods you have used  (e.g.,  chi-square  and t-test)  are understood by the reader and do not need definitions, but you do need to say which techniques you have used and why.

▪ 10 points will be awarded for presentation.

Submission checklist

These documents should be submitted to separate Turnitin links on Moodle.

The data file is “bsa2019_poverty_open.dta” . You can download this file in the usual way from Moodle once you have agreed to the conditions of use. A brief user guide can be foundhere.

The variables in the “bsa2019_poverty_open.dta” dataset that you might require are:

DOs and DON’Ts

-     DON’T include raw variable names in the text or tables

-     DON’T use too many decimal places, but be consistent

-     DON’T include unedited R output or you will lose marks

-     DO make sure tables and figures have titles and are referred to in the text

-     DO make sure your tables and figures can be understood without reading the text

-     DO make sure you have given a clear enough description of what you have done so that the reader can reproduce any numbers/results that you present

-     DO be careful how you use the terms ‘significant’ and ‘correlation’ because they have specific meanings in social statistics.

Report brief

You have been commissioned by the Resolution Foundation think tank to write a report on public perceptions of benefit fraud. You will use data  from the British  Social Attitudes Survey 2019. You should start your report with an introduction describing the data and variables you will use and present the complete case sample characteristics in a table (10 points). Your report should end with a concluding section, using no more than 400 words, summarising how your findings can inform whether public opinion is in line with the Department for Work and Pensions official estimates of benefit fraud, as well as explaining any limitations to your data and methods. The conclusion should also address which groups should be targeted in efforts to reduce misinformation about benefit fraud (20 points).

You should present your results using a maximum of eight figures (tables or plots). Ensure all results are interpreted substantively and statistically. You may be required to recode the data to answer certain questions. You should specify an alpha level of 0.05.

1.   How does mean public perception of benefit fraud vary within and between the following measures: sex, education, and marital status? (10 points)

-     Calculate an appropriate measure of central tendency and dispersion for public perception of benefit fraud for categories of sex, education, and marital status. Report a table and describe its contents.

2.   Is the mean  estimate  of perceived benefit  fraud different to  an estimate from Department for Work and Pensions data of fraudulent applications of 1. 1%? (5 points)

-     Use an appropriate statistical test to test a hypothesis about the sample mean of perceived benefit fraud. Report the results of your test.

3.   Do people who think most benefit claims are fraudulent read news via a social media more regularly? (10 points)

-     Produce a cross tabulation between whether a respondent thinks more than 50% of benefit claims are fraudulent and their frequency of reading new via social media.  Describe  the  finding  from  your  table  and  report  the  result  of an appropriate statistical test to determine whether there is a relationship between the  two  variables.  Present  only  a  final  cross-tabulation  that  meets  the assumptions of the test. You will need to derive your own variable determining whether a respondent thinks most people (i.e., 50% or more) claim benefits fraudulently.

4.   Are those with children in the household more likely to think more people claim benefit fraudulently? (15 points)

-     Use  an  appropriate  statistical  test  to  test  this  hypothesis  and  demonstrate whether  your  data  meet  the  assumptions  to  conduct  the  test.  State  your hypothesis and report the results of your test in a table. Describe whether the data meet four assumptions required for the test using the data in the table and using up to two separate plots.